Full text of Economic Review (Federal Reserve Bank of Cleveland) : Fall 1983

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

Economic Review
Fall 1983




Plant Closings and
Worker Dislocation ...............................2
This article examines the justifications for
plant-closing laws, analyzing whether the
objectives for plant-closing laws can be
achieved. Authors Daniel A. Littman and
Myung-Hoon Lee find that dislocated workers
as a group do not suffer hardships any more
severe than do the long-term unemployed.
The article concludes that plant closings are
a relatively small source of national unem­
ployment and that plant-closing laws would
not necessarily inhibit closings. The article
further finds that such laws unintentionally
create incentives to reduce the size of the
work force in affected plants, especially
among high-seniority workers.
Prevailing Wage Laws,
the Federal Reserve, and
the Service Contract A c t ................... 19
Under the provisions of the Service Contract
Act, Federal Reserve Banks must ensure that
hired service contractors pay their employ­
ees according to wage scales set by the U.S.
Department of Labor. Since Federal Reserve
Banks also must charge for the services pro­
vided to depository institutions at prices
reflecting the full costs of those services, the
Service Contract Act affects the role of the
Federal Reserve System in the payments
mechanism. Author Mark S. Sniderman
explains how the Service Contract Act
increases Federal Reserve expenses and dis­
cusses the implications for priced transporta­
tion services.
Economic Review is published quarterly by the
Research Department of the Federal Reserve Bank
of Cleveland, RO. Box 6387, Cleveland, Ohio 44101.
Telephone: 216/579-2000. Editor: Pat Wren.
Design: Jamie Feldman. Typesetting: Lucy Balazek.
Opinions stated in the Economic Review are those
of the authors and not necessarily those of the
Federal Reserve Bank of Cleveland or of the Board
of Governors of the Federal Reserve System.
Material may be reprinted provided that the source
is credited. Please send copies of reprinted materials
to the editor.

Daniel A. Littman
is an economist with
the Federal Reserve
Bank of Cleveland;
Myung-Hoon Lee
is a researcher/con­
sultant with the
World Bank. Roger
Hinderliter and
Gary Wyckoffpro­
vided insightful
comments on drafts
of this article, as did
many other mem­
bers of the Research
Department of the
Federal Reserve
Bank of Cleveland.

Plant Closings and
Worker Dislocation
by Daniel A. Littman and
Myung-Hoon Lee

1. Although several
closing-related pro­
posals have been in ­
troduced to congres­
sional committees
in recent years, as
yet the federal gov­
ernment has not im ­
plemented policies
to address the plantclosing problem.

Federal Reserve Bank of Cleveland



In any dynamic economy, capital investments
must be retired at the end of their useful
lives and replaced by more productive invest­
ments. The industrial structure and its geo­
graphic distribution respond to a variety of
powerful economic forces, including changes
in prices, consumer preferences, production
technologies, and international trade compe­
tition. The opening of new plants and closing
of obsolete plants are part of this vital pro­
cess. An economy is efficient when produc­
tive resources are allowed to move freely in
response to this ever-changing environment.
It follows that some magnitude of capital
turnover, reflected in part by plant openings
and closings, may signify economic health,
as it assists growth and competitiveness (see
Schumpeter 1950). Permitting a relatively
uninhibited flow of capital among firms, in­
dustries, and geographic locations therefore
may lead to net economic benefits for society.
Plant closings also can impose tremen­
dous adjustment costs on particular economic
actors—laid-off workers (hereafter, dislocated
workers) and their families, local govern­
ments, and local businesses linked to the clos­
ing plants. Plant closings are attracting
increased media and public attention because
of the problems that such closings cause or
symbolize—socioeconomic hardships for dis­
located workers, increased unemployment,
decay of the local employment base, and fis­
cal distress. Public officials have begun to
consider policy options to strengthen the
social safety net and to moderate the pace of
plant closings. Proposals include incentives
for employee ownership of marginal plants
(employee stock ownership plans, or ESOPs),
expanded use of industrial investment and
employment/wage incentives, and increased
occupational training and income mainte­
nance assistance. One popular and compre­
hensive policy option considered by state and
local governments is plant-closing laws,
which are examined in this article.1 Since
1975, five states (California, Maine, New
Jersey, Rhode Island, and Wisconsin) and at
least three municipalities (Philadelphia,
Pittsburgh, and Vacaville, CA) have enacted
plant-closing laws. During their 1981 and

1982 sessions, 21 state legislatures considered
almost 60 proposals to alleviate the problems
surrounding plant closings. State legislatures
have found it difficult to weigh the merits of
policy alternatives. Certainly one reason for
this difficulty is that the existing literature
does not embrace thorough economic anal­
yses of the plant-closing problem.
In section I of this article, we review
the characteristics of existing and proposed
plant-closing laws. We also outline the major
Table 1 Typical Provisions of
Plant-Closing Laws
Provisions of existing and proposed plant-closing
laws are listed in order of descending frequency. The
number of states having proposed or existing laws
with such provisions is listed in the right column.

Provision

N umber of
plant-closing
laws with
provision

Prior notice of closing

20

Firm-paid severance benefits

14

“Good-faith” sale efforts,
incentives for employee owner­
ship and firm reimburse­
ments for employee retraining

11

Continuation of health-insurance
coverage for specified period
after termination

10

Effects bargaining, or require-

10

ment for employer to discuss
effects of plant-closing with
workers
Employer payments to state of
specified proportion of annual
wage bill

9

Firm reimbursement for employee
relocation expenses

8

Paid leave time for workers prior
to shutdown

7

Preferential transfer rights for
affected workers

6

Decision bargaining, or require-

2

ment for employer to discuss
decision to close with workers

Economic Review • Fall 1983



justifications for plant-closing laws, such as
reducing unemployment, reducing hardships
for dislocated workers, and alleviating fiscal
distress of local governments. In section II
we review the evidence on dislocated worker
hardship and compare these hardships with
the problems experienced by other unem­
ployed workers. We conclude that dislocated
workers as a group cannot be distinguished
from other unemployed workers on the
basis of the severity of their problems. In
section III we examine whether, and to what
extent, plant closings cause unemployment.
Although plant closings are associated with
the shrinking of local labor markets, such
closings appear to be only a relatively small
source of national unemployment. In section
IV we analyze whether statutory provisions
that inhibit plant closings would effectively
reduce unemployment and the probability
of fiscal distress. The analysis shows that
plant-closing laws would not necessarily
delay or otherwise inhibit closings. In sec­
tion V we examine the effects of plant-closing laws on resource allocation, finding that
the laws may unintentionally create incen­
tives to reduce the size of the work force in
affected plants, especially among highseniority workers.

I. Plant Closing Laws
Existing and proposed plant-closing laws
are rather diverse with respect to compre­
hensiveness, coverage tests, and statutory
obligations placed on public authorities and
affected firms (see table 1). The proposed
laws are more comprehensive and more ambi
tious than existing statutes, in part because
these laws have yet to run the gauntlet of
legislative scrutiny. In Wisconsin, for exam­
ple, firms that wish to close facilities employ­
ing more than a specified minimum number
of workers must give employees and public
officials at least a 60-day prior notice of such
action. Maine’s plant-closing law requires a
60-day prior notice of closing and severance

pay to laid-off workers equivalent to their
average weekly pay multiplied by their years
of service. In marked contrast, the proposed
plant-closing law in Hawaii would require
employers to give a three-year prior notice of
closing. Ohio’s proposed law would require
a one-year prior notice and severance pay,
along with a variety of other obligations for
public authorities and the firm that is plan­
ning to close a plant (see table 2).
Advocates of plant-closing laws justify
such labor-market intervention on the
grounds of reducing hardships for dislo­
cated workers, reducing flows into national
and local unemployment, and insulating
local governments from fiscal distress that
a plant closing could cause. Advocates of
plant-closing laws recognize that plant clos­
ings impose severe socioeconomic hardships
on dislocated workers and attempt to alleviate

Table 2

Features of Ohio’s Proposed Plant-Closing Law

Coverage tests for firm
Employs 100 or more workers
Cannot be political subdivision or nonprofit firm
Must have operated in state for 5 years or more; if
affected facility was acquired from another firm,
purchaser succeeds to seller’s obligations
Affected by permanent shutdown for reasons other
than bankruptcy
Affected if transferring operations an “unreasonable
distance” and reducing work force by 10 percent
or more
Affected if one or more parts of an operation are being
phased out, resulting in an overall work force
reduction of 50 percent or more over 2 years

Obligations of state authorities
Establish Employee and Community Readjustment
Administration (ERCA)
Establish rules for severance payments and firm
payments to Community Readjustment Fund
Receive employer notification of closing, relocation,
or reduction in operations
Investigate failures to provide prior notice
Receive and evaluate economic impact statements
Have subpoena powers
Notify and coordinate activities of county authorities

Federal Reserve Bank of Cleveland



these hardships. It is argued that dislocated
workers usually face more severe hardships
than other unemployed workers, in terms
of unemployment duration, income losses,
and health and family difficulties. Advocates
maintain that the existing policies to protect
the unemployed in general do not address
the particular difficulties experienced by dis­
located workers. The features of plant-closing
laws that specifically address dislocated
worker hardships include cash-severance
benefits, extra paid leave time, continuation
of health insurance benefits after closing,
occupational training, job counseling and
placement, and income support.
Advocates of plant-closing laws maintain
that plant shutdowns are a major source of
local, regional, and national unemployment.
It is argued that governments have a respon­
sibility to inhibit or delay closings, thereby

Obligations of county authorities
Establish local citizens’ council within 45 days of
closing notice
Establish Community Readjustment Fund
Administer Community Readjustment Fund to pro­
vide or maintain local employment opportunities,
job finding and job creation assistance, planning
services, emergency tax relief, community devel­
opment projects
Have subpoena powers
Obligations of firm closing plant
Submit 1-year prior notice of closing, relocation, or
reduction in operations to ERCA, workers, union,
and community officials
Prepare economic impact statement within 90 days
of notice
Give lump-sum severance payment to all affected
workers with over 5 years seniority equal to
average weekly earnings over past 2 years multi­
plied by years and partial years of service
Allocate lump-sum payment to Community Readjust­
ment Fund equal to 5 percent of annual payroll
Continue health insurance coverage for affected
workers for a period not to exceed 6 months
Allow affected workers to transfer to other facilities
Allow adequate relocation reimbursement

2. While reluctance
to relocate suggests a
voluntary element
in the unemploy­
ment of dislocated
workers, it should
be remembered that
older workers are
more likely to have
stronger community
roots, families, and
homes—all of which
explain their reluc­
tance to move.

reducing the flow of workers into unemploy­
ment. The provisions of plant-closing laws
aimed at this issue include discouraging
closings through the threat of financial pen­
alties for firms contemplating such action;
providing incentives for employee ownership;
and/or requiring firms that intend to close
plants to offer preferential transfer rights
and relocation assistance to affected workers.
Another reason for supporting plantclosing laws is that shutdown-related unem­
ployment of dislocated workers can cause fis­
cal distress for local governments. Local tax
revenues might be reduced through the loss
of firm-paid property taxes, worker-paid local
income and sales taxes, and reductions in
property values. Simultaneously, a major
closing could increase government outlays
in such areas as General Assistance, Unem­
ployment Insurance, and Aid for Families
with Dependent Children.

II. Plant Closings and Hardships
for Dislocated Workers
The existing literature on plant closings is
filled with case studies that quantify the
primary and secondary effects of permanent
closings on workers and their families. Typi­
cal primary effects include extended unem­
ployment, losses in income and occupational
status, and failure to regain steady employ­
ment. Secondary effects include stress-related
health and family difficulties.
Plants often experience a lengthy period
of decline in output, productivity, capital
investment, and employment prior to closing.
As work forces decline in size, new hiring is
curtailed and pre-closing layoffs tend to be
concentrated among younger, lower-seniority
workers. Case studies show that the typi­
cal dislocated worker is older (40 years to
55 years old), has higher seniority (15 years
to 25 years) and occupational status, is less
well-educated (7 years to 10 years of formal
education), and has achieved relatively higher
earnings than other workers in the labor
pool. These five characteristics—age, senior­

Economic Review • Fall 1983



ity, occupational status, education, and
earnings—often prove to be labor-market
handicaps once such workers become unem­
ployed and begin to search for new jobs. Em­
ployers tend to screen older and less-educated
workers in the hiring process. Conversely,
many dislocated workers tend to be more
selective about the jobs they will accept,
because they have become accustomed to
higher occupational status and higher earn­
ings. The labor-market difficulties of dislo­
cated workers are sometimes compounded by
the fact that they often are dislocated from
a declining industry or during a recession;
in addition, the dislocated worker often is
reluctant to relocate to areas where job pros­
pects are better.2
The case studies indicate that, among
dislocated workers, 10 percent to 15 percent
find new jobs immediately or accept inter­
plant transfers, 10 percent to 25 percent
permanently drop out of the labor force, and
60 percent to 80 percent tend to experience
very long unemployment spells. Estimates of
the mean duration of the initial spell vary
widely (ranging from 10 weeks to 13 months),
depending on the industry involved, demo­
graphic and occupational composition of the
work force, plant location, and stage of the
business cycle at which the plant closing
occurred. Our research adjusts for time
period and business cycle by comparing the
average duration of unemployment for dislo­
cated workers in each case study with the
average length of unemployment spells for all
workers in the nation for the reference year.
We found that dislocated workers experience
average unemployment spells that are 50 per­
cent to 80 percent longer on average than the
average for all workers.
Lengthy unemployment, combined with
lower earnings in subsequent jobs, results
in permanent earnings losses for dislocated
workers. Jacobson and Thomason (1979)
found earnings diverged from their pre­
closing trend by 11 percent to 46 percent
during the first two years after closing, and
10 percent to 30 percent after three years.
Holen, Jehn, and Trost (1981) estimated that
dislocated workers’ earnings were 27 percent
below their projected earnings two years

3. See Dohrenwend
and Dohrenwend
(1974), Ferman and
Gordus (1979), and
Moen (1979).

after layoff. These two studies, along with
Holen (1976), also found that dislocated
workers’ earnings converged to the pre­
closing trend within five years of closing,
although such workers experienced a reduc­
tion in current income (see figure 1). In addi­
tion to lengthy unemployment and reduced
earnings, dislocated workers suffer a number
of secondary effects. Shostak (1980) found
higher incidence of physiological health prob­
lems, mental disorder, divorce, alcoholism,
suicide attempts, and anomie among dislo­
cated workers than he found among
employed workers.
The identification of hardships for dislo­
cated workers does not necessarily imply a
need for new policies to address such hard­
ships. Special help for dislocated workers, in
addition to existing policies for the unem­
ployed, can be justified only if the difficulties
of dislocated workers are found to be decid­
edly more severe than those of other unem­
ployed workers. While the case studies dem­
onstrate significant differences between
dislocated workers and the “average” unem­

Fig. 1
Earnings Losses of Dislocated W orkers

ployed worker, dislocated workers generally
cannot, as a group, be distinguished from
the long-term unemployed. Older, less edu­
cated workers with relatively high pre­
termination earnings and occupational status
are disproportionately represented among
the long-term unemployed, just as they are
among the dislocated. Consequently, the
labor-market and income problems of dislo­
cated workers would appear to be the same as
the difficulties faced by many other individu­
als suffering relatively long unemployment
spells. The social, personal, and family prob­
lems experienced by dislocated workers also
have been shown to be pervasive among
the long-term unemployed.3 It follows that
dislocated workers are served just as well or
just as poorly by existing unemployment pro­
tections as are other long-term unemployed
workers, and there is no strong justification
for policy targeting of the select group of dis­
located workers. The structure of existing
unemployment measures, both public and
private, provides a substantial amount and
range of protection to the unemployed in gen­
eral and to dislocated workers who are part of
that population (see table 3). Dislocated
workers are likely to benefit from unemployment-insurance coverage, as well as from
other federal and state programs, the
National Labor Relations Board, corporate
personnel policies, and union contracts.

III. Local and National
Unemployment
Several researchers have examined the
effects of plant closings on local labor
markets. Bagshaw and Schnorbus (1980)
and Aronson and McKersie (1980) found that
plant closings increased local unemployment

Time, years

6

Federal Reserve Bank of Cleveland




rates but that the increase vanished after
12 months. The local unemployment rate
returns to trend levels as a result of re­
employment, voluntary and involuntary
labor-force dropouts (e.g., retirement vs.
“discouraged workers”), and worker reloca­
tions. As a direct result of such adjustments,
the local labor market tends to shrink as the
unemployment rate once again approaches
trend levels. The speed of adjustment and
the decline in local employment levels may
affect local government tax revenues and ex­
penditures. Among three communities studied
by Aronson and McKersie (1980), two exper­
ienced slight increases in real-estate tax de­
linquencies, and one suffered a slight decline

Number of Jobs Affected by Plant Closings
There is a body of research based on the pioneering
work of Birch (1979) with the Dun and Bradstreet
Market Identifier (DMI) files to arrive at estimates of
the number of jobs affected each year by plant closings.
This research is exemplified by Bluestone and Harri­
son (1982), Weiss and Shapira (1982), and Leighton,
Roderick, and Folbre (1981). Bluestone and Harrison,
for example, found that between 1969 and 1976 on
average 3,000,000 jobs per year were terminated
through closings. This figure implies that plant clos­
ings are the most important single source of unem­
ployment in the United States.
However, the DMI files seem to be inappropriate
data for deriving plant-closing estimates. Because the
files are not designed for social-science research, ana­
lysts have faced significant methodological difficulties
in using them to answer policy questions (see Birch
1979, Miller 1980, and Harris 1983). The DMI files
contain no evidence of motive for inter-year “disap­
pearances” of business establishments. It is likely that
a large proportion of inter-year disappearances reflect
acquisitions, mergers, changes of address, and Dun and
Bradstreet data collection and input errors.
Even if the DMI files could provide a crude assess­
ment of job losses attributed to plant closings, the
figure cited by Bluestone and Harrison would be
grossly overestimated from a public-policy standpoint.
Most existing and proposed plant-closing laws use 100
as a minimum number of employees for a firm to be
obligated under statute. Yet, the empirical results
reported by Birch (1979) suggest that the vast majority
of inter-year disappearances are firms that employ
50 or fewer workers.

Economic Review • Fall 1983



in sales-tax revenues. In all three cases, the
divergence from trend was evident only in the
year of the plant shutdown. Neither declines
in real-estate values nor increases in local
outlays for social services were found in any
of the communities (although one reported a
marked increase in food-stamp distribution).
Plant closings are considered to be a
major source of U.S. unemployment and a
significant constraint on the expansion of
U.S. employment. Appraisals of the magni­
tude of the plant-closing problem have been
hampered by measurement difficulties and
a lack of historical information. Since 1979,
Conway Publications has tracked closings
of U.S. manufacturing plants, and the Cali­
fornia Employment Development Department
(CEDD) has collected similar data for all
industries in California. The Bureau of
National Affairs (BNA) began to track the
national private nonfarm business sector in
1982. Conway and BNA base their estimates
on many bibliographic sources and an exten­
sive network of contacts with trade associa­
tions and federal, state, and local public
agencies. CEDD’s estimates are based pri­
marily on the state’s ES-202 data collection
program, conducted under the auspices of the
national payroll survey of the Bureau of
Labor Statistics.
These data sources are the basis for 1982
estimates of the number of jobs lost through
plant closings. The estimates can be com­
pared with measures of labor-market activity
to determine the extent to which plant clos­
ings cause labor-market problems in the
aggregate (see table 4). While the estimates
differ in terms of geographic coverage, indus­
try, and underlying definition, it is readily
apparent that dislocated workers do not con­
stitute a very large proportion of the national
labor force, employment, and unemployment.

In 1982, for example, the BNA recorded 619
plant closings involving 215,500 jobs. Unem
ployment spells of workers dislocated by

Table 3

these closings accounted for less than 1 per­
cent of the total unemployment spells expe­
rienced in the United States in 1982. Sim-

Existing Protection for Dislocated Workers
Protection

Coverage

Applicability

Federal and State
Unemployment compensation

Income maintenance for unemployed

89% of U.S. workers in 1982

Comprehensive Employment and
Training Act, Titles IIB, IIC,
and VII

Employment, training, and counseling
for structurally unemployed adults

In FY 1982, 835,234 individuals
served by IIB and IIC;
125,994 served by VII

Trade Adjustment Assistance

Income maintenance and CETA-like
services for workers adversely
affected by imports

30,480 workers received first TAA
payments in FY 1982

U.S. Employment Service

Job referral, counseling, and other
supportive services

14.3 million workers served
in FY 1982

Redwood National Park Act
amendments of 1978

Income supplements for workers
adversely affected by expansion of
Redwood National Park

Primarily California-based
lumber workers

Civil Aeronautics Board and
Airline Deregulation Act
of 1978

Income supplements and job guarantees
for workers adversely affected by
mergers and airline deregulation

Workers in air-transportation
industry

Regional Rail Reorganization Act
of 1973, Urban Mass Transpor­
tation Act, and High Speed
Ground Transportation Act
of 1965

Income supplements to workers
adversely affected by these pieces
of legislation

Mainly railroad workers, espe­
cially with Amtrak, Conrail,
and their predecessors

Employee stock ownership
plans (ESOPs)

Internal Revenue Code and Treasury
regulations offer very favorable tax
treatment to ESOPs

ESOPs nationally; potential
ESOPs in 6 states in 1982

Prior notice of separation

National Labor Relations Board (NLRB)
decisions require that unions have
sufficient time to bargain over the
rights of workers affected by closings,
out-sourcing, and subcontracting

Union workers

Effects bargaining

NLRB decisions require that unions be
allowed to discuss effects of closing;
usually upheld by courts

Union workers

Decision bargaining

NLRB decisions require that unions be
given the opportunity to bargain over
decision to close, out-source, or sub­
contract; usually overturned in courts

Union workers

Health insurance

Conversion to individual policy with­
out waiting period or physical
examination upon layoff

40 states in 1982

Federal Reserve Bank of Cleveland



ilarly, dislocated workers constituted less
than 1 percent of the total number of workers
who experienced unemployment spells in the

Protection

year. Dislocated workers achieved a some­
what larger representation among the long­
term unemployed and among unemployed

Coverage

Applicability

Federal and State (cont.)
Health insurance (cont.)

Employer required to continue health
insurance coverage for 3 months
to 12 months after layoff

22 states in 1982

Severance pay

Separation for economic cause, 2-week
lump-sum payment to 2 weeks pay
times years of service in lump sum

66% of companies and 56% of
workers in 1977

Supplemental unemployment
benefits

Supplements to income after layoff

15% to 27% of workers in 1981

Relocation assistance and/or
retraining

Reimbursement of relocation expenses;
company-paid retraining

45% of companies in 1977

Outplacement counseling

Counseling, job referral, supportive
services

53% of companies in 1977

Continuation of life and/or
medical insurance

Company continues to pay insurance
premiums for period of 1 month
to 3 years

70% of companies in 1977

Advance notice of separation

1-6 months advance notice; mode of
2-4 weeks (47% of companies)

95% of companies in 1977

Prior notice of closing

2 weeks to 1-year prior notice

19.8% of union workers in 1980

Decision bargaining

Negotiations with union over
decision to close

1.5% of union workers in 1982

Effects bargaining

Negotiations with union over
effects of closing

10.1% of union workers in 1982

Supplemental unemployment
benefits

Income supplements to laid-off workers

27.7% of union workers in 1980

Severance pay

Lump-sum payment upon termination

37.8% of union workers in 1980

Relocation assistance

Reimbursement for relocation expenses

32.0% of union workers in 1980

Transfer rights

Preferential transfer and bumping
rights, with retention of seniority

49.2% of union workers in 1980

Corporate Personnel Policies

Collective Bargaining
Agreements

SOURCES: U.S. Department of Labor, (1972, 1981a, 1981b, 1982); Scarry (1982); Miner (1978); Gorlin (1981); Industrial Union Department,
AFL-CIO (1982); Gacek (1981); Millen (1979); Goldfarb (1980); and Freedman (1978).

Economic Review • Fall 1983



4. The empirical
literature on plant
relocations also
suggests that the
number of workers
affected by intra­
state moves is small
relative to labormarket activity.
Birch (1979) found
that emigration
of entire facilities
across state lines
affected between
0.03 percent and
0.10 percent of
the national labor
force each year from
1969 to 1976.
M iller (1982) con­
cludes that “reloca­
tions play a very
m inor role in real­
locating manufac­
turing employment
among regions . . .
and that when relo­
cations do occur, the
majority involve
short-distance, intra­
state moves by non­
corporate affiliated
employers” (p. 34).

manufacturing workers.4 Based on the CEDD
count, plant closings and worker dislocation
would appear to be more common in Califor­
nia than in the nation as a whole. However,
the CEDD defines plant closings and worker
dislocation more broadly than the other
sources. While BNA and Conway count only
complete plant closings and the number of
workers immediately affected by such actions,
CEDD measures job losses from the plant’s
employment peak during the 24 months prior
to closing and includes partial closings and
“massive” permanent layoffs from continu­
ing plants that affect at least 50 percent of a
facility’s work force.
The foregoing analysis demonstrates
that dislocated worker hardships are not
unique; indeed, such hardships are “special”
only insofar as the dislocated constitute
roughly 5 percent of the long-term unem­
ployed. Plant-closing laws are intended to
make our economic system fairer and more
equitable, yet their narrow applicability
creates inequities. If the existing structure
of unemployment protection is inadequate
to the needs of the dislocated worker, it is
likewise inadequate for the much larger mass
of long-term unemployed. It follows that the
most appropriate response for public policy
would involve the modification and expan­
sion of the existing structure of unemploy­
ment assistance that would more adequately
address the needs of all long-term unem­
ployed workers, including dislocated workers
experiencing relatively lengthy unemploy­
ment spells.

IV. The Timing of Plant Closings
The existing and proposed plant-closing laws
generally require firms to make a lump-sum
severance payment to dislocated workers,
equivalent to their most recent weekly earn­
ings times the number of years of service for
each worker. Firms with no severance-pay

protection or with an existing arrangement
less generous than required by law would
be affected by this provision. Advocates of
plant-closing legislation argue that such laws
should discourage and/or delay closings,
thereby maintaining local employment stabil­
ity and helping local governments to avoid
fiscal distress. Unfortunately, existing plantclosing laws are too recent, have little history
of enforcement, and are not sufficiently wide­
spread to facilitate an empirical analysis of
their economic effects. Thus, the analysis in
this article relies on theoretical tools to indi­
cate the direction but not the magnitude of
the economic effects of severance-pay re­
quirements. In this section, we examine
whether severance-pay requirements and
resulting adjustments by firms would influ­
ence the timing of plant closings and, if so,
in what direction.
We assume that firms would continue
to operate a plant as long as the net present
value of future returns from operation re­
mains greater than the net present value
of returns from closing, or, alternatively, as
long as the internal rate of return (IRR) from
operation is larger than the IRR from closing.
Let
T

2 pt/(1

r) T-s

t=s

represent the net present value of an oper­
ating plant’s profit stream over a period of
T-s + 1 years, where the plant becomes obso­
lete at year T. The symbols Pt and r denote
profit in the year t and the discount rate,
respectively. The Pt stream is assumed to be
nonnegative for simplicity. The net present
value of returns from continuous operation of
the plant as of year s (NPVOs) can be
expressed as

(1) NPVOs = 2
t=s

(1 + r)t-s

+

SVT - TSPt
(1 + r)

T-s

where SV^and TSPt are the plant’s scrap
value and total severance payments at
10

Federal Reserve Bank of Cleveland




closing, respectively, in year T. The net
present value of returns from immediate
closing of the plant as of year s (NPVCS)
may be written as

(3) NPVCS > NPVOs ,
or, equivalently,
T

(2) NPVCS = SVs - TSPs .

(4)

Closing of the plant at year s rather than at
year T makes the firm better off if, and only
if, the following condition is satisfied:

SVs - TSPs > 2 J-s
t=s (1 + r)
SVT - TSPt
+
a + r) T-s

Table 4 Losses from U.S. Plant Closings: 1982
Alternative sources of dislocation data relative to measures of labor-market activity

Measures

BNA:
U.S., all
industries

BNA:
U.S. m anu ­
facturers

Conway:
U.S. m anu ­
facturers

CEDD:
California,
all industries

619

424

72

304

215,500

146,900

63,723

57,406

Unemployment spells, %

0.64

na

na

na

Workers who experienced
unemployment, %

0.96

2.49

1.08

na

Unemployment spells exceeding
6 months, %

4.27

11.30

4.90

na

Labor force, %

0.20

0.64

0.28

0.47

Employment, %

0.22

0.72

0.31

0.52

Unemployment, adjusted %b

0.95

2.48

1.08

2.23

Unemployment exceeding
6 months, adjusted %b

5.07

11.96

5.19

15.76

Number of plant closings
Number of jobs affected
Annual Work Experience
Supplement to Current
Population Survey3

Current Population Survey3

a. The percentages reflect the estimated contribution of dislocated workers to 1982 unemployment spells,
workers experiencing unemployment, and unemployment spells of more than six months duration (from the
Annual Work Experience Supplement to Current Population Survey, Bureau of Labor Statistics) and annual
averages for labor force, employment, unemployment, and unemployment of more than six months duration
(from the BLS Current Population Survey).
b. To derive a more precise measure of the contribution of plant closings to the average monthly level of
unemployment and the average monthly level of workers unemployed for at least six months, we have adjusted
the number of dislocated workers to reflect case-study findings on the average duration of dislocated worker
unemployment. Aronson and McKersie (1980) provided supplementary unpublished data that were particularly
helpful in making these adjustments.

Economic Review • Fall 1983



Changes in the severance-pay formula (TSPs
and TSPt ), as well as changes in scrap
values (SV^ and SVT), the profit stream (Pt),
and the discount rate (r), can change a firm’s
decision on the optimal timing for closing by
changing the direction of the inequality in
equation 4. As an introduction to our main
concern—the effects of changes in the sev­
erance-pay formula—let us first discuss the
potential effects of changes in r, Pt, SV^, and
SVT. First, ceteris paribus, an increase in the
profit stream (Pt) or the scrap value at year T
(SVT) would delay the closing by increasing
NPVOs, while an increase in the scrap value
at year s (SJ£) would accelerate the closing by
increasing NPVCS. Second, the effect of
an increase in r on optimal timing of clos­
ing is uncertain, depending on the sign of
PT + SVT - TSPt . T o describe this more
clearly, we may rewrite equation 1 by moving
PT/( 1 + r)T~s from its first term to its second
term, as follows:
T-

1

p

(5) NPVOs = X -----t=s (1 + r)1 5
PT+ SVT- TSPt
(1 + r)T~s
Assuming the profit stream to be nonnega­
tive (Pt > O, t = s, . . T-1) and to be posi­
tive for at least one period, an increase in the
discount rate (r) always decreases the first
term of equation 5. The discount rate sensi­
tivity of the second term of equation 5
depends on the sign of PT + SVT - TSPf. an
increase in r decreases, does not affect, or
increases the second term of equation 5 as
PT + SVT- TSPt is positive, zero, or neg­
ative, respectively. When the two terms are
combined, an increase in r unambiguously
decreases NPVOs and accelerates the clos­
ing as long as PT + SVT - TSPt is nonnega­
tive. If the expression PT + SVT - TSPj is
negative, an increase in the discount rate

Federal Reserve Bank of Cleveland



reduces the first term and increases the
second term in equation 5, thereby making
the effect of a discount rate increase on the
sequence of closing-related events uncertain.
An increase in TSPs alone, ceteris paribus,
would delay closing by decreasing the value
of NPVCS\likewise, an increase in TSPt
alone would accelerate closing by decreasing
NPVOs. However, the effect of changes in the
severance-pay requirement on the optimal
timing of closing is uncertain when TSPs and
TSPt change simultaneously, depending on
the size of the discount rate (r) relative to the
growth rate of TSP over time (expressed as
m). Let us begin from a situation in which
NPVCSequals NPVOs, or, equivalently,
I
p>
(6) SFs - TSPs = X -----t=s (1 + r f s

SVT
(1 + r)T~s

TSPt
(1 + r)T~s

Then the net present value of returns from
closing is the same as that from remaining in
operation until time T, and there is no incen­
tive for a firm to advance its expected tim e of
closing from T toward s. If concurrent
changes in TSPs and TSPt are such that
(7) ATSPs = ATSPt /(1 + r)T~s ,
then the changes do not alter the equilibrium
of equation 6, since both sides of equation 6
are simultaneously being changed by the same
amount. However, closing would be acceler­
ated if TSPs and TSPt change such that
(8) ATSPs < ATSPt /(1 + r)T~s,
because NPVCSdecreases less rapidly than
NPVOs. Equation 8 may be rewritten as

5. The foregoing
analysis was re­
stricted to the effects
of severance-pay
requirements on
plants already
located in a given
state and effects
on the local labor
market. The analy­
sis would not, how­
ever, be complete
without brief com­
ment on the effects
of plant-closing laws
on potential new in ­
vestments in a given
locality. The litera­
ture on this capitalflow question has
been lim ited by the
absence of empirical
evidence. However,
McKenzie (1982),
among others, has
argued that plantclosing laws reduce
the attractiveness to
new investment by
increasing the cost
of doing business in
a place that enacts
such a law. This
assertion is sup­
ported by the lit­
erature on the
effects of certain
discretionary gov­
ernment policies,
such as environ­
mental protection.
We have some res­
ervations concern­
ing the applicability
of these examples to
plant-closing laws,
and we reserve judg­
ment on the issue
of plant-closing laws
and capital inflows,
leaving the question
to further research.

13

(9)

A TSPs

>

(1 + r)

T-s

V. Distortions of Resource
Allocations

A TSPt

or,
ATSPt
(10)

y

----- +
A TSPs

Assuming, for simplicity, that the total
severance-pay obligation (TSP) changes over
time at a constant annual rate of m, such
that TSPt = TSPs(1 + m)T~s, equation 10 may
be rewritten as
ATSPs(l + m)
(1 1 )

Having shown that the timing of a plant clos­
ing can be accelerated by plant-closing laws,
we now shall show that severance-pay re­
quirements can change the optimal mix of
production inputs by changing the perceived
prices of the inputs. In the framework of
comparative static analysis, we assume that
a single-plant firm maximizes its profits
(PRF) subject to its production function
Q= Q (K,L),

T-s

> (1 + r)

T-s

A TSPC

where K and L denote capital and labor,
respectively, and
(14) max PRF = P Q (.K,L) - WL - RK,

( 12)

(1 + m)T~s > (1 + r)T' s

(13)

m > r.

Thus, the optimal time of closing can be
accelerated by severance-pay requirements
if employers observe that the growth rate of
severance-pay liability (m) exceeds the dis­
count rate (r). The size of m depends on the
structure of the laws and on technical and
institutional constraints that each plant
faces: m is likely to be larger for plants whose
production functions allow less substitution
between capital and labor and less substitu­
tion between skilled and unskilled labor.
Likewise, plants will have greater difficulty
in reducing m if their production functions
are relatively inflexible, and if the work place
is governed by relatively restrictive work and
seniority rules. We therefore conclude that
the severance-pay provision would not unam­
biguously delay or otherwise inhibit plant
closings. Indeed, depending on institutional
and technological conditions, severance-pay
requirements may result in earlier closings
than otherwise would be the case. Thus, it is
possible that plant-closing laws could accel­
erate both flows into unemployment and the
pace of local employment decay.5

Economic Review • Fall 1983




where P, W, and R represent prices of the
product (Q), labor, and capital, respectively.
In the short run, where capital is fixed and
labor is assumed to be the only variable
input, PRF is maximized when the wage rate
equals the value of labor’s marginal product.
Taking the partial derivative of PRF
with respect to L yields the following firstorder condition:
(15)

W= P

dQ
dL

where the second-order conditions are as­
sumed to be satisfied. In the long run, when
the capital stock is allowed to change, profit
maximization requires that the price of capi­
tal equal the value of its marginal product:
(16) R = P

dQ
dK

In the long run, the profit-maximizing input
combination (K/L)* before the imposition
of the severance-pay requirement can be ob-

6. Another way
to look at this is
through the insu­
rance concept. I f
a company could
insure against the
new liability of sev­
erance pay, assum­
ing that no moral
hazard prevailed,
the insurance pre­
mium that the firm
would like to pay to
the insurance com­
pany per each em­
ployee would be
equal to FXW, the
amount of wage-rate
increase offered by
the firm .

14

tained from equation 17, which is derived by
dividing equation 16 into equation 15:
(17)

W _ dQ/dL
R

dQ/dK

That is, profits are maximized when the ratio
of each factor’s marginal product equals
the ratio of each factor’s price.
Suppose plant-closing laws would neces­
sitate severance payments in the amount of
SP per eligible employee. Although the
amount of the severance pay does not in­
crease the “actual” price of labor as long as
the plant is in operation, it certainly would
increase the “contingent” price of labor
unless the firm’s subjective probability of
closing were zero. Consider a hypothetical
situation where plant-closing laws allow
voluntary transactions between employers
and employees regarding the newly imposed
severance-pay requirement and where the
transactions costs are relatively low. The
firm apparently is “worse off” than in the
case of no severance payment, but the
workers are “better off” because they can
anticipate additional income should the plant
be closed. Such an imbalance creates an
environment conducive to trading between
the employer and employees. Conceivably,
the firm might prefer to increase the current
level of workers’ wages in exchange for the
workers’ voluntary releasing of the firm from
the severance-pay obligation. Likewise,
employees may prefer an increase in current
wage levels in exchange for future contingent
severance pay.
Such transactions would not occur if
bargaining failed to achieve a wage increase
satisfactory to both parties. Technically
speaking, this would take place if the firm’s
maximum wage offer (FXW) remained below
the minimum wage increase acceptable to
workers (ENW ). The optimal wage increase
would, of course, lie between FXW and ENW,
but the probability of satisfactory agreement

Federal Reserve Bank of Cleveland




depends on the relative bargaining strength of
the parties and the structure of negotiations.
It is not our purpose to elaborate on the
likelihood that such voluntary transactions
would arise, or the actual determination of
the equilibrium level of wage increases. It
suffices, rather, to indicate that the subjec­
tive wage rate has increased from W to
W +FXW, and the subjective input price
ratio between L and K has increased from
W/R to {W + FXW)/R as a result of the
severance-pay requirement.6 The level of
FXW would be higher if the firm were rela­
tively risk averse and if the level of SP were
relatively high. Excluding the extreme case
of perfect factor substitution, the resulting
increase in the subjective wage rate would
raise the marginal cost of production, in turn
reducing the profit-maximizing level of out­
put. In the short run, the entire burden of the
output reduction is borne by labor: as the
wage rate rises from W to W + FXW, the use
of labor input would be curtailed to satisfy
equation 15.
In the long run, the increase in the mar­
ginal cost of production would be smaller
than in the short run, because the capital
stock would no longer be fixed. Therefore,
excluding the extreme cases of zero or infi­
nite factor substitution, the profit-maxi­
mizing level of output would fall with the
imposition of severance pay, but by less than
in the short run. In contrast to the short-run
effects, the burden of reduced output in the
long run is shared by capital and labor.
Nevertheless, labor still would shoulder a
disproportionate share of the burden arising
from reduced output and from the substitu­
tion effect: since the severance-pay require­
ment increases the subjective input price
ratio, the right-hand part of equation 17 must
increase to satisfy the profit-maximizing con­
dition. With the assumption of decreasing
marginal product, this implies that the
optimal input combination (K/L)* must
increase through a greater reduction in labor
input (L) than in capital input (K). These

would be perverse and unintended conse­
quences of the severance-pay requirement—
reduced production and reduced labor
demand—considering that increased employ­
ment constitutes a key objective of plantclosing laws.
Just as severance pay can affect the
quantity of labor used in production, it may
also affect the quality of labor. For conven­
ience, we define skilled labor as higher-wage
and higher-seniority workers. Suppose the
optimal input mix of capital and labor has
been pre-determined with regard to the con­
siderations discussed previously. Plant man­
agement may wish to manipulate labor qual­
ity and quantity to reduce the contingent
liability associated with severance-pay
requirements. Assuming a production func­
tion that allows substitution between the
types of labor, plants could hire more skilled
labor to keep overall labor productivity con­
stant and reduce the number of workers. Let
us suppose that the severance payment to
employee h (SPh) is a simple function of
minimum-service years (MSY), years of
seniority (SYh), and current wage level (Wh),
as follows:

In this case, SPh is a flat payment of equal
amounts to employees regardless of years of
service or wage level. Plants can reduce TSP
by hiring more skilled labor and reducing
their overall number of employees (//), while
keeping total labor productivity constant.
(b) a > 0, b > 0.
In this case, SPh is a flat payment of equal
amounts to each employee, plus a variable
payment that increases in accordance with
Wh and SYh (as long as SYh is greater than
MSY). The direction of employment impact is
uncertain, because an increase in skilled
labor and a reduction in total employment
use reduces TSP by lowering the first term of
equation 19, a ■H, but increases TSP by rais­
ing the second term (by increasing SYh
and/or Wh). Thus, the direction of change
depends on the relative sizes of a • H on the
one hand, and
H

b • 5) (SYh - MSY) • Wh
h=l
on the other.

(18) SPh = a + b • [(SYh - MSY) • Wh].
Total severance payments (TSP) for the plant
may then be defined as
H

(19)

TSP = X { a + b- [(SYh - MSY) • Wh]}
h= 1
H

= a • H + b - ]g(SYh - MSY) • Wh,
h=1

where H is the total number of employees.
Consider three combinations of signs of a
and b, as follows:
(a) a > 0, b = 0.

Economic Review • Fall 1983



(c) a - 0, b > 0.
There is no flat payment in this case, and
SPh is a function solely of SYh, MSY, and Wh.
This severance-pay formula is more relevant
than the previous two, since it closely resem­
bles the formulas incorporated in existing
and proposed plant-closing legislation. If
there were a requirement for minimum years
of service to be eligible for severance pay (five
years in most laws), plants may wish to add
new workers relative to high-seniority skilled
workers, attempting to reduce TSP by
decreasing SYh - MSY and Wh. Firms can
unambiguously reduce TSP by deliberately
hiring new, unskilled workers and letting
them go before they have worked MSY years.

7. Indeed, given
that the case-study
evidence shows that
some dislocated
workers experience
no unemployment
or only brief spells
of unemployment
and still qualify
for benefits, plantclosing laws may
exacerbate the in ­
equities between
employed and un­
employed workers.
8. I f plant-closing
laws were more
effective policy
instruments, their
implementation
might harm na­
tional economic effi­
ciency. Less capital
and labor would be
available for their
most productive
and profitable uses.
Resources already
located in the
affected jurisdic­
tion would be
trapped in unpro­
ductive pursuits,
and resources that
might otherwise
enter the jurisdic­
tion would be dis­
couraged by higher
operating costs.

Therefore, in the absence of effective worker
opposition, plants have incentives to increase
work-force size, decrease average seniority
and average wages, and increase employee
turnover. Early retirement of high-seniority
workers is one of many methods to reduce
average seniority. As a result, high-skilled,
high-seniority workers would tend to be
worse off with this severance-pay formula.

VI. Policy Implications
Plant closings can result in serious hardship
for dislocated workers and their families,
including periods of extended unemploy­
ment, losses in income and occupational
status, deterioration of health, and family dif­
ficulties. Yet, the foregoing analysis shows
that such hardships (alone or in combination)
are not unique to dislocated workers. The
same kinds of hardships are prevalent among
the long-term unemployed, of which dislo­
cated workers constitute a relatively small
part. Plant-closing laws are intended to make
our economic system more fair and more
equitable, by transferring scarce economic
resources to the unemployed. It follows that
plant-closing laws might help equalize the
disparity between the employed and the
unemployed members of our society. The
laws have the opposite effect, however, within
the pool of unemployed workers. Plantclosing laws are designed to benefit an arbi­
trarily defined subgroup of the unemployed:
to qualify, a worker must have lost his/her
job in a permanent facility shutdown. Dislo­
cated workers receive additional benefits not
available to other unemployed workers who
share the hardships of unemployment.7
Plant-closing laws do not appear to be
effective policy instruments. The laws are
intended to inhibit closings, yet the standard
severance-pay requirement, in many cases,
would create incentives for plants to close
earlier than they would otherwise. While the
laws are intended to reduce the number of
unemployed workers (especially highseniority workers), they may create incen­

Federal Reserve Bank of Cleveland



tives to reduce overall labor use, increase
employee turnover, and shed high-seniority
workers more rapidly.8
The design of plant-closing laws could be
modified to make them less ineffective policy
instruments, although the adverse effects
on within-group distributional equity would
remain. The inherent bias against highseniority workers might be eliminated by ex­
changing the severance benefit based on
years of service and pay for a fixed severance
payment per worker. This approach would
not, however, eliminate the incentives to use
less labor per unit of output. A second pos­
sibility would involve shifting the penalty for
closing from a labor-based to an asset-based
formula: if firms engaged in closings were
required to pay a proportional tax on the
plant’s scrap value at closing, the laws
might delay closings in the short run without
causing distortions in the optimal mix of
productive resources. This approach would
not, however, provide much direct financial
assistance to dislocated workers, and it might
create incentives for firms to keep plants
operating with only skeletal work crews.
A more comprehensive unemployment
assistance policy, aimed at the structurally
unemployed and financed by tax revenues
and experience-rated employer payments,
would avoid many of the equity, efficiency,
and design problems associated with plantclosing laws. Such a policy would deliver
general- and specific-skills training, income
support, career and job-search planning
services, and relocation assistance to the
unemployed. It could also include additional
incentives for the establishment of ESOPs.
This proposal closely resembles the existing
structure of unemployment protections, and
it could be built on a well-developed adminis­
trative delivery system and substantial past
program experience. While the existing frame­
work of unemployment protection is by no
means perfect (and could certainly benefit
from increased coordination), both dislocated
workers and the pool of unemployed workers
probably would obtain greater benefit from
a broad-based approach than from the selec­
tive intervention of plant-closing laws.

References
Aronson, Robert L., and Robert B. McKersie.
Final Report: Economic Consequences of
Plant Shutdowns in New York State. Alba­
ny, NY: State Employment and Training
Council, 1980.
Bagshaw, Michael L., and Robert H.
Schnorbus. “The Local Labor-Market
Response to a Plant Shutdown,” Economic
Review, Federal Reserve Bank of Cleveland,
January 1980, pp. 16-24.
Birch, David. The Job Generation Process.
Cambridge, MA: MIT Program on Neigh­
borhood and Regional Change, 1979.
Bluestone, Barry, and Bennett Harrison. The
Deindustrialization of America.
New York, NY: Basic Books, 1982.
Bureau of National Affairs. Layoffs, Plant
Closings, and Concession Bargaining:
Summary Report for 1982. Washing­
ton, DC: BNA, 1983.
Carlisle, Rick, and Michael Redmond. Com­
munity Costs of Plant Closings: Bibliography
and Survey of the Literature. Washing­
ton, DC: C&R Associates for the Federal
Trade Commission, 1978.
Cropper, Maureen, and Louis Jacobson. The
Earnings and Compensation of Workers
Receiving Trade Adjustment Assistance.

Alexandria, VA: Public Research Institute,
1983.
Dohrenwend, Barbara S., and Bruce P
Dohrenwend, Eds. Stressful Life Events:
Their Nature and Effects. New York, NY:
John Wiley & Sons, Inc., 1974.
Ferman, Louis, and Jeanne Gordus. Mental
Health and the Economy. Kalamazoo, MI:
W.E. Upjohn Institute for Employment
Research, 1979.
Freedman, Audrey. Security Bargains Recon­
sidered: SUB, Severance Pay, Guaranteed
Work. Report No. 736. New York, NY: Con­
ference Board, 1978.

Economic Review • Fall 1983



Gacek, Stanley A. “An Employer’s Duty to
Bargain on the Termination of Unit
Work,” Labor Law Journal, vol. 32, nos. 10
and 11 (October and November 1981),
pp. 659-78, 699-724.
Goldfarb, Robert S. “Compensating Victims
of Policy Change,” Regulation,
October 1980, pp. 22-30.
Gordus, Jeanne Prial, Paul Jarley, and
Louis A. Ferman. Plant Closings and Eco­
nomic Dislocation. Kalamazoo, MI:
W.E. Upjohn Institute for Employment
Research, 1981.
Gorlin, Harriet. Personnel Practices I:
Recruitment, Placement, Training, Com­
munication. Bulletin 89,1981; Personnel
Practices II: Hours of Work, Pay Practices,
Relocation. Bulletin 92, 1981; Personnel
Practices III: Employee Services, Work
Rules. Bulletin 95, Conference Board, 1981.
Hammerman, Herbert. “Five Case Studies of
Displaced Workers,” Monthly Labor
Review, vol. 87, no. 6 (June 1964),
pp. 663-70.
Harris, Candee S. Small Business and Job
Generation: A Changing Economy or Differ­
ing Methodologies. Washington, DC: Brook­
ings Institution, 1983.
Hekman, John, and John Strong. “Is There a
Case for Plant Closing Laws?” New Eng­
land Economic Review, Federal Reserve
Bank of Boston, July/August 1980,
pp. 34-51.
Holen, Arlene. Losses to Workers Displaced by
Plant Closures or Layoff: A Survey of the
Literature. Arlington, VA: Public Research
Institute, 1976.
_____ , Christopher Jehn, and Robert D.
Trost. Earnings Losses of Workers Displaced
by Plant Closings. Alexandria, VA: Public
Research Institute, 1981.
Industrial Union Department, AFL-CIO.
Comparative Survey of Major Collective
Bargaining Agreements. Washington, DC:
AFL-CIO, 1982.

Jacobson, Louis. Community-Wide Labor Ad­
justment to Declining Employment. Alexan­
dria, VA: Public Research Institute, 1982.

Analysis of the Studebaker Shutdown,”
Industrial and Labor Relations Review,
vol. 21, no. 2 (January 1968), pp. 234-50.

______, and Janet Thomason. Earnings Losses
Due to Displacement. Alexandria, VA: Cen­
ter for Naval Analyses, 1979.

Scarry, Donald M. Employee Stock Ownership
Plans: A Programfor NewJersey. Trenton, NJ:
New Jersey Department of Labor, 1982.

Leighton, Julia C., Melissa R. Roderick, and
Nancy R. Folbre. “ ‘Pick Up Your Tools and
Leave—The Mill is Down’ Plant Closings
in Maine, 1971 - 1981.” Brunswick, ME:
Bowdoin College, 1981.

Schumpeter, Joseph A. Capitalism, Socialism
and Democracy. New York, NY: Harper &
Row Publishers, Inc., 1950, pp. 81-6.

Lipsky, David B. “Interplant Transfer and
Terminated Workers: A Case Study,”
Industrial and Labor Relations Review,
vol. 23, no. 2 (January 1970), pp. 191-206.
McKenzie, Richard B., Ed. Plant Closings:
Public or Private Choices? Washington, DC:
Cato Institute, 1982.
Mick, Stephen S. “Social and Personal Costs
of Plant Shutdowns,” Industrial Relations,
vol. 14, no. 2 (May 1975), pp. 203-08.
Millen, Bruce H. “Providing Assistance to
Displaced Workers,” Monthly Labor Re­
view, vol. 102, no. 5 (May 1979), pp. 17-22.
Miller, James P. Nonmetro Job Growth and
Locational Change in Manufacturing
Firms. Washington, DC: U.S. Department
of Agriculture, 1980.
______ “Manufacturing Relocations in the
United States, 1969-75,” in Richard B.
McKenzie, Ed. Plant Closings: Public or
Private Choices? Washington, DC: Cato
Institute, 1982.
Miner, Mary Green. Separation Procedures &
Severance Benefits, Personnel Policies
Forum Survey No. 121. Washington, DC:
Bureau of National Affairs, 1978.
Moen, Phyllis. “Family Impacts of the 1975
Recession: Duration of Unemployment,”
Journal of Marriage and the Family, vol. 41,
no. 3 (August 1979), pp. 561-72.
Palen, J. John, and Frank J. Fahey. “Unem­
ployment and Reemployment Success: An

Senia, Al. “A Dark Cloud of Plant Closings
Hangs over Sunny California,” Iron Age,
vol. 226, no. 8 (March 16, 1983), pp. 28-31.
Sheppard, Harold L., Louis A. Ferman, and
Seymour Faber. Too Old to Work—Too
Young to Retire: A Case Study of a Perma­
nent Plant Shutdown. Washington, DC:
Government Printing Office, 1960.
Shostak, Arthur B. “The Human Cost of
Plant Closings,” AFL-CIO American Federationist, vol. 87, no. 8 (August 1980),
pp. 22-5.
Site Report, vol. 2, nos. 1-6, Atlanta, GA:
Conway Publications, Inc., 1982.
Stern, James. “Evolution of Private Man­
power Planning in Armour’s Plant Clos­
ings,” Monthly Labor Review, vol. 92,
no. 12 (December 1969), pp. 21-8.
______ “The Consequences of Plant Clo­
sure,” Journal of Human Resources, vol. 7,
no. 1 (Winter 1972), pp. 3-25.
U.S. Department of Labor, Bureau of Labor
Statistics. Employee Benefits in Industry,
1980. Bulletin 2107, 1981; Employee Bene­
fits in Medium and Large Firms. Bul­
letin 2140, 1982; Layoff, Recall and Work­
sharing Procedures. Bulletin 1425-13, 1972;
Plant Movement, Interplant Transfer and
Relocation Allowances. Bulletin 1425-20,
1981.
Weiss, Marc A., and Philip Shapira. Series of
Briefing Papers on Assembly Bill 2839,
nos. 1-5. Berkeley, CA: University of Cali­
fornia, 1982; processed.
Wilcock, Richard C. “Employment Effects of
a Plant Shutdown in a Depressed Area,”
Monthly Labor Review, vol. 80, no. 9 (Sep­
tember 1957), pp. 1047-52.

Federal Reserve Bank of Cleveland



This article was
abstracted from a
more comprehensive
Federal Reserve
System staff study,
“Effects of the Ser­
vice Contract Act
on the Transporta­
tion Contracts of the
Federal Reserve Sys­
tem, the Expenses
of the Federal Re­
serve Banks, and
the Efficiency of the
N ation’s Payments
Mechanism, ”
October 1982;
processed.

Prevailing Wage Laws,
the Federal Reserve,
and the Service
Contract Act
by Mark S. Sniderman

Mark S. Sniderman
is an assistant vice
president and econ­
omist, Federal
Reserve Bank of
Cleveland. Merphil
Kondo provided
many and substan­
tial contributions to
this research pro­
ject. Robert Goldfarb offered helpful
comments and sug­
gestions, as did
Jennifer Johnson,
Dan Littman,
Walker Todd,
Richard Patterson,
Richard Harris,
Robert Gay, and
many other Federal

Reserve System
employees.

19

Economic Review • Fall 1983




Private firms entering into certain kinds of
contracts with the federal government must
agree to compensate their employees working
on the contracts according to standards set
forth by the U.S. Department of Labor. The
various laws from which the Department of
Labor derives this authority are informally
called prevailing wage laws, since their intent
is to guarantee that compensation of em­
ployees performing work for the federal
government is no lower than compensation
for comparable jobs in that geographical loca­
tion. Economists are typically critical of pre­
vailing wage laws, because there is virtually
no evidence to suggest that the equity consid­
erations that motivate such laws outweigh
the economic inefficiencies they promote.
The Federal Reserve System complies with
the requirements of one of these prevailing
wage laws, the Service Contract Act (SCA),
in engaging contractors to furnish various
services to the Federal Reserve Banks. Em­
ployees of these engaged contractors are com­
pensated according to criteria established
by the Department of Labor (DOL) in Wash­
ington, DC. One effect of compliance with the
SCA is costs of contract services to Federal
Reserve Banks that are greater than would
be expected if the SCA did not exist. These
greater Reserve Bank expenses might be
construed as evidence that the SCA is having
its intended effect, i.e., raising the compen­
sation of covered employees. However, this
article presents evidence that the SCA actu­
ally boosts compensation rates above the
average levels of the marketplace. And, since
the Federal Reserve is mandated by law to
sell its services to depository institutions at
fees that reflect the full direct and indirect
costs of providing those services, the SCA
therefore causes the Federal Reserve to set
higher fees than would occur in the act’s
absence. The pricing obligation stems from
the Depository Institutions Deregulation
and Monetary Control Act of 1980, wherein
Congress sought to encourage competition
between the Federal Reserve and the private

1. Robert S. Goldfarb and John E
M orrall III, " The
Davis-Bacon Act:
An Appraisal of
Recent Studies, ”
Industrial and
Labor Relations
Review, vol. 34,
no. 4 (Janu­
ary 1981),
pp. 191-206.

sector so that the nation’s payments system
would become more efficient. Evidence also
suggests that the SCA will gradually lose its
influence over labor costs for the transporta­
tion of checks, coins, and currency.
Compliance with the requirements of the
SCA hampers the Federal Reserve’s ability to
achieve the objectives of the Monetary Con­
trol Act. The problem is most acute in the
area of contracts for the transportation of
checks, coin, and currency. To the extent that
transportation costs are greater for the Fed­
eral Reserve than for its actual or potential
competitors, the Reserve Banks cannot press
for payments system services that are costeffective. This article describes the labormarket consequences of the SCA, with par­
ticular emphasis on Federal Reserve System
costs and operations.

1982, for example, the Reserve Banks pro­
cessed approximately 15 billion checks
(almost one-half of the nation’s total), 11 bil­
lion pieces of currency, 17 billion coins,
and 58 million electronic funds transfers.
Since all payments mechanism services are
provided directly to depository institutions
and government agencies, the public has little
opportunity to observe these functions.
The Monetary Control Act (MCA) requires
the Federal Reserve System to establish a
schedule of fees to be charged to depository
institutions for certain Federal Reserve servi­
ces, including check clearing, check collec­
tion, and transportation of currency and coin.
The MCA requires that these services be
made available to all depository institutions,
with the same fee schedules and other terms
of service for all depository institutions.

I. The Nation’s Payments System
and Priced Federal Reserve
Services

II. The SCA and Prevailing
Wage Laws

A tremendous number and variety of trans­
actions take place each year among U.S. con­
sumers, businesses, depository institutions,
and governments. Most of these transactions
require money payments for their completion,
either in cash or in transfers among deposit
accounts at depository institutions. The sys­
tem for making money payments often is
referred to as the payments mechanism, a
complex structure of instruments, proce­
dures, and financial institutions. Payments
usually are timely and convenient, and the
cost per transaction of making payments is
typically very low compared with the value of
the payment.
The Federal Reserve System is a unique
part of the payments mechanism. The Fed­
eral Reserve Banks provide currency and coin
to banks and other depository institutions,
and participate in the clearing of a major
share of all checks and other payments in­
struments, such as money orders, travelers’
checks, and electronic funds transfers. In

20

Federal Reserve Bank of Cleveland




The SCA is the most recent of the nation’s
three prevailing wage laws. The others are
the Davis-Bacon Act, which applies to fed­
eral construction contracts, and the WalshHealey Act, which applies to federal supply
contracts. Virtually all economic research on
prevailing wage laws centers on the DavisBacon Act, the oldest of the prevailing wage
laws; data availability appears most condu­
cive to Davis-Bacon analysis. The DavisBacon studies point clearly to many ineffi­
ciencies in the contracting process for federal
construction.1 Although the SCA covers dif­
ferent government purchases from DavisBacon, the economic principles used to evalu­
ate Davis-Bacon are appropriate to the
assessment of the SCA.
Most proponents of prevailing wage laws
argue that, in their absence, locally sited
federal agencies either might purchase
cheaper labor services from distant locations
or use their market power to drive local
wages below a more natural level. Thus,
these wage laws are supposed to insulate
local labor markets from competitive forces

2. Federal Regis­
ter, vol. 46, no. 157
(August 14, 1981),
p. 41397.

21

external to the local market, or to protect
workers from a potentially exploitative
employer, the federal government. In the
absence of prevailing wage laws, the federal
government presumably would incur lower
expenses in the conduct of its business.
However, the greater labor costs currently
incurred have been justified by proponents on
equity grounds, i.e., that taxpayers should
pay more taxes so that employees of govern­
ment contractors are treated fairly.
Critics of prevailing wage laws claim that
the federal government rarely has enough
market power to affect the wages of nonfederal employees in most localities, and that
there is little evidence that federal agencies
have imported cheaper labor from one local­
ity to another. Even if this importation did
occur, it could be justified on the grounds
that there is insufficient competition in the
“high-wage” market. Furthermore, critics
argue that prevailing wage laws, at best,
transfer money from taxpayers to already
well-paid employees. At worst, these laws
distort labor markets and promote inefficien­
cies by artificially increasing the cost of labor
and reducing the number of jobs.
Aside from these theoretical considera­
tions, a practical issue commands substantial
attention. The most valid objectives of pre­
vailing wage laws can be somewhat cor­
rupted through the administrative process.
Many critics fault the methods of DOL in its
determination of the prevailing wages for
occupations in a locality. Indeed, accurately
collecting the required data for literally hun­
dreds of occupations in each of hundreds of
localities is an immense task, especially if the
information must be current. Not surpris­
ingly, DOL’s collection and interpretation of
the required labor market data have been
subject to intense criticism.
The administrative challenge of the SCA
can better be understood by considering how
the SCA is actually implemented by DOL.

Economic Review • Fall 1983




The following description of this process
appeared in the Federal Register.
The SCA establishes standards for minimum
compensation and safety and health protection of
employees performing work for contractors and
subcontractors on service contracts entered into
with the Federal Government and the District of
Columbia. It applies to contracts entered into pur­
suant to negotiations concluded or invitations for
bids . . . The provisions of the Act apply to con­
tracts, . . . the principal purpose of which is to
furnish services in the United States through the
use of service employees. Under its provisions,
every contract subject to the A c t. . . must contain
stipulations .. . requiring, (a) that specified min­
imum monetary wages and fringe benefits deter­
mined by the Secretary of Labor (based on wage rates
and fringe benefits prevailing in the locality or, in
specified circumstances, the wage rates and fringe
benefits contained in a collective bargaining agree­
ment applicable to employees who performed on a pre­
decessor contract) be paid to service employees by the
contractor or subcontractor. . . (emphasis added]

There are two types of relevant wage and
fringe-benefit determinations: (1) those pre­
vailing in the locality, and (2) those contained
in collective bargaining agreements applica­
ble to previous contractors. The latter deter­
minations often are required when a contrac­
tor is awarded a contract that had previously
been awarded (i.e., not a first-time contract),
even if the contractor is the incumbent. If
a contract were awarded to a new contractor,
and if the previous contractor had a collec­
tive-bargaining contract with his employees,
the successor contractor must pay his SCAcovered employees at rates no lower than the
previous contractor would have paid his
covered employees.
In either of these two cases, after the con­
tracting Reserve Bank notifies DOL of its
intent to contract for services, DOL informs
the Reserve Bank of the minimum wages any
contractor must pay his service employees
who perform on that contract. This informa­
tion is conveyed to the Reserve Bank in the
form of a document called wage determina­
tion. According to DOL regulations, the wage
determination is based on either (1) wage and
fringe benefits “prevailing in the locality” of

the service to be performed, as determined by
DOL, or (2) the terms set forth in applicable
collective-bargaining agreements. Thus, once
a unionized firm services a contract, it is dif­
ficult for future contracts to entail wages
paid below the union wage, regardless of how
unrepresentative the union wage may be of
the market wage.
Prevailing wages may be determined with
reference to Area Wage Surveys conducted
by the Bureau of Labor Statistics (BLS) at
DOL. The DOL may also consider informa­
tion provided by employers, employees, and
their trade associations. In particular,
collective-bargaining agreements may be used
as a readily available source of information.
Where a single wage rate is paid to a majority
of workers in a job category in a certain local­
ity, that wage rate automatically is deter­
mined by DOL to be the prevailing wage. For
example, if no more than 1,000 janitors are
employed in a certain locality and 501 or
more janitors are paid identical wages, their
pay standards automatically become the deter­
mined wage. Situations such as this are likely
to occur only when a collective-bargaining
agreement is in effect, because equal wage
and fringe benefits (i.e., to the penny) among
so many people do not usually occur other­
wise. When prevailing wages cannot be deter­
mined through this majority method, DOL
may use either the median or mean wages of
the BLS survey and other data.
DOL is permitted a substantial amount of
judgment in determining prevailing wages.
In part, this judgment is exercised through
its choice of “locality” and its evaluation
of compensation data once the locality is
defined. The DOL enjoys great latitude in
selecting locality, being able to use city,
county, SMSA, state, or regional designa­
tions. Regardless of the procedure used by
DOL in arriving at its wage determination,
the contracting agency and other interested
parties (actual or potential bidders, employee
organizations, other government agencies)
may appeal. Appellants may claim that the
wage determination varies substantially from

Federal Reserve Bank of Cleveland



true prevailing wages, or that DOL erred in
its process by ignoring relevant data, or incor­
rectly defined the geographic locality.
Consider a hypothetical Federal Reserve
Bank that requests bids to service an
armored-carrier route (see figure 1). In this
example, the successful bidder services 12
depository institutions along the route, with
initial and terminal stops at the Reserve
Bank. Notice that all depositories are located
in counties other than the county in which
the Reserve Bank is located. The Reserve
Bank is located in the center of a large met­
ropolitan area; the depositories are located
outside this area in smaller cities. The crucial
issue related to the bidding process for this
route is the determination of the locality to be
served. Under current regulations, once a
locality is designated, all bidders—regardless
of their location—must conform to the pre­
vailing wages in that locality. For many
situations, this policy is reasonable, espe­
cially where the work is to be performed at a
single location such as a particular building.
In the case of transportation service con­
tracts, however, the service may not inher­
ently be confined to a small geographic area.
Should the locality be the city in which the
Reserve Bank is located, or the surrounding
metropolitan area? Or, should the locality be
one or both of the counties in which the de­
positories are located? Or, would a combina­
tion of the counties and the metropolitan area
be most appropriate? Since the wages that
prevail in one locality may differ significantly
from those prevailing in another, the choice of
locality very much influences both the wage
determination and the winning bid price.
The definition of locality also can influence
the set of the actual bidders. In our example,
suppose that DOL chooses the city in which
the Reserve Bank is located as the locality,
and that wages prevailing there are 20 per­
cent greater than wages prevailing in the out­
lying counties. Since the contract must be
awarded to a contractor paying the wages

■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ flH n n H n H H B S B B n H H H

prevailing in the Reserve Bank’s city, con­
tractors based in the outlying areas (i.e.,
counties A and B) may not bid on the route.
Were one to win the bid, that contractor
would be required to pay some of its em­
ployees 20 percent more on the Reserve
Bank’s route than they are paid on other
routes. This situation could be considered
untenable and thus would discourage bidding
by those contractors.
Under new rules adopted by DOL, a twostep wage-determination process would be
used in cases where the performance of the

Fig. 1

service is unknown at the time of bidding.
An example would be a contract for data pro­
cessing that could be performed in any one of
several localities. First, the contracting
agency would request bids and identify all
interested bidders; DOL then would issue a
wage determination to each bidder, requiring
conformance to wages prevailing in his local­
ity. This procedure certainly could be super­
ior to the current rules. It is unclear whether
DOL would agree that transportation con­
tracts of Federal Reserve Banks are siteunknown at the time of bidding.

Relationship between a Reserve Bank and an Armored-Carrier Route

County A

County C

Qi

Depository in non-metropolitan area

Depository in metropolitan area

Each grouping represents a city.




County B

3. Federal Register,
vol. 45, no. 157
(August 14, 1981),
pp. 41382-83.
4. Robert S. Gold­
farb and John S.
Hey wood, "A n
Economic Evalua­
tion of the Service
Contract A ct,”
Industrial and
Labor Relations
Review, vol. 36,
no. 1 (October
1982), pp. 56-72.

III. Labor-Market Consequences:
The Wage Impact of SCA Coverage
In the SCA, the judgment of the federal
government is substituted for the judgment
of the marketplace, at least where the wages
of government service contractors are con­
cerned. The SCA has no effect on most ser­
vice contractors when wage determinations
are at or below market wages. When deter­
minations are greater than market rates,
however, there are direct and indirect effects.
The direct SCA effect is the cost increase
associated with paying greater-than-market
wages. One indirect SCA effect is the longerrun cost of restricting the number of vendors
who would bid on government contracts. If
the number of participants in a market is
appreciably reduced, contract bidding can
become less than competitive. This is espe­
cially important in the case of transportation
services, where federal and state laws
already limit by regulation the number of
carriers in a market.
The methodology used to calculate the SCA
impact on an occupation in an area is quite
explainable. First, sum the wages actually
paid to all of the employees in the occupation
in the given area. This dollar amount is
called the wage bill. Next, locate the em­
ployees paid less than the determined wage.
Sum up, for these individuals, the differences
between their actual wages and the deter­
mined wage. This calculation yields the SC4
burden, because it assumes that all belowSCA wage rates must be raised to SCA levels.
Finally, the SCA burden is divided by the
wage bill and converted to a percentage fig­
ure. This step puts the dollar figure for
the SCA burden in perspective relative to the
dollar size of the wage bill. Mathematically,
the formula is
N
X f ( wi)(wD - wi)

SCA impact = -----------X 100 ,
M

7= 1

24

Federal Reserve Bank of Cleveland




where
wD =
wi =
f(w •) =
N =

the wage determined by DOL,
the *th wage wi < wD,
the number of employees earning w{,
the number of different wage rates
less than WD in the occupation,
M = the total number of different wage
rates in the occupation.

Using this methodology in a recent analysis
of economic impact, the DOL concluded that
in 1981, for 23 technical and clerical occupa­
tions in 30 areas of the United States, the
SCA impact was only 4 percent.3 Several
points need to be raised regarding this find­
ing. The DOL did not state how either the
occupations or the locations were selected for
its analysis. Nor can one determine how rep­
resentative the sample is of all contracts
subject to the SCA. Furthermore, DOL under­
stated the overall impact of SCA wage deter­
minations by choosing technical and clerical
occupations. Because collective-bargaining
agreements are so important to some wage
determinations (especially via successorship
requirements), SCA burdens should be rela­
tively larger in transportation-related occupa­
tions than clerical and technical occupations
because of the relatively stronger union pres­
ence in the transportation industry. In addi­
tion, the wage measure should, in principle,
include only private-sector wages not already
affected by the SCA.
Goldfarb and Heywood (1982) provided an
estimate similar to that of DOL (5 percent)
for a 1981 sample of 6 technical and clerical
occupations in 11 cities.4 However, Goldfarb
and Heywood noted that in their sample
the SCA effect was as small as 0 percent (for
a draftsman in Memphis) and as large as
12 percent (for a draftsman in Columbus).
This degree of variation in the SCA effect is
of critical importance because it indicates
that it is possible for the SCA to increase
wages consistently in a locality regardless of
occupation or in an occupation regardless of
locality. The variation demonstrates that the

5. The Fourth Fed­
eral Reserve District
includes all of the
state of Ohio, west­
ern Pennsylvania,
eastern Kentucky,
and the panhandle
of West Virginia.

SCA might regularly increase wages by a
large amount for some occupations without
appreciably raising aggregate wages in a
large sample of occupations and localities.
Consequently, the SCA impact on wages may
be seen as moderate when viewed across a
large sample of occupations and/or localities.
However, a government agency that must
contend with the SCA for a few occupations
or localities may view the SCA impact as
being quite large.
For this study I compared wage determina­
tions issued by the DOL with Area Wage
Survey data compiled by the BLS for 12 areas
in which Reserve Banks purchase transporta­
tion services. Although my sample of loca­
tions was not random, it was not chosen to
reveal unusual circumstances; the same
methodology should provide similar findings
for other areas. The chosen occupations—
drivers and guards—constitute virtually
100 percent of labor costs for armored trans­
portation services purchased by Federal
Reserve Banks.
The SCA burden was shockingly greater
when applied to transportation service
employees than for employees in technical
and clerical occupations (see table 1). For
armored-car drivers, the SCA impact ranged
from 1 percent in Buffalo, NY, and Hart­
ford, CN, to 30 percent in Poughkeepsie, NY.
For armed guards, the SCA impact ranged
from 5 percent in New York City, NY, to
59 percent in Pittsburgh, PA. Most impacts
for guards were close to 30 percent. Across
all locations and both occupations, the SCA
impact averaged 17 percent. These findings
indicate that wage determinations applicable
to Federal Reserve Bank contracts generally
exceed market wages.
Surely, the differences between the
nationwide SCA impact of 4 percent cited by
the DOL and these figures for drivers and
guards must lead to the conclusion that Fed­
eral Reserve transportation contracts are
considerably more costly because of the SCA.
These added expenses must show up in the
prices charged for Federal Reserve services
because of the requirements of the MCA.
Economic Review • Fall 1983




IV. A Case Study: The Fourth
Federal Reserve District5
The Cleveland experience demonstrates how
the SCA can dramatically increase transpor­
tation expenses. The Federal Reserve Bank of
Cleveland solicits bids in some areas from
common carriers, which are exempt from the
SCA by statute. Common carriers are trans­
port companies that receive freight from the
public and charge according to tariffs estab­
lished by a regulatory body (state or federal).
Common carriers are exempt from the SCA,
because their freight rates are already regu­
lated. Carriers not common are contract carri­
ers that ship according to agreements nego­
tiated by the firm and its customers. Many
Federal Reserve Banks cannot take advantage
of the common-carrier exemption to SCA, be­
cause carriers licensed to operate in their dis­
tricts are often only contract carriers. Few
Federal Reserve Banks have many opportuni­
ties to solicit bids from several competing
common carriers. When competitors bid for
Federal Reserve transportation routes, a
common carrier paying market wages usually
has an edge over contract carriers paying
above-market wages.
During summer and fall 1978, the Pitts­
burgh office of the Federal Reserve Bank of
Cleveland bid 18 intrastate armored routes
regulated by the Pennsylvania Public Utili­
ties Commission. Company A, a contract car­
rier, was the dominant armored carrier in
western Pennsylvania, although independent
common carriers and one small contract car­
rier provided armored service in some re­
stricted areas. Prior to bidding, company A
serviced the 18 routes.
Company B, a Pennsylvania common car­
rier, was the dominant carrier in eastern
Pennsylvania, but did not operate out of
Pittsburgh although licensed to do so. Com­
pany B bid $30,000/month on 15 of the 18
routes, while company A bid $68,000/month.
Company B’s costs were lower, in part,
because it hired at a market wage below the
union wage paid by company A and because

it staffed at levels not subject to the work
rules in company A’s union contract. Com­
pany B was awarded these 15 routes, and
company A was awarded the remainder. Al­

though the Cleveland Fed was subject to the
SCA at the time it awarded routes to com­
pany B, company B was exempt from the
SCA because of its common-carrier status.

Table 1 Im pact of SCA on Wage Costs in Selected Cities in 1980
Location/
category

Boston, MA
Drivers
Guards
Buffalo, NY
Drivers
Guards
Cincinnati, OH
Drivers
Guards
Eugene, ORb
Drivers
Guards
Hartford, CT
Drivers
Guards
Los Angeles, CAC
Drivers
Guards
New York, NY
Drivers
Guards
Pittsburgh, PA
Drivers
Guards
Portland, ME
Drivers
Guards
Portland, ORb
Drivers
Guards
Poughkeepsie, N Y d
Drivers
Guards
Trenton, NJd
Drivers
Guards

Area Wage Survey

SCA-determined
wage, dollars

Wage impact,
percent

8.77
8.63

7.56
52.55

8/1980

Medium truck drivers (all)
Guards (Class A)

5.41
5.32

1.26
16.34

10/1980

Medium truck drivers (all)
Guards (Class A)

8.03
7.91

10.60
19.08

7/1980

Medium truck drivers (all)
Guards—manufacturing (all)

8.08
8.52

1.74
6.43

8/1981

Medium truck drivers (all)
Guards—manufacturing (all)

5.65
6.20

1.29
14.21

3/1980

Medium truck drivers (all)
Guards—manufacturing (Class A)

8.28
7.94

12.69
30.59

10/1979

7.82
7.67

7.32
5.09

5/1980

Medium truck drivers (all)
Guards—manufacturing (Class A)

8.27
8.22

8.71
59.15

1/1980

Medium truck drivers (Mfg)
Guards (Class A)

5.95
5.85

18.22
11.89

12/1980

Medium truck drivers (all)
Guards—manufacturing (all)

9.70
9.53

2.59
29.39

6/1981

Medium truck drivers (all)
Guards—manufacturing (all)

7.82
7.67

29.86
31.63

6/1980

Medium truck drivers (all)
Guards—manufacturing (all)

7.87
7.67

2.49
21.80

9/1980

Medium truck drivers (all)
Guards—manufacturing (all)

Date

Wage distribution3

Medium truck drivers (all)
Guards (Class A)

a. Occupations selected represent the closest available category incorporating drivers and guards in each
Area Wage Survey.
b. SCA-determined wages not available for Eugene and Portland, OR, in 1980.
c. Wage determination of May 1979.
d. New York City’s determinations were assigned to carriers in these cities, as was done by the DOL.

Federal Reserve Bank of Cleveland




Had the services of company B not been
available and its routes awarded to company
A, the Cleveland Fed would have paid an
additional $456,000/year for the transporta­
tion services on these 15 routes.
In May and June 1979, seven remaining
armored routes running out of Pittsburgh
were bid, five intrastate and two interstate.
Company A bid $39,000/month for all seven
routes. Company C, a common carrier, bid
$11,000/month for the five intrastate routes.
Company B, now operating as a contract car­
rier, bid $9,000/month for the two interstate
routes. The combined total of the bids from
companies B and C ($20,000/month) was
52 percent of company A’s bid. It is notable
that company B raised its bid from $6,000 to
$9,000 only 14 days after discovering that it
was subject to SCA’s wage provisions. Com­
panies B and C received awards for these
seven armored-carrier routes. One year later,
company B lost its two interstate routes to a
new market entrant operating as a common
carrier. The routes rate fell from
$9,000/month to $5,600/month. In addition,
company A ceased operation in Pittsburgh. At
this time, company A had routes costing
$5,000/month; these routes were awarded to
common carriers for $3,800/month.
In spring 1981, armored-carrier routes in
the Cincinnati, OH, area were bid. All carri­
ers in Ohio were obliged to adhere to a wage
determination under the successor provisions
of the SCA. The Federal Reserve Bank of
Cleveland was informed by one carrier, who
had the necessary operating authority, that
he would not bid for the routes because his
wage rates were 50 percent below the wage
determination. If this potential bidder had
been an actual bidder, he would have bid
about $22,000/month. Costs before the bid
were $38,000/month; costs after the bid were
$42,000/month. Because, in the absence of
the SCA, contract costs would have been
about $22,000/month, the impact of the SCA
raised the contract price by roughly
$20,000/month, or $250,000/year.

Economic Review • Fall 1983




Currently, approximately $100,000/month
is paid for transportation contracts elsewhere
in Ohio on routes that have never been bid
competitively among common carriers. Real­
izing that these routes would soon be bid, a
number of contract armored carriers applied
for and received permission from the Public
Utilities Commission of Ohio to convert to
common carrier status. Consequently,
through competitive bidding, the Cleveland
Fed reduced its expenses for these routes by
about 40 percent, or nearly $500,000/year.

V. Summary and Conclusions
The SCA increases the federal government’s
cost of obtaining many services, even when
those services are contracted through a
“competitive” bidding process. Although the
SCA, in theory, protects labor markets
against federal government abuses, in prac­
tice the federal government may abuse labor
and product markets through its applica­
tion of the SCA. The usual consequences of
such circumstances are revenue losses to the
Treasury, fewer services provided to the
public, and fewer people employed.
For some government agencies, econom­
ically efficient operation is not of paramount
importance. In these agencies, the SCA and
other prevailing wage laws may encumber
operations, but they probably do not conflict
with prime directives. This situation is not
the case for the Federal Reserve.
In the Monetary Control Act, Congress in­
structed the Federal Reserve System to pro­
mote an efficient payments mechanism in the
United States by pricing Federal Reserve ser­
vices at their long-run costs. Because of the
SCA, some Federal Reserve costs are greater
than private-sector (competitive) costs; thus,
Federal Reserve prices for some services are
not as likely to force depository institutions
to be more efficient.
The Federal Reserve services that are most
affected by the SCA are those requiring
ground transportation—check processing and
cash services. These services are somewhat
labor intensive and therefore price-sensitive

to labor costs. Data from DOL wage determi­
nations and Area Wage Surveys indicate that
the SCA increases the wages that Federal
Reserve contractors must pay for truck
drivers and armed guards by 17 percent on
average. The increases generally are quite
large for drivers and even more substantial
for guards.
A difficulty associated with estimating the
SCA burden on Federal Reserve transporta­
tion activities is the fact that transportation
services themselves are usually regulated
by some government agency. Even in the
absence of the SCA, some Federal Reserve
Banks would find few firms licensed to ser­
vice some of their routes. The SCA screens
firms from already thin markets. Firms with
common-carrier status, exempt from the
SCA, are frequently able to offer substantial
savings, a point that is amply illustrated
by the case study of the Fourth Federal
Reserve District.
Despite the paucity of qualified firms com­
peting in some localities, deregulation of the
trucking industry is steadily leading to more
competition for Federal Reserve business
throughout the nation. The Federal Reserve
System has been expanding its contractor
base during the past few years, and it will
continue to do so. While the SCA will impede
this diversification, it will not prevent it.
In markets where the SCA dramatically in­
creases Federal Reserve costs, depository
institutions not subject to the SCA may con­
tract for their own transportation services. A
Federal Reserve Bank simply would become
another stop on the route. Market forces
eventually would push market participants
toward the most efficient mode of service
provision, regardless of the SCA. Those highwage jobs currently “protected” by the SCA
would be lost to more efficient contractors.
The problems in letting market forces
alone undermine the SCA are twofold. The
process may take a considerable amount of
time, especially because the transportation
and financial services industries are not com­
pletely competitive. Furthermore, the Federal
Reserve would not be able to discipline the
Federal Reserve Bank of Cleveland




market with low-cost service if it were the
only competitor affected by the SCA.
Variance hearings are a cumbersome
method for eliminating the SCA burden.
Nevertheless, Federal Reserve Banks have
successfully used, and are currently using,
this process to alleviate undue SCA burdens.
Variance hearings are not completely satis­
factory for another very important reason,
however. In some labor markets, wages might
be high precisely because the SCA prevents
potential competitors from entering the
market. When this happens, it might be diffi­
cult to estimate correctly the true market
wage for drivers and guards in the armoredcarrier industry.
Unfortunately, there seems to be no simple
remedy for the Federal Reserve. The DOL
may grant exemptions to the SCA when it
is “necessary and proper in the public inter­
est” and to avoid the “serious impairment of
Government business.” The Federal Reserve
has requested from the DOL an exemption
for its priced transportation services in the
interest of a more efficient national payments
system. In rejecting the request, DOL did not
agree that the statutory criteria for exemp­
tion were met.
If the social goals of the SCA are indeed
valid, then merely showing that they inter­
fere with other valid social objectives does
not, by itself, imply an exemption is war­
ranted. However, the evidence and analysis in
this article argue that the DOL cannot ensure
Federal Reserve contracts to high-wage firms
in transportation markets that are becoming
increasingly populated by SCA-exempt firms.
Furthermore, as private financial institutions
become more aware of the profit potential in
these transportation services, they will seek
to engage contractors directly and compete
against the Federal Reserve. In the end,
although DOL may not exempt the Federal
Reserve System from the SCA, market forces
operating through the explicit pricing of Fed­
eral Reserve services will ensure that market
wage rates ultimately prevail. The potential
inefficiencies of the SCA will become less
important because of the law’s increasing
ineffectiveness.

The Federal Reserve
Bank of Cleveland
publishes an infor­
mative research
periodical called
Economic Commen­
tary. Following are
the titles published
since January 1982.
I f you are interested
in receiving this
publication, either
future or back issues,
please contact our
Public Information
Center, Federal Re­
serve Bank of Cleve­
land, P.O. Box 6387,
Cleveland, OH
44101.

Economic
Commentary
Financial Services and Small Businesses
Paul R. Watro
1/11/82
Methods of Cash Management
John B. Carlson
4/05/82
Unemployment Insurance: An Old
Lesson for the New Federalism?
Michael E Bryan
4/19/82
Bank Holding Companies’ Participation
in Credit Insurance Underwriting
Paul R. Watro
5/03/82
The Steel Trigger Price Mechanism
Gerald H. Anderson
5/17/82
The Problem of Seasonally
Adjusting Money
John B. Carlson
5/31/82
Performance of Ohio’s
Independent Banks
Gary Whalen
6/14/82
Union Wage Concessions
Daniel A. Littman
6/28/82
Anatomy of a Price-Fix
Michael E Bryan
7/12/82
The Strength of Consumer
Balance Sheets
KJ. Kowalewski
7/26/82
Safe-Harbor Leasing: Separating the
W heat from the Chaff
Amy L. Kerka and Owen E Humpage
10/04/82

29

Economic Review • Fall 1983




The Shift to Western Coal
Gerald H. Anderson
10/18/82
Do Deficits Cause Inflation?
Owen E Humpage
11/01/82
Social Security: Issues and Options
Amy Kerka
1/10/83
Soil Conservation: Market Failure and
Program Performance
Paul Gary Wyckoff
1/24/83
Issues in the 1983 Auto-Sales Outlook
Michael E Bryan
3/07/83
Loan Quality of Bank Holding Companies
Gary Whalen
3/21/82
Economic Outlook for 1983
Paulette Maclin and Joanne Bronish
4/04/83
Exchange Rates and U.S. Prices
Gerald H. Anderson and Owen E Humpage
4/18/83
Velocity and Monetary Targets
William T. Gavin
6/06/83
The Mythology of Domestic Content
Michael E Bryan
6/20/83
Economic Recovery and the
Fourth District
Robert H. Schnorbus and Sandra Pianalto
7/05/83
Geographic Banking Markets
Paul R. Watro
9/12/83
The Japanese Postal Savings System:
A State-Run Financial Monster?
Laura A. Kuhn
9/26/83
Banking and Commerce:
To Mix or Not to Mix?
Thomas M. Buynak
12/5/83

Michael L. Bagshaw
is a statistician with
the Federal Reserve
Bank of Cleveland;
W illiam T. Gavin
is an economist,
also with the Fed­
eral Reserve Bank
of Cleveland.

Working Paper
Review
Michael L. Bagshaw and
William T. Gavin
Forecasting the Money Supply
in Time Series Models
Working Paper 8304,
December 1983.
23 pp. Bibliography.

In this paper the authors develop a multi­
variate simultaneous equation model of
the dynamic correlations among growth in
money, credit, the interest rate, output, and
prices. The Tiao-Box multivariate time series
procedure is used to identify and estimate
the model.
The Tiao-Box procedure is interactive,
similar in principle to that used in single­
equation Box-Jenkins modeling. The steps in­
volved are (1) tentatively identifying a model
by examining autocorrelations and cross­
correlations of the series; (2) estimating the
parameters of this model; and (3) applying
diagnostic checks to the residuals. If the
residuals do not pass the diagnostic checks,
then the tentative model is modified and
steps 2 and 3 are repeated. This process con­
tinues until a satisfactory model is obtained.
The model was estimated using both
pre-seasonally adjusted and “raw” data. As
one would expect, the conventional census
X-ll seasonal adjustment procedures intro­
duce a significant change in the dynamic
cross-correlations among the variables.

Federal Reserve Bank of Cleveland



Using not-seasonally adjusted data results
in a forecasting model that is block recur­
sive with two independent leading blocks, the
price equation by itself, and the money and
interest-rate equations. The credit equation
depends on the money and interest-rate block.
The output equation depends on both leading
blocks. This result suggests that a bivariate
model including just the interest rate and
M-l would predict M-l as well as the fivevariate model. Both should outperform a uni­
variate model of the money supply process.
Using seasonally adjusted data results in
a block recursive forecasting model in which
the credit equation forms the leading block,
the money and interest equations form the
second block, the inflation equation is the
third block, and the output equation is the
final block. In this case the forecasts of M-l
from the five-variable model should outper­
form the bivariate and univariate models.
Forecasts of M-l from the five-variate
model were compared with forecasts from
univariate and bivariate models. The results
from the forecasting experiment were mixed.
In five of the eight experiments, the fivevariate model gave better forecasts than the
smaller models. In two of the other cases, the
results were very close. This was a turbu­
lent period for monetary policy. The Federal
Reserve adopted a new operating procedure in
October 1979. The change in regimes was fol­
lowed by unpredicted swings in the interest
rate and more volatile growth in the money
supply. In spite of this, the out-of-sample
quarterly prediction error of M-l was on the
order of 1 percent when we intervened for the
period of credit controls. This error is of the
same magnitude as that found when stan­
dard econometric models are used. Overall,
the forecasting results from this short period
do not distinguish sharply between the three
time series models.

Economic Trends

Economic Trends is published monthly by the
Research Department of the Federal Reserve
Bank of Cleveland. Originally a booklet of
charts, the periodical now includes charts
and explanations of what is happening in a
particular series. In addition, the periodical
includes a current overview of the economy,
“The Economy in Perspective.”
Different series are featured in each month’s
Economic Trends, depending on current eco­
nomic events. Following is a list of contents
of the December 1983 Economic Trends:
Gross National Product and Components
Industrial Production
Capacity Utilization
Personal Income
Retail Sales
Auto Production and Sales
Housing
Capital Investment
Business Inventories and Sales
Labor Compensation
Consumer Prices
Producer Prices
Federal Budget
Money Supply: M-l
Money Supply: M-2
Reserve Aggregates
Money Markets
Capital Markets
Stock Market Measures
Bank Loans
Financial Condition of Savings and Loans
U.S. Merchandise Exports
Foreign Exchange Rates
If you are interested in receiving this pub­
lication on a regular basis, please contact
our Public Information Center, Federal
Reserve Bank of Cleveland, P.O. Box 6387,
Cleveland, OH 44101 (216/579-2000).

Economic Review • Fall 1983



The working paper
series is published by
the Research Depart­
ment of the Federal
Reserve Bank of
Cleveland to stim ­
ulate discussion and
critical comment.
Copies of working
papers are available
through our Public
Information Center,
Federal Reserve
Bank of Cleveland,
PO. Box 6387,
Cleveland, OH 44101.

Working Paper
Series

8101
The Welfare Implications of Alternative
Unemployment Insurance Plans
Mark S. Sniderman
April 1981, 20 pp.
8201
M ultibank Holding Company Organiza­
tional Structure and Performance
Gary Whalen
March 1982, 30 pp.
8202
Stability in a Model of StaggeredReserve Accounting
Michael L. Bagshaw and William T. Gavin
August 1982, 27 pp.
8203
A Micro View of the Transactions
Money Market
Mark A. Zupan
September 1982, 31 pp.
8301
Non-Nested Specification Tests and the
Intermediate Target for Monetary Policy
Mitsuru Toida and William T. Gavin
June 1983, 14 pp.
8302
Holding Company Organizational Form
and Efficiency
Gary Whalen
July 1983, 20 pp.
8303
Extension of Granger Causality in
M ultivariate Time Series Models
Michael L. Bagshaw
August 1983,11 pp.
8304
Forecasting the Money Supply in
Time Series Models
Michael L. Bagshaw and William T. Gavin
December 1983, 23 pp.

32

Federal Reserve Bank of Cleveland