BREAKING NEWS! WARNING! A CALL TO ACTION

UP-COMING >>> MAJOR ISSUE >>> HEADLINES
UNFAIR TRADE: MEANS THE END OF US PROSPERITY, SOVERENTY; AND, DEMOCRACY
STOP
WHAT?

STOP 10s of 1000s [layoffs???] "FIRINGS'
AMERICAN PLANT CLOSURES! You are killing America!

Rebuild Our Manufaturing Base; and sell High Tech

US Products! Do Not Sellout America - Its People!
It is BUY-AMERICAN! Not; Come Here and Buy America!

Stop Dismembering and Exporting America! Piece By Painful Piece!!!

STOP! Isuing visas to foreign workers to replace American Tech Professionals! Stop importing cheap labor! These are our educated - highly-skilled - children!

Stop it now!!!

---

STOP

---

THE CONGRESS
AND
THE ADMINISTRATION

---

FROM SELLING-OUT
THE
REMAINDER OF AMERICA.

GLOBALIZATION
 

---

THE GREAT

GLOBALIZATION DEBATE

HAS COMMENCED

Kudos to Journalist Thomas Friedman and Senator Byron Dorgan for courageously appearing on PBS to discuss this controversial topic. This is a question , we, as a Nation, must decide.  Once we know the  direction we want to  take 21St Century America; getting there,  will not obstruct determined Americans.

---

SOME SITE RESEARCH SUBJECTS:
Globalization 

Mad-Rush Dump USA 

Dump U.S. = Dump U 

Dump GM, Ford Delphi 

MergeGM,FORD,Delphi 

UnTrade 

Predatory Trade 

Dumping Imports 

Defect. Component Parts 

Defect. Military Parts 

Exploit Global Poor 

Trade Slavery 

Sweat Shops 

China: 

Communist Aristocrats 

Slave-Waged Chinese 

Tade Deficit 

Prison Child Female Labor 

China Price 

Wal-Martization 

China Militarism 

In Western Hemisphere?

Unspinning The Globe

 

Printer Friendly Version
EMail Article to a Friend

Page 1
Centre for Research on Globalisation and Labour Markets, School of Economics,
University of Nottingham
The Centre acknowledges financial support from The Leverhulme Trust under
Programme Grant F114/BF
CENTRE FOR RESEARCH ON GLOBALISATION AND LABOUR MARKETS
Research Paper 2000/9
Do Hostile Mergers Destroy Jobs?
by
M. Conyon, S. Girma, P. Wright and S. Thompson
Page 2
The Authors
Martin Conyon is Reader and Head of the Corporate Performance Research Unit in Warwick
Business School, Sourafel Girma is Research Associate in the Leverhulme Centre for Research
on Globalisation and Labour Markets (GLM), Steve Thompson is Professor in Economics in
the Department of Economics, University of Leicester, and Peter Wright is Lecturer in the
School of Economics, University of Nottingham.
Acknowledgements
The authors would like to thank Claude Menard, editor of the Journal of Economic Behavior
and Organization, and two anonymous referees for comments and encouragement and
participants at the 1998 Royal Economic Society Conference, especially Andrew Dickerson and
Stephen Machin, for helpful suggestions on an earlier version of this paper. Emma Angell,
Helena Gorse and Fiona Singer provided technical assistance and the research was supported by
the ESRC (grant number R000221779) and the Leverhulme Trust (Programme Grant F114/BF),
to whom the authors are indebted.
Page 3
Do Hostile Mergers Destroy Jobs?
by
M. Conyon, S. Girma, P. Wright and S. Thompson
Abstract
This paper provides a systematic empirical analysis of the employment effects of hostile
takeovers in the United Kingdom for the period 1983-1996. It finds no evidence for
distinguishing between friendly and hostile acquisitions in terms of their impact on labour
demand. Indeed, each type of transaction appears to have an immediate negative impact on
labour demand, equivalent to about 7.5 percent of the pre-merger level. However, the paper
does find that the absolute number of employees falls substantially, along with output, in the
hostile merger case alone. This appears to be the consequence of a high level of post-merger
divestment that distinguishes hostile transactions.
Outline
1. Introduction
2. Hostile Mergers and Their Employment Consequences
3. Sample, Data and Modelling Strategy
4. Results
5. Conclusions
Page 4
Non-Technical Summary
The objective of the study was to construct a large panel database of UK firms in order to determine the
impact of completed acquisitions on the demand for labour of the firms concerned. We identify as many
acquisitions as possible within the set of quoted UK companies, over the period 1967-1996, using the
London Business School Share Price Database, the Cambridge DTI database of firm accounts, the Office
of National Statistics datasheets and 'Acquisitions Monthly'.
The paper finds that acquisition activity in the UK is associated with substantial and statistically significant
falls in the employment of merged firms. However it finds no evidence for distinguishing between friendly
and hostile acquisitions in terms of their impact on labour demand. Indeed, each type of transaction
appears to have an immediate negative impact on labour demand, equivalent to about 7.5 percent of the
pre-merger level. However, the paper does find that the absolute number of employees falls substantially,
along with output, in the hostile merger case alone. This appears to be the consequence of a high level of
post-merger divestment that distinguishes hostile transactions.
In contrast to the earlier US literature which suggests that control changes produce insignificant
employment effects, or at least produces effects which are limited to a small number of white collar
employees, our results are indicative of mergers being used as a major tool of industrial restructuring.
Page 5
1
1 Introduction
“In the light of the lacklustre performance of other disciplinary devices, hostile takeovers
are probably the most effective way for shareholders to get rid of non-value-maximizing
managers without bribing them.” Andre Shleifer and Robert Vishny (1988) p. 11.
“…[T]he greater the threat of the contested takeover, then the higher the premium
companies feel that they must earn to keep their shareholders happy. This is the
fundamental weakness of the British system.” Will Hutton (1995) p. 157.
The purpose of this paper is to provide a systematic empirical investigation of the widely
held - but strangely under-researched - proposition that hostile takeovers result in job losses.
The hostile takeover remains the most controversial instrument in the Anglo-American
system of corporate governance, as the above quotations illustrate. Its supporters see it as
the ultimate disciplinary sanction on managerial incompetence or malfeasance. Critics
suggest that it is imprecise, open to abuse and the source of substantial specific and systemic
costs. Following an influential contribution by Shleifer and Summers (1988), it has also
been considered as a device to enforce an ex post renegotiation of the firm’s explicit and
implicit labour contracts. This, according to the authors, involves a “breach of trust” with
employees and an expropriation of value - via job losses and/or revisions in the wage/effort
bargain - from labour and its redistribution to shareholders. Such an outcome has efficiency
implications if it undermines the ex ante commitment of employees to the firm and to their
investment in firm-specific human capital (Blair, 1995).
This debate over hostile takeovers gives rise to a policy dilemma: If the market for corporate
control works as its proponents believe then any restrictions on hostile bids will encourage
inefficiency. However, if the specific and systemic costs of hostile transactions exceed the
benefits, then limitations, including those designed to safeguard the interests of
“stakeholders” [Kay and Silberston (1995)], including employees, may be appropriate. Such
issues can be resolved only by evaluation of the empirical evidence. The need to assemble
such evidence provides the principal motivation for this paper.
The existing evidence on the employment consequences of hostile mergers - see below - is
fragmentary, inconclusive and largely confined to studies from the USA. This paper uses a
dynamic demand for labour model to explore the immediate impact of mergers on
Page 6
2
employment and also the adjustment process through time. Since our design incorporates
explicit controls for output changes in the post-merger period, it addresses the changing
demand for labour more precisely than would be possible through a consideration of
employment numbers alone. The model is estimated on an unbalanced panel of 433 firms,
with 240 identified stock market acquisitions over the period 1983-96.
The paper is organised as follows: Section II explores the hypothesised relationships
between hostile acquisitions and job losses. Section III outlines the sample, presents some
descriptive statistics and then discusses the econometric methodology it be employed. The
results are discussed in section IV and a brief conclusion follows.
2 Hostile Mergers and Their Employment Consequences
The ‘market for corporate control’ hypothesis contends that the interests of shareholder-
principals are best protected by unrestricted competition for the stewardship of corporate
assets using a take-over market [Manne (1965)]. Such a view rests on three propositions.
First, it assumes the existence of a liquid and efficient stock market that allows low cost
trades in equity claims at prices that reflect relevant available information
i
. Second, for the
threat of a take-over bid to be fully effective, it must be possible that a potential acquirer can
take control despite the opposition of the incumbent management - i.e. hostile bids must be
feasible. And third, actual and potential competition for equity voting rights should be
sufficient to ensure that any benefits resulting from a change in control will accrue wholly or
partly to incumbent shareholders
ii
.
Critics have questioned whether the hostile acquisition is cost-effective as a disciplinary
device. First, such critics point to the evidence on the characteristics of acquired and
acquiring firms - see Hughes (1993) for a comprehensive survey - to question the precision
of the take-over process in reallocating resources to demonstrably superior management
teams. Second, it is pointed out that equating a hostile approach with a disciplinary one is
not necessarily justifiable: hostile acquisitions may be used to further managerial aspirations
for growth, particularly by cash-rich managements [Mueller (1969), Jensen (1986)]. Finally,
there are non-trivial costs associated with pursuing and opposing hostile bids, these include
the explicit costs incurred in specific contests as well as the systemic costs, associated with
the absorption of managerial efforts etc.
iii
Page 7
3
Given the multiplicity of possible motivations for acquisition [see Caves (1989)], including
managerial aspirations for growth as noted above, it may be reasonable to ask why hostile
takeovers are so widely considered likely to result in injurious consequences for labour?
After all, disciplinary transactions in the sense of Manne (1965) or acquisitions intended to
restructure redundant capacity could be achieved –and probably are – through agreed
takeovers. Two principal reasons for expecting hostile takeovers to be distinctive have been
advanced: First, a hostile (or contested) merger is one that, by definition, is initially opposed
by the board of the target firm. To the extent to which this indicates that the two parties
could not agree upon a mutually acceptable price, it suggests that low bidders (among which
are presumably many cost-minimising disciplinary acquirers) will be disproportionately
represented. Second, since hostile acquisitions are typically followed by the displacement of
existing senior managers [Franks and Mayer (1996), Hirshleifer and Thakor (1994)] they
represent an opportunity to take decisions that these managers may have been reluctant to
approve. Managerial preferences for size, or for a quiet life without excessive change, or
management’s identification with the interests of employees may inhibit the sale,
reorganisation or closure of under-performing business units. A new management team, with
a different motivation, may have fewer inhibitions. Jensen (1993) attributes a major role to
the hostile takeover in this regard in stimulating US corporations’ exit from many traditional
manufacturing activities in the 1980s.
The change in managerial motivation is likely to be supplemented by an increase in
management credibility in the event of any confrontation with labour. Not merely is success
in the takeover contest itself some indication of a willingness to engage in confrontation, but
hostile takeover specialists, including Leveraged Buyout associations, may bring with them a
reputation for re-structuring, perhaps reinforced by debt commitments.
Shleifer and Summers (1988) reject this efficiency-enhancing interpretation of hostile
transactions. They view the control change as an opportunity to renege on implicit aspects of
the employment contract and an occasion for the new management to renegotiate
employment, effort and pay levels on terms less favourable to the interests of labour. They
suggest that such a change is not merely an ex post redistribution of value in favour of
shareholders. It is also - to the extent that it is seen as a “breach of trust” by employees with
long-term job expectations - a discouragement to other workers in making ex ante
Page 8
4
commitments to the firm. Thus it may discourage employees from acquiring job-specific
human capital or from accepting pay profiles with deferred compensation elements. If these
arrangements have output benefits, as has been widely argued, their discouragement will
have systemic costs and not merely re-distributive consequences.
These arguments have reinforced the voices of those seeking to restrict hostile acquisition
activity out of concern for job losses. In the USA a number of states have enacted anti-
takeover legislation for this purpose, while in the UK there have been widespread calls, from
the trade unions and from proponents of “stakeholder capitalism” [e.g. Kay and Silberston
(1995), Hutton (1995)], for changes in shareholder voting procedures to retard hostile
approaches. However, even if systemic costs are associated with hostile takeovers, it does
not follow that these should be outlawed. Shleifer and Summers (1988) themselves merely
suggest that the distributional consequences of such transactions be addressed by making the
acquired firm’s shareholders, who typically reap the anticipated benefits via a bid premium,
responsible for compensating adversely affected employees. Jensen (1993) goes further and
argues that the costs are simply an inevitable corollary of using implicit contracts. He
suggests that the latter’s use here is precisely because it is prohibitively costly to specify
contracts that allow for all demand and supply uncertainties.
These policy issues require an informed debate, but the extent of hard evidence on the
employment effects of hostile acquisitions is extremely limited. Most existing empirical
evidence on the employment implications of mergers [e.g. Brown and Medoff (1988),
McGuckin et al. (1995)] does not typically distinguish their consequences by merger type.
This is unfortunate if, as Morck et al (1988) point out, the differences between the
characteristics of friendly and hostile targets are sufficiently pronounced that: “ … research
results on friendly bids may have little to say about hostile bids, and vice versa.” Such
evidence as is available is strictly limited but suggestive of differing effects at different levels
in the organisation: board members are clearly subject to abnormally high levels of
displacement following successful hostile bids in the USA [Hirshleifer and Thakor (1994)]
and the UK. Franks and Mayer (1996), for example, report that 88 percent of UK inside
directors resigned following a hostile acquisition compared to 50 percent in “friendly”
mergers. [Of course, the absolute number of displaced directors will be quite small and
some of these will be replaced.] White-collar employees appear to be most vulnerable in
Page 9
5
hostile take-overs, particularly where headquarters can be consolidated. Bhagat et al (1990)
provide case study evidence on 30 hostile acquisitions in the US, out of which 14
experienced substantial white-collar job losses, averaging 660 each. However, Bhagat et
al’s use of press reports as the primary source of labour data makes it difficult to be specific
about what was happening elsewhere in their sample. Lichtenberg and Siegel (1990)
examining US leveraged buyouts, which may be considered as alternative disciplinary
transactions, also report supervisory job losses; Blue-collar employees appear to experience
fewer job losses in US deals. However, whilst Bhagat et al (1990) report higher
proportionate white collar job losses in hostile take-overs than in “white knight” outcomes,
the reverse is true for blue-collar jobs.
3 Sample, Data and Modelling Strategy
3.1 Sample and Data
The starting point for this research was to use the London Share Price Database to identify
all domestic acquisitions among the set of quoted UK companies between 1983 and 1996.
This allowed the identification of 721 mergers and acquisitions made by 536 firms.
Takeovers involving foreign or state-owned enterprises were omitted because of potential
data and comparability problems. Since the intention of the research was to study
employment effects in the context of a dynamic labour demand model, the entire potential
sample of acquired and acquiring firms was screened for data availability on employment,
wages and sales for at least three consecutive years surrounding the merger event.
Companies making two or more substantial acquisitions in any year were excluded, to avoid
conflating the effects of different events.
Takeovers were next classified into hostile or friendly, on the basis of the target firm’s
response to the initial bid from the subsequently successful bidder, using data from
Acquisitions Monthly and the Financial Times. Additional information was obtained from
Datastream and from the Office of National Statistics. Unavailability of data here and
relating to employment, wages etc. reduced our sample to 240 potentially useful cases, made
by some 195 acquirers. The frequency distribution of these mergers is given in Table 1.
A control group of firms was also selected from the population of UK quoted companies
over the period, using the following criteria. First, that the firm had made no major
acquisition during the sample period. Secondly, that data on wages, employment etc. were
Page 10
6
available for at least three consecutive years. Thirdly, that in no year did the growth rate of
total assets exceed 100%, an event which was considered likely to indicate an unrecorded
acquisition. And fourthly, that the 2-digit sectoral composition of the control matched that
of the sample itself. This resulted in a control group of 238 firms. These non-acquiring
control firms were, as expected, significantly smaller than the acquirers
iv
. For example, the
latter had an average number of employees (pre acquisition) of 11,041 compared to 973 for
the control. Similarly, their average levels of real output (in 1990 £k) were 632,449 and
103,235, respectively. The balance of the entire panel is given in Table 2.
A preliminary comparison of the hostile and friendly cases is shown in Tables 3 and 4. For
the purposes of displaying descriptive statistics, we have here concentrated on a restricted
sample of companies for which six consecutive years of data were available and which made
only a single acquisition over the period. Table 3 indicates within the set of friendly
acquisitions, the acquiring firm was significantly larger and paid a significantly higher mean
wage rate than the acquired firm. Hostile acquirers were also significantly larger than their
targets, although the wage rate difference was insignificant. Mean labour productivity,
measured here as real sales per employee, was higher for the acquired firms in the friendly
case and for the acquiring firm in the hostile case, but neither difference was statistically
significant.
In Table 4 the post-merger trajectories for the same variables are compared to their pre-
merger (t-1) values. It can be seen that there is a considerable difference between the
friendly and hostile cases. In the former, there is an insignificant fall immediately after the
merger, perhaps consistent with the elimination of duplicated facilities, followed by a
significant increase that is sustained until at least year t+4. In the hostile merger case the
immediate fall in employment is significant and equal to a substantial 15 percent of the pre-
merger level. There is some recovery in year t+2, but this is not sustained over the following
years. By contrast, productivity fluctuates substantially in a both the hostile and friendly
cases, but shows no significant change.
A partial explanation for the employment differences between the two cases is evident in
Table 4, where real output levels are compared. It is clear that while friendly acquirers’ mean
output increased gradually over a six year period, the hostile acquirers experienced a severe
17.9 percent fall in output between the years immediately preceding and following the
Page 11
7
acquisition. There is some recovery in each of the following three years, but even four years
after the event average real output only stands at 86.7 percent of its pre-merger level. This
pattern is consistent with the substantial post-merger divestiture of activities in the case of
hostile transactions, an interpretation that is considered below. However, for a more detailed
analysis it is necessary to examine employment changes in the context of a demand for
labour model.
3.2 The Modelling Strategy
If the merger of two firms results in a different optimal employment size to that previously
obtaining for them as separate entities, then a profit-maximizing management will need to
effect an adjustment in the labour force. However, movement to the newly desired level of
employment is unlikely to be instantaneous, and the process of adjustment will depend on
the balance of costs between changing employment levels and being away from the
optimum. Different specifications of a dynamic labour demand function may be derived
depending on the assumptions that are made concerning the form of adjustment costs, the
production function, the predetermination of production and the capital stock. We follow
Nickell (1984) in assuming that individual firms face quadratic costs functions, Cobb-
Douglas technology and are output constrained. This results in general adjustment equation
of the form:
(1)
it
i
it
it
it
it
it
it
f
Q
Q
w
w
l
l
e
d
d
ß
ß
a
+
+
+
+
+
+
=
-
-
-
1
1
0
1
1
0
1
Where l, w and Q denote the log of employment, log of real wages relative to the user cost
of capital and the log of real output respectively and the
i
f denote firm-level fixed effects
v
.
In the present context it is also necessary to allow for the possibility that mergers have
additional effects on labour demand, for example by changing the efficiency with which
labour is used. This is done by the introduction of dummy variables to capture the
contemporaneous and subsequent effects of hostile (H and PH, respectively) and friendly (F
and PF, respectively) acquisitions. Adding time dummies to control for possible cyclical
effects gives the estimating equation:
(2)
it
i
it
it
it
it
it
it
it
it
it
it
dummies
time
f
PF
PH
F
H
s
s
w
w
l
l
e
?
?
?
?
d
d
ß
ß
a
+
+
+
+
+
+
+
+
+
+
+
=
-
-
-
1
0
1
0
1
1
0
1
1
0
1
Page 12
8
Where i and t index firms and time periods respectively; l, w and s denote logarithms of
employment, real wages and real sales respectively
vi
. In recognition of the problems of
measuring inter-firm variations in the user cost of capital, we follow much of the demand for
labour literature – see Hamermesh (1993) – in implicitly normalising this to unity and thus
using the real wage terms to capture the relative factor price effects. Finally, e is an
equation disturbance term.
Following the seminal work of Anderson and Hsiao (1982), consistent estimates of the
parameters of dynamic panel models with fixed effects may be secured by applying
appropriate instrumental variables techniques to the first differenced equations
vii
. Since the
disturbances of the first differenced models are correlated within firms by construction, see
Nickell (1981), a generalised instrumental variables, or a generalised method of moments,
estimation is required.
4 Results
The first-differenced version of equation (2) was estimated using a generalised instrumental
variable approach, with an instrument set which included lagged values of employment,
wages, output and fixed assets. The results are given in Table 6. Here it can be seen that the
labour demand equation performs as expected. The wage and output terms have the
anticipated signs and are each statistically significant. The positive and significant coefficient
on the lagged dependent variable additionally indicates that the employment level exhibits
inertia and wages and output have persistent effects. In all cases the parameter estimates fall
within the expected ranges. The equation performs well statistically, with an absence of
second order serial correlation while the J-test indicates that instrumental validity cannot be
rejected.
Turning to the impact of mergers and acquisitions on labour demand, Table 5 indicates that
friendly and hostile transactions are each accompanied by a statistically significant fall of
approximately seven and a half percent in the acquirer’s derived demand for labour. A test
for equality of the two coefficients could not be rejected (p=0.96). Thus having controlled
for output changes each form of acquisition appears to stimulate a similar immediate fall in
employment. The subsequent effects are investigated using the post-merger dummies PF and
PH. These attract negative signs and coefficients of substantial magnitude (equivalent to 4.9
percent and 6.1 percent, respectively), but they do not achieve significance at the 10 percent
Page 13
9
level. Thus there is no statistical evidence to distinguish between the effects of hostile and
friendly transactions, although the relatively small number of hostile cases (39 out of 240)
and the high standard error on the PH dummy together suggest some caution is necessary.
In section II of this paper we reviewed a number of arguments, that each tended to suggest
that hostile takeovers would be particularly associated with job losses. This paper’s results
suggest such a view is substantially incorrect. It is the case that on average the level of
employment in the new entity created by the merger falls steeply in the immediate aftermath
of a hostile acquisition. Furthermore, this decline is sustained and is even reinforced over a
four-year post-merger interval. By contrast, the average level of employment in the friendly
merger case actually increases, at least after a probable initial decline. However, the decline
in employment in the hostile merger case appears to be a consequence of the immediate fall
in output that follows a hostile transaction in our sample. This output fall is not present in
the case of friendly mergers. Once a demand for labour formulation is adopted, in which
output changes are explicitly controlled for, the distinctive employment effects of the hostile
transaction disappear. Both friendly and hostile transactions appear to be followed by a
reduction in the derived demand for labour of about 7.5%. Thus we find no clear support for
the Shleifer and Summers (1988) argument that the completion of a hostile transaction
provides a unique occasion for the new management to renege on implicit labour contracts
resulting in a “breach of trust” of employees’ expectations. Control changes do appear to be
followed by substantial falls in labour demand, but no more so for hostile mergers than for
friendly ones.
How then should we reconcile our result that hostile and friendly acquisitions have an
approximately equal negative impact on labour demand with the finding reported here - and
indeed elsewhere [e.g. Bhagat et al (1991)] - that the level of employment falls particularly
sharply after hostile acquisitions? It is conjectured that the answer lies in the substantial
volume of voluntary divestment that is associated with hostile acquisitions. Haynes et al.
(2000), in a forthcoming paper, suggest that the UK economy was exhibiting very high
levels of voluntary divestment over the period of our investigation. They suggest that this
frequently represented a reversal of an earlier trend towards corporate diversification. Such
an outcome was perhaps a response to a decline in the comparative advantage of the
multidivisional form of organisation for the multi-product firm, and/or a recognition of the
Page 14
10
stock market’s apparent preference for firms with a narrower focus of activities [Hoskisson
and Turk (1990)]. Haynes et al. (2000) report that merely the threat of a hostile approach
was sufficient to generate a significant increase in the level of divestment activity among
their sample.
A full investigation of the relationship between acquisition activity and divestment would
require detailed information on the business units divested, which lies beyond the scope of
this paper. However, it was possible to use published accounting data to make an
exploratory investigation. Table 6 reports the results of a random effects model panel
regression of reported fixed asset sales (i.e. disposals of property, plant and equipment) on
contemporaneous and lagged hostile and friendly merger dummies, with controls for
industry growth and the pre-merger level of divestment. It can be seen that the post-hostile
coefficient is highly significant (t=69.4) and approximately three times the size of the
positive but insignificant post-friendly effect. This result is strongly supportive of the paper’s
conjecture that hostility is particularly associated with divestment.
Three further qualifications should be added to our results: First, assuming that the observed
output fall is generated by divestment, as we have conjectured, it is not necessarily the case
that hostile takeovers cause divestment. As Franks and Mayer (1996) have pointed out,
disagreements over proposed divestment plans might be an important reason why incumbent
boards reject takeover approaches in the first place. Second, our employment results relate
strictly to observed activity in the firm that emerges as an outcome of the merger. If
divestment activity does indeed follow hostile takeovers to a substantial extent, as
suggested, then a full analysis of employment consequences would require an investigation
of the second (plus any subsequent) round control change effects as divested business units
are themselves acquired. It is possible that employees of these units experience Shleifer-
Summers effects to a disproportionate extent. Third, as seen in Table 1, the proportion of
hostile cases, is quite small (under one sixth). It is conceivable, given the respective
magnitude of the Post Friendly and Post Hostile coefficients, in Table 6, and the high
standard error of the latter that a change in sample period would produce a significant
difference between the two.
Page 15
11
5 Conclusions
It has been widely contended that hostile takeovers have adverse employment consequences.
Firstly, hostility has been interpreted as signalling a disciplinary acquisition whose objective
is the substitution of a new set of managers to raise the return on corporate assets. Such
transactions are considered likely to be associated with increased labour productivity and job
losses. Secondly, following Shleifer and Summers (1988), it has been conjectured that a
hostile acquisition offers a unique opportunity for employers to renege on the explicit and
implicit terms of employment of workers in the acquired company. This would allow a
transfer of value from labour to capital, at least some of which will take the form of job
losses.
This paper has used a dynamic demand for labour model estimated across a panel of UK
firms over the period 1983-1996 to explore the impact of hostile mergers on employment.
Our empirical results suggest that hostile and friendly acquisitions are each similarly
associated with an apparent initial decrease in labour demand, averaging about 7.5 percent.
There is some weak evidence that each type of acquisition produces a further negative shock
in subsequent periods, although the coefficients here fall below generally acceptable levels of
significance. There is, however, no evidence that hostile acquirers’ labour demand falls more
strongly than that of the friendly acquirers. That is, using a demand for labour framework
that explicitly controls for contemporaneous and lagged output and wage effects, we are
unable to distinguish any separate hostile takeover effect. In the UK, at least, the hostile
takeover is not uniquely associated with the ‘dirty job’ of laying off redundant workers, as
Shleifer and Vishny (1991) and others have contended. It follows that calls to reform the
corporate governance system to outlaw or severely curtail hostile acquisitions, with a view
to protecting the interests of labour, may be seriously misplaced.
There is, however, one important respect in which this paper has shown that hostile and
friendly takeovers differ and which may have given rise to some of popular distrust of the
former. It has been seen that hostile transactions are associated with immediate substantial
falls in output and employment, which are not present after friendly transactions. It has been
conjectured here that these falls are a consequence of a high level of post-merger divestment
of divisions and subsidiaries. Such divestments constitute a second order of control
transactions. In the absence of further information it is not possible to say whether these
Page 16
12
transfers of business units have employment consequences that differ from the acquisition of
quoted companies themselves. If they do experience a second wave of post-transaction
employment changes, then the Shleifer-Summers argument is vindicated, albeit in an indirect
form.
Page 17
13
Table 1
Frequency of Sample Mergers by year
Year
Agreed
Hostile
1983
5
3
1984
10
3
1985
17
1
1986
23
8
1987
24
6
1988
24
0
1989
21
3
1990
14
1
1991
12
5
1992
13
3
1993
11
2
1994
13
1
1995
13
2
1996
1
1
Total
201
39
Table 2
Balance of the Panel
Number of time series
Acquirers
Controls
3
7
68
4
13
30
5
11
16
6
8
20
7
16
17
8
11
14
9
10
10
10
5
8
11
3
8
12
18
7
13
67
25
14
25
14
16
1
1
Total
195
238
Page 18
14
Table 3
Paired t-tests for Acquiring and Acquired
Firms a Year Prior to Merger
Merger type
And variable
Acquiring
Acquired
p-value
Friendly
Employment
4154
1529
.00
Wage rate
12.38
10.99
.02
Labour Productivity
60.97
66.87
.23
Hostile
Employment
9819
3943
.01
Wage rate
12.57
11.10
.20
Labour Productivity
74.01
61.47
.30
Notes:
1.
All values are simple means
2.
Productivity is defined as real sales per employee
Table 4
A Comparison of Pre- and Post-event Employment, Wage, Productivity and Real
Output Performance in Friendly and Hostile Mergers
Merger type
And variable
t-1
t+1
t+2
t+3
t+4
Friendly
Employment
5684
5342
6081*
6175*
6024
Wage rate
11.83
13.39
12.26
12.19*
12.60*
Labour Productivity
58.66
65.59
59.85
60.40
62.06
Real Output
334,423
350,381
363,948
372,970
373,849
Hostile
Employment
13762
11706*
12240*
12004*
11229*
Wage rate
12.31
12.11
12.34
13.60
14.20*
Labour Productivity
69.74
67.32
65.14
69.08
74.12
Real Output
959,762
788,048
797,314
829,236
832,293
Notes:
1.
The first column refers to the combined (acquired and acquiring) values of the respective variables one year prior to the mergers.
2.
(*) Denotes significant differences (at 10% level) from the pre-merger values.
Page 19
15
Table 5
Estimates of First-Differenced Labour Demand Model: Dependent variable l
it
Independent
Variables
Coefficient
(standard error)
l
it-1
.610 (.144)
w
it
-.680 (.269)
w
it -1
.387 (.121)
s
it
.829 (.096)
s
it -1
-.419 (.139)
F
it
-.074 (.024)
H
it
-.076 (.038)
PF
it
-.049 (.033)
PH
it
-.061 (.039)
No. of obs.
2679
R-squared
.607
Time effects
Yes
Serial correlation
No
(p-value = .72)
J-stat p-value
.551
Table 6
Exploratory Analysis of the Determinants of Divestment:
Random Effects GLS Estimates
Independent Variable
Coefficient (standard error)
Prior Divestment
-.911
**
(.074)
Industry Growth
27.11 (74.20)
Friendly
-13.83 (49.81)
Post-Friendly
54.67 (44.48)
Hostile
44.62 (97.62)
Post-Hostile
149.04
**
(2.15)
No. of obs.
960
R
2
.29
Hausman p-value
.96
Notes
1.
The dependent variable is real sales of fixed assets (i.e. proceeds from sale of property, plant and equipment, Datastream item 423). The
data are restricted to four years after merger.
2.
** Indicates significance at the 1% level.
3.
Industry Growth is the growth rate of the acquired industry (3 digit level). Replacing this variable with industry’s dummies produced a
similar result.
4.
Friendly and Hostile are dummy variables for the first year of post-merger activity.
5.
Post-Friendly and Post-Hostile are dummy variables for the second and subsequent years of post-merger activity.
Page 20
16
References
Anderson T. and Hsiao C. (1982) ‘Formulation and Estimation of Dynamic Models using
Panel Data’, Journal of Econometrics, 18, 67-82.
Auerbach A.J. (ed.) (1988) ‘Corporate Takeovers: Causes and Consequences’, University
of Chicago Press: Chicago and London.
Baltagi B.H. and Griffin J.M. (1997) ‘Pooled Estimators versus their Heterogeneous
Counterparts in the context of the Dynamic Demand for Gasoline', Journal of
Econometrics, 77, 303-327.
Bhagat S, Shleifer A. and Vishny R. (1990) ‘Hostile Takeovers in the 1980s: the Returns to
Corporate Specialisation’,
Brookings Papers on Economic Activity:
Microeconomics, 1-72.
Blair M (1995) ‘Ownership and Control: Rethinking Corporate Governance for the 21
st
Century’. Brookings: Washington DC.
Brown, C. and Medoff J.L. (1988) ‘The Impact of Firm Acquisition on Labor’, in
A.J.Auerbach op. cit.
Caves, R.E. (1989) ‘Mergers, Takeovers and Economic Efficiency: Foresight versus
Hindsight’, International Journal of Industrial Organization, 7, 151-76.
Franks J.R. and Mayer C (1996) ‘Hostile Takeovers and the Correction of Management
Failure’, Journal of Financial Economics, 40, 163-181.
Grossman S. and Hart O.D. (1980) ‘Takeover Bids, the Free Rider Problem and the Theory
of the Corporation’, Bell Journal of Economics, 11, 42-64.
Hamermesh D.M. (1993) Labor Demand, Princeton University Press.
Hart O.D. (1995) Firms, Contracts and Financial Structure, Clarendon Press: Oxford.
Haynes, M., Thompson S. and Wright M. (2000) ‘The Determinants of Corporate
Divestment in the UK’, International Journal of Industrial Organization
[forthcoming].
Hirshleifer D and Thakor AV. (1994) ‘ Managerial Performance, Boards of Directors and
Takeover Bidding’, Journal of Corporate Finance, 1, 63-90.
Hoskisson R. and Turk T. (1990) ‘Corporate Restructuring: Governance and Control Limits
of the Internal Capital Market’, Academy of Management Review, 15, 459-77.
Hughes A (1993) UK Mergers and Economic Performance: A Survey of the Empirical
Evidence, 1950-1990, in M Bishop and J Kay (eds.) European Mergers and Merger
Policy, Oxford University Press.
Hutton W. (1995) The State We’re In, Jonathan Cape: London.
Jensen M. (1986) ‘Agency Costs of Free Cash Flow, Corporate Finance and Takeovers,’
American Economic Review, 76, 323-329.
Jensen M. (1993) ‘The Modern Industrial Revolution, Exit and the Failure of Internal
Control Systems,’ Journal of Finance, 48, 831-880.
Kay, J. and Silberston, A. (1995) 'Corporate Governance’, National Institute Economic
Review, 153, 84-97
Lichtenberg F. and Siegel D. (1990) ‘The Effects of Leveraged Buyouts on Productivity and
Related Aspects of Firm Behaviour’, Journal of Financial Economics, 27, 165-194.
Page 21
17
Lichtenberg F. and Siegel D. (1992) ‘Takeovers and Corporate Overhead.’ In Corporate
Takeovers and Productivity, F. Lichtenberg, (ed.), Cambridge, Massachusetts: The
MIT Press, 45-67.
Manne H.G.(1965) ‘Mergers and the Market for Corporate Control’, Journal of Political
Economy, 73, 693-706.
McGuckin R.H., Nguyen S.V. and Reznek A.P. (1995) ‘The impact of ownership change on
employment, wages and labor productivity in US manufacturing 1977-1987.’ Center
for Economic Studies, US Bureau of the Census, 95-8, April.
Morck R., Shleifer A. and Summers L. (1988) ‘The Characteristics of Hostile Takeovers’,
in Auerbach op. cit.
Mueller D.C. (1969) ‘A Theory of Conglomerate Mergers.’ Quarterly Journal of
Economics, 83, 643-59.
Nickell S (1981) ‘Biases in Dynamic Models with Fixed Effects’, Econometrica 49, 1417-
26.
Nickell S. (1984) ‘An Investigation of the Determinants of Manufacturing Employment in
the United Kingdom.’ Review of Economic Studies, LI(4), # 167, 529-558.
Nickell S, Wadwhwani S, Wall M (1992) 'Productivity growth in UK companies 1975-
1986.' European Economic Review, 1992, Vol.36, No.5, pp.1055-1085
Pesaran H. and Smith R. (1995) ‘Estimating Long-Run Relationships from Dynamic
Heterogeneous Panels’, Journal of Econometrics, LXVIII, 79-113.
Shleifer A. and Summers L (1988) ‘Breach of trust in hostile takeovers’, in Auerbach op.cit.
Shleifer A. and Vishny R (1991) ‘The Takeover Wave of the 1980’s’, Journal of Applied
Corporate Finance, 49-56.
Thompson, S. and Wright M. (1995) 'Corporate Governance: The Role of Restructuring
Transactions', Economic Journal, 105, 690-703.
Yarrow G.K. (1985) ‘Shareholder Protection, Compulsory Acquisition and the Efficiency of
the Takeover Process’, Journal of Industrial Economics, 34, 3-16.
Page 22
18
i
In an illiquid or an inefficient market, incumbent shareholders’ interests will not necessarily benefit from a
take-over approach.
ii
Hart (1995) demonstrates that the (widely adopted) one-share-one-vote decision rule affords maximum
protection to the interests of incumbent shareholders in the event of a control contest. Grossman and Hart
(1980) showed that in the absence of compulsion, free riding by the target’s shareholders would require that
all potential gains be redistributed to incumbent shareholders. Here the incentive to mount a control contest
depends upon prior toehold stakes or upon institutional devices to coerce minorities - see Yarrow (1985).
iii
Blair (1995) and others point out that these could be substantial if a fear of hostile acquisition encourages
myopic neglect of long-term investments.
iv
There have been a number of comparisons of acquiring and other firms in the UK literature (see Hughes
(1993) pp48-52). These have shown that acquiring companies are on average substantially larger but not
neceThis is the html version of the file http://www.belkcollege.uncc.edu/economics/tamura2.pdf.
G o o g l e automatically generates html versions of documents as we crawl the web.

Google is neither affiliated with the authors of this page nor responsible for its content.

This is the html version of the file http://www.cherry.gatech.edu/refs/student/elder-2004-prof-paper.pdf.
G o o g l e automatically generates html versions of documents as we crawl the web.
To link to or bookmark this page, use the following url: http://www.google.com/search?q=cache:u5hXugA8MOwJ:www.cherry.gatech.edu/refs/student/elder-2004-prof-paper.pdf+deindustrialization+US+opposition&hl=en&gl=us&ct=clnk&cd=6

Google is neither affiliated with the authors of this page nor responsible for its content.
These search terms have been highlighted:  deindustrialization  us  opposition
Page 1
Effects of Job Displacement in the U.S.:
A Cross-Case Analysis
Professional paper submitted as part of the requirements
for the degree of Master of Science in Public Policy, Spring 2004.
Delia C. Elder
MSPP Candidate,
Georgia Institute of Technology
dcelder@alumni.princeton.edu
Page 2
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
2 of 21
Abstract
Since the 1970s, the U.S. has experienced job loss in established industrial sectors that
have traditionally provided jobs with high wages, stability, and good benefits. Due to a
number of factors, including deindustrialization, the decline of unions, and a stagnation in
the relative minimum wage, jobs being created today are characterized by lower wages,
underemployment, less permanence, and fewer benefits. However, due to a variety of
differences among studies, researchers have been unable to provide a coherent view of
the degree or impact of these changes on U.S. workers. This paper uses a cross-case
analysis of twenty-two relevant studies to identify shared values, goals, and conclusions.
The results of this analysis show that researchers have yet to develop a standardized
framework for investigation of this topic. Weakness exists in that approaches, definitions,
and differences among datasets render results somewhat incomparable. Much effort
needs to be made towards standardizing research efforts across time periods and datasets
in order to generate better policy recommendations.
Page 3
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
3 of 21
I. INTRO
A. Context of Increased Earnings Inequality
Since the 1970s, the manufacturing sector has experienced an overall decline in
number of people employed, compared to total US employment (Figures 1 and 2,
below). Coupled with the reduction, after 2000, of many service sector jobs (Figure
3), it seems that a perception exists that employment is more insecure and less
permanent than it used to be (Kuhn 2003). This insecurity has important
consequences for workers as jobs become at once scarce and less rewarding
(financially). As a result, a vast body of research has emerged on both the causes and
the effects of increasing earnings inequality.
Figure 1: Total Non-farm Employment, 1970-2004, Seasonally Adjusted
Ψ
Ψ
Data for figures taken from the National Bureau of Labor Statistics, summarized from the Current
Population Survey. <http://www.bls.gov>
Page 4
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
4 of 21
Figure 2: Manufacturing Employment, 1970-2004, Seasonally Adjusted
Figure 3: Service-Sector Employment, 1970-2004, Seasonally Adjusted
B. Definition of Policy Issue
A lack of consensus among researchers regarding both the causes and the effects
of increasing wage inequality, let alone its existence, has led to a variety of policy
changes with the intent of assisting displaced workers, either through job training,
“safety-net” assistance programs, or economic interference in certain job sectors. The
Manpower Development Training Act of 1962, the Trade Adjustment Assistance
(TAA) Act of 1962/74, and the Worker Adjustment and Retraining Notification
(WARN) Act of 1989 are all examples of policymakers’ desire to assist the
unemployed and to reduce the impact of adverse effects caused by economic changes.
However, these efforts have had mixed results. It is clear that in order to
construct effective policies to deal with new economic and workforce challenges,
research must provide coherent, if not precise estimates of the impacts of worker
Page 5
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
5 of 21
displacement and earnings inequality. A framework for evaluating research goals and
outcomes remains to be formulated.
C. Statement of Intent
The first step in piecing together a cohesive framework is to ascertain why
previous research has achieved such diverse and fragmentary results. This paper will
use a cross-case analysis to identify shared goals, measurements, and results among a
selected body of research concerning wage inequality and displacement effects.
II. PROBLEM STATEMENT
A. Background of the Problem
Levy and Murnane (1992) reviewed a series of studies in the 1980s that showed
that, beginning in the early 1970s, the US, along with a number of OECD countries,
experienced a dramatic increase in earnings and wage inequality among workers, a
rise which is contributing to a growing differential in total family income. McCall
(2000) also found increasingly high levels of within-group income inequality.
Although the literature is relatively consistent on the issue of rising wage
inequality both within and between groups, there is a lack of agreement on the causes
underlying the increase, as well as the degree of wage inequality and its impacts.
Suspected causes fall into three categories:
• decline in institutional constraints on inequality (such as the fall of the minimum
wage and a decline in unionization),
• changes in the economy (such as increased foreign trade and changes in industrial
structure) and
• changes in technology (skill-biased technological change.)
The next section will detail a number of studies that focused on these three areas
of causation and the effects they found, in order to provide a sense of the disparity of
results. It should also be noted that any of these “causes” could be difficult to
separate from the others, due to each exerting an influence on the others.
Institutional Constraints
The real minimum wage fell 44% in the 1980s. As Gottschalk and Smeeding
point out, this is consistent with a decline at the bottom of the wage distribution, but
does not explain the increase at the top of the distribution or within educational
groups. However, several studies have shown that the impact of the falling minimum
wage is certainly a factor, perhaps responsible for as much as 30 percent of the
change in wage distribution.
Wolff (2002) highlights the post-war decline in rates of unionization as another
institutional factor in the inequality trend. As unions experience a decline in
membership due to layoffs, the union is less able to exert bargaining powers which
previously kept wages at similar levels across jobs and industries. However, neither
the decline of unionization nor the fall in the minimum wage can explain the rise in
skill intensity among industries, because these changes would have made less-skilled
workers cheaper, leading firms to decrease their emphasis on skill-intensive
production.
Page 6
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
6 of 21
Economic Changes
Although it is clear that a decline of institutional constraints on wages exerts some
effects on the wage distribution, a variety of demand-side shifts exert pressure as
well.
Deindustrialization, or the shift in U.S. production from manufacturing to service
sector output, has been cited as a major cause for the increased demand for college-
educated workers. However, several studies have shown that deindustrialization is
only part of a larger demand-side trend. Murphy and Welch (1993), for example,
estimated that changes in U.S. industrial structure could account for only 16% of the
increased demand for skilled workers.
Increased globalization of the U.S. economy, beginning in the 1960s, is another
major economic shift that has displaced low-skilled U.S. labor. Borjas and Ramey
(1995) define this shift as a chain of causation: foreign competition in certain
industries with low skill requirements causes a shift towards high-skilled industry,
causing unskilled workers to enter the non-tradeable sector, which causes a decline in
their relative wages. However, several authors have pointed out that relative demand
shifts towards skilled workers took place in all industries, not just the tradeable goods
sector (Johnson 1997). Additionally, this demand shift has been occurring over at
least the past forty years, not just since the 1980s, although there has been a slight
acceleration.
Skill-biased Technological Change
“Skill-biased technological change” refers to the theory that as firms move
towards more skill- or technology-intensive production techniques, their demand for
highly skilled, educated, and adaptive workers increases. Skilled workers become
more productive in jobs they already perform, or more efficient in jobs formerly done
by unskilled workers. This theory forces the less-skilled population into service sector
jobs and less technology-intense industries, typically jobs with fewer benefits and
lower pay.
Additionally, “skill-biased technological change” may have the effect of pushing
less-skilled workers into industries other than manufacturing such as the service
sector, which inherently exhibit lower rates of unionization. This would result in a
cyclical process of de-unionization dependent on the severity of skills mismatch
among industries (Levy and Murnane 1994).
The effects of technological change are difficult to measure; for a while, skill-
biased change represented a factor inferred from the lack of explanation by other
factors. However, a number of studies emerged in the 1980s and 1990s that used case
studies or regressions on R&D investment, patent petitions, and employment data in
an attempt to analyze the observable effects of technological change. The U.S.
Bureau of Labor Statistics (1994) conducted several case studies of the effects of
changing production processes by industry, while Berman, Bound and Griliches
(1993) examined the ratios of expenditures on research and development to sales and
changes in skill demand. Bartel and Sicherman (1999) used panel data to assess the
role of the education premium associated with technological change. Lloyd-Ellis
(1999) and Casselli (1999) developed general equilibrium models of the effects of
Page 7
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
7 of 21
technological change. The results of these studies consistently indicated that
technological change is a determinant of increased skill demand.
B. Consequences and Effects of Change
Disagreement on the causes (or existence) of earnings inequality can lead to
imprecision in policy decisions, but it is equally if not more important to identify the
consequences of changes leading to inequality. The long-term impacts of change
cannot be neglected, especially the effects of displacement on earnings. Given that
the U.S. services sector now comprises almost 70% of U.S. jobs, many studies have
addressed the trend towards growth in service sector jobs, which generally afford less
benefits, have lower rates of unionization, provide fewer hours, and pay less than
other jobs (Bartel and Sicherman 1999). Nelson and Lorence (1988), for example,
found that earnings inequality in metropolitan areas was increased significantly by
service sector growth. Lorence (1991) further found that earnings inequality differed
among types of services provided; entry into the trade and personal services had the
largest impact on wages, whereas professional and business service sectors exhibited
fewer effects.
Some researchers have found that a consequence of the shift towards the service
sector is actually reduced earnings inequality, but only because men’s wages have
gone down, bringing them closer to those of women workers (Chevan and Stokes
2000).
Until there is a more unified measure of the impact of rising inequality,
policymakers will continue to speculate over the most appropriate course of action.
This paper seeks a consensus among selected research in order to identify shared
measurement goals and results.
C. Mixed Results At Home and Abroad
Wage Changes: Cross-National Comparisons
A variety of studies have attempted to examine wage changes in other nations.
Due to the prevalence of multinational firms and the quick dissemination of changes
in production processes, one would expect to see trends in rising inequality in other
countries as well as the United States. Although country studies can vary due to data
incompatibilities, several studies have shown strong evidence that the relative
demand for skilled labor is rising at a rapid rate (Machin, Ryan, and Van Reenin
1996; Gottschalk and Smeeding 1997).
One problem with conducting cross-country analysis is the lack, to date, of
comparable data sets and time periods, a factor which undoubtedly is responsible for
lack of consensus among US studies as well.
Page 8
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
8 of 21
Disparities
Disparities such as those reported above have often been explained by
irreconcilable differences in data sets, methods, and time periods among studies.
However, several studies indicate that the unit of analysis used in the study affects the
strength of the impact of deindustrialization on wage inequality (Morris and Western
1999; Lorence 1991). Different units of analysis, i.e. metro, state, regional, or
national geographic areas determine the size of income inequality effects.
While some data sources are more suited to answering certain kinds of research
questions, variation among data sources and the vagaries of data collection can also
lead to inconsistent results.
D. Income Inequality and Displacement: Current Policy Environment
Most programs, beginning in the 1960s, attempted to help soften the initial blow
of lost income following displacement (Kletzer 1998). The Manpower Development
Training Act of 1962 was one of the first programs to anticipate worker displacement
in the face of increasing automation. The same year saw the enactment of the Trade
Adjustment Assistance Act, which sought to provide income-replacement support to
workers in industries that might be adversely affected by the relaxation of certain
trade restrictions. The Act was amended in 1974 to relax eligibility requirements,
and again in 1994 in response to the North American Free Trade Agreement
(NAFTA).
These programs, along with unemployment insurance, are intended to function as
a “safety net” for displaced workers. They provide job training, re-training, income,
and job search assistance for eligible candidates. Job training programs have been
shown to be effective in getting workers back to work, but the effect of training on
long-term earnings has not been similarly established (Chevan and Stokes 2000).
In 1989, Congress passed the Worker Adjustment and Retraining Notification Act
(WARN), requiring employers to provide at least 60 days notice in the event of a
plant closing or layoffs. Before WARN, only three states required such notification.
The Act was intended to allow workers to begin searching for new employment
sooner (and thus avoid or reduce the length of joblessness), but some studies have
questioned the effectiveness of the Act, partly due to lax notification requirements. In
addition, it is unclear what kind of notice (verbal, written) or how soon such notice
should be given in order to produce maximum effectiveness (Kletzer 1998).
III. METHODOLOGY
A. Concepts and Variables
Earnings:
Hourly, weekly, or annual wages for a given individual.
Family Income: Used instead of individual earnings to estimate the full impact of
wage changes; presumably a more accurate “quality-of-life” indicator. Includes all
sources of income, including earnings/wages.
Wages:
Usually measured in hourly, quarterly, or yearly increments, and
does not include sources other than earnings paid by an employer.
Tenure:
Cumulative time spent with a prior employer; used to measure
transferability of skills among jobs, firms, or industries
Page 9
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
9 of 21
Upskilling:
Character of job changes such that a job requires more skill than
previously; high-tech sector jobs, and some manufacturing jobs, are presumably
experiencing “Upskilling.”
Deskilling:
Character of job changes such that it requires less skill than
previously; mass assembly resulted in “deskilling” in the early twentieth century.
Multiskilling:
The ability of workers to maintain or build multiple skills or skill
types.
Displaced Workers: Workers who experienced involuntary separation from a
previous job due to mass layoff or plant closing. A problem with this definition is that
underemployment or wage stagnation may induce a worker to quit voluntarily,
without them being captured as “displaced.”
Data Sources
The Displaced Workers Survey (DWS), a part of the Current Population Survey
provided by the Bureau of Labor Statistics, has become the “industry standard” for
job displacement research (Kletzer 1998). It draws a large, random sample of 60,000
US households, weighted to be representative of the US work force. The DWS lacks
data regarding unionization rates, however, and prior to 1994 did not identify workers
displaced from unionized jobs.
The Panel Survey of Income Dynamics (PSID) and National Longitudinal Survey
(NLS) provide information on individuals who were at risk of displacement but
experienced no job loss. A disadvantage of both is that they contain relatively small
samples of displaced workers. The PSID, unlike the CPS or NLSY, contains
observations on people of all ages, and covers a longer time period
The National Longitudinal Survey of Youth (NLSY) is a representative sample of
men and women who were between the ages of 14 to 22 when first interviewed in
1979. It is administered annually, and provides information on individuals both
before and after potential job loss. The NLSY also allows for distinction between
layoffs and firings, and (unlike the PSID) is not based on household heads, so does
not leave out a large proportion of female earners (Kletzer 2003).
The Longitudinal Employer Household Dynamics (LEHD) database is compiled
by the Census Bureau, utilizing state-level Unemployment Insurance records and ES-
202 data. It thus captures employer and employee data on almost all employees in a
state over long periods of time (Andersson et al. 2003).
B. Methodology
The methodology used in this paper is a cross-case analysis. Cross-case analysis
shares some characteristics with Research Value Mapping (RVM), an attempt to lend
structure to case study or anecdotal data collected in the course of a project.
Originally developed to assess (1) the value of university-industry partnerships, (2)
the development of science and technology human capital, and (3) the creation of
knowledge value communities, RVM methodology allows a more quantitative
evaluation of case study or anecdotal results, as well as a “lessons-learned” or “best
practices” summarization across study results (Bozeman and Kingsley 1996).
Additionally, the type and degree of study outcomes can be compared with a view
towards individual case weakness or bias.
Page 10
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
10 of 21
Traditionally used with case studies, RVM entails full case coding and statistical
manipulation. Cross-case analysis, on the other hand, presents a useful way to
compare outcomes without such extensive coding. Youtie (2003) examined a number
of case studies to identify shared goals and results among participants in the
Manufacturing Extension Partnership (MEP). This cross-case (or meta-case) analysis
attempted to standardize procedures for case study reporting. Although less
quantitative than the RVM approach, the cross-case approach works by
operationalizing the original case inputs and outputs and using these as variables in a
cross-case analysis.
Although this paper attempts to derive values similar to those found using RVM,
this study will attempt to apply a cross-case analysis to academic and think tank
research in the field of labor economics. This will allow for a “big-picture” approach
to the issues being examined by individual research efforts. Research can then be
examined for institutional bias, weaknesses, and differences in underlying
assumptions, region studied, datasets, and variables.
C. Research Strategy
In order to formulate an evaluatory framework, the inputs and outputs of each
study have been categorized, and certain questions asked of each to ensure
standardization. Questions regarding research focus include:
• Major concern or research focus
• Examines causes of wage inequality?
• Examines degree of wage inequality? (high loss, moderate loss, weak or
no loss)
• Examines differences between industry switchers and stayers?
Additionally, each study was coded according to:
• Scope (international, national, state, multi-metro, or metro)
• Funding body and/or Institutional Affiliation
• Dataset (i.e., CPS, DWS, PSID, NLSY)
• Dependent variables (family income, individual earnings, weekly, annual)
• Independent variables (gender, race, age, education, length of
displacement)
• Additional variables (unionized vs. non-unionized workers or industries,
white vs. blue-collar job placement, and part time / full-time distinction)
Page 11
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
11 of 21
IV. ANALYSIS
Table 1 – DATA SOURCE
DATA SOURCE
# OF CASES USING DATA
SOURCE
ASM (Annual Survey of Manuf.)
1
CBP (County Business Patterns)
1
Census
2
CPS (Current Population Survey)
2
DWS (Displaced Workers Survey)
7
LEHD (Longitudinal Employer Household Dynamics)
1
NLSY (National Longitudinal Study on Youth)
2
PSID (Panel Study on Income Dynamics)
3
QES (Quality of Employment Survey)
1
State Records (i.e. Unemp. Insurance)
3
International Sources (i.e. LIS, FES, RPP)
3
The majority of studies in this sample utilized the DWS as a data source.
Although some combined the DWS with other sources or methods of analysis, it should
be noted that the DWS has a number of potential problems. First, it is administered
infrequently, every three to five years. This poses problems not only in length of time,
but also because respondents may remember job separations that were more traumatic
more than other displacements (Rodriguez-Planas 2003). Additionally, the wording of
the DWS questions changed substantially between survey periods, making comparison
difficult over time (Farber 2003, Carrington and Zaman 1994). Finally, the DWS suffers
from sample bias, in that it refers to displaced workers by definition. The NLSY, in
comparison, provides a control group of workers who did not experience a job loss, as
well as a pre-test and post-test group of workers who did experience loss, both before and
after job separation (Kletzer 2003).
Individual studies mentioned a few more problems inherent with using datasets
such as the DWS, PSID, or the NLSY. For example, workers who leave the workforce to
work in an industry outside of dataset scope (non-salaried) are not measured, as are
neither self-employed workers nor those who retire.
Table 2 – SCOPE
REGION
# OF CASES USING
REGION
Metro
1
State
2
Country (US)
14
Multi-metro
1
Multi-state
1
Multi-country
3
Page 12
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
12 of 21
It appears that comparisons of earnings inequality focused on particular regions;
many authors cite the availability or completeness of datasets as one criterion for
selecting a particular scope. The US, for example, provides a number of accessible
datasets for analysis through the Census. However, as some researchers pointed out,
there have not been large efforts to make multi-country datasets comparable.
One problem that arises in comparing data across countries is that there are large
differences in variation among regions, even within countries. Second, there is a
pervasive problem inherent in definitions between countries (Gottschalk and Smeeding
1997); many countries define “income” as earnings, whereas others define income as a
combination of not only earnings, but also items such as unemployment insurance.
Table 3 – FUNDING SOURCE
FUNDING
# OF CASES USING SOURCE
Academic
10
Foundation
5
Government
5
Think Tank
2
The majority of studies appeared to be of an academic nature, although some were
published or reviewed by institutes such as the Upjohn Institute for Employment
Research. Although both of the studies labeled as “think tank” came from the relatively
liberal Brookings Institute, it was difficult to find studies published by a contrasting
conservative institute, such as the Cato Institute.
1
Table 4 – INDEPENDENT VARIABLES
Author
Title
Date of
Study
Unions?
Gender?
Race?
Age
Education?
Farber
"Job Loss in the US, 1981-
2001"
May 2003 NO Y
Y
Y
Y
Leonard and
Jacobsen
"Wage Trends and the Job
Creation Debate"
May 1990 Y
*N/A
Chevan and
Stokes
"Growth in Family Income
Inequality, 1970-1990:
Industrial Restructuring and
Demographic Change"
Aug. 2000 NO Y
Y
NO Y
1
Robert Z. Lawrence published earlier studies, sponsored by the Cato Institute, on plant
shutdowns and displacement in the early 1980s, but those were not as timely as the
majority of studies used in this research.
Page 13
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
13 of 21
Rodriguez-Planas "New Evidence on Layoffs,
Recalls, and Unemployment"
February
2003
Y
Y
NO Y
Y
Kuhn
"Use It or Lose It: Human
Capital Investment Strategies
and the Earnings of Displaced
Workers"
January
2003
Y
NO NO NO Y
Peter
"Skill-Biased Transition: the
Role of Markets, Institutions,
and Technological Change"
Sept. 2003
NO
NO Y
Kletzer 1998
"Job Displacement"
1998
NO Y
Y
NO Y
Kletzer and
Fairlie 2003
"The Long Term Costs of Job
Displacement for Young Adult
Workers"
2003
NO Y
NO Y
Y
Carrington and
Zaman
"Inter-industry Variation in the
Costs of Job Displacement"
Apr. 1994 Y
NO NO Y
Y
Jacobsen,
Lalonde, Sullivan
"The Costs of Worker
Displacement"
Feb. 1992 NO Y
NO Y
NO
Devereux
"Do Employers Provide
Insurance Against Low
Frequency Shocks?"
May 2002 Y
Y
Y
NO Y
LeFranc
"Labor Market Dynamics and
Wage Losses of Displace
Workers in France and the
US"
Aug. 2003 NO NO NO NO Y
Neal
"Industry-specific human
capital"
1995
Y
Y
Y
NO Y
Kletzer 1996
"The Role of Sector-Specific
Skills in Post-displacement
Earnings"
Oct. 1996
NO NO NO Y
Gottschalk and
Smeeding
"Cross-National Comparisons
of Earnings and Income
Inequality"
June 1997 Y
Y
NO Y
Y
Andersson,
Holzer, & Lane
"Worker Advancement in the
Low-Wage Labor Market: The
Importance of Good Jobs"
Oct. 2003 NO Y
Y
NO Y
Bartel and
Sicherman
"Technological Change and
Wages: an Interindustry
Analysis"
1999
NO Y
Y
Y
Y
Page 14
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
14 of 21
Hamermesh
"The Costs of Worker
Displacement"
Feb. 1987 NO NO NO NO NO
*N/A : blank spaces indicate non-applicable
Even among studies that purported to look at earnings inequality, researchers by
and large used a very selective range of variables. For example, the basic variables of
gender, race, age, and education were not examined in 6, 9, 10, and 2 of the studies
respectively. These factors are extremely correlated with degrees of increasing wage
inequality, as most of the studies indicate up front. Gender, for example, has been
suggested as one of the primary forces behind wage changes over the past thirty
years, as more women entered the workforce. Race becomes important when
examining earnings distributions, specifically when looking at impacts among the
low-wage labor market. Similarly, education is linked to skill premiums and a
worker’s ability to find new employment.
Another significant variable, at least when examining wage changes, is the
importance of unions within a given industry or firm. Almost half (7) of the eighteen
studies neglected to even estimate industry unionization rates, or to clarify whether or
not workers were unionized. Carrington and Zaman (1994), for example, did not look
at unionization rates, even while making comparisons among industries.
Page 15
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
15 of 21
Table 4 – RESULTS: DEGREE OF INEQUALITY
Sixteen of the twenty-two studies explicitly examined the existence of wage
inequality. Individual studies were coded for results according to the degree and
direction of wage inequality found:
High Loss = 5
Moderate Loss = 4
Weak Loss = 3
No Loss = 2
Moderate Gain = 1
High Gain = 0
High and moderately high loss was indicated by measurable loss either within one
industry or across industries or sectors. Weak loss was indicated by a lack of
conclusive results, or indistinct causes. No loss indicated that no direction, i.e. either
increasing or decreasing wage inequality could be found.
DEGREE
# OF CASES FINDING DEGREE
High Loss
5
Moderate Loss
5
Weak Loss (or dependent on other
factors)
4
No Loss
1
Moderate Gain
1
High Gain
0
The vast majority of the cases found a high to moderately high degree of increasing
wage inequality. Notably, findings of increasing inequality came not only from
institutions, but also from purely academic researchers, as well as those sponsored by
the government. It is clear that further research, perhaps concentrating on specific
industries or regions, is necessary in order to shed light on this phenomenon.
Table 4 – POLICY ALTERNATIVES
Of the studies that found increasing wage inequality, some attempted to suggest
policy changes or recommendations to better protect workers from growing
inequality. A variety of suggestions emerged:
ALTERNATIVE
# OF CASES SUGGESTING
Warn Against Disincentivizing
2
Education or Job Training
3
Job Search Assistance
2
Changes to Existing Policies like WARN
1
Changes to Redistributive Policies
2
Insurance Against Industry Shocks /
Importance of Multiskilling
4
Page 16
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
16 of 21
A couple of studies warned against disincentivizing normal market corrections to
inequality, i.e. warned against redistributive policies or government intervention in
economics. Although this recommendation ignores the short-term impact of
inequality, the authors rationalized that overcorrection by the government can prove
disastrous. However, Gottschalk and Smeeding (1997) pointed out that across
countries, those that have been most successful in mitigating wage inequality are
those with the strongest institutional safety nets. Their study was one of two that
advocated better redistributive and tax policies. Another study advocated changes to
existing labor policies like WARN in an effort to make them more effective and
timely, arguing that such simple changes as wording or time periods may be enough
to assist workers.
Many studies underlined the importance of rising returns to education over the
past few decades. However, Chevan and Stokes (2000) noted that not everyone will
experience increasing returns to education. Similarly, some studies noted that job
training is desirable, but this generally leads to firm-specific skills, which a number of
authors warned against, rationalizing that workers should aim to multiskill in an effort
to remain “employable.” Firm-specific skills tend to depreciate, or disappear, if a
worker loses a job or an industry declines overall (Kletzer 2003).
Finally, one author pointed out that policy change may be irrelevant, as the
market may correct for firm-specific skills; decreased job security in the US may
make US workers more competitive in the long run as they shift from firm-specific
skills to multiskilling (Kuhn 2003).
V. CONCLUSIONS AND RECOMMENDATIONS
A. Research Recommendations
It appears that researchers are keen to find out the degree or impact of inequality
in the US, since most of the studies used US data sources instead of state- or metro-
level sources. However, it has been noted that a number of variations exist between
the regional and the national level, and therefore, it may be more useful to
policymakers if researchers concentrate on micro-level changes as well as macro-
level changes.
In addition, in light of the disparate results of comparisons between countries,
researchers must seriously consider differences in definition between countries for
purposes of data comparison. With a view towards this, Carrington and Zaman
(1994) suggested incorporating empirical Bayes techniques for econometrics
purposes.
Furthermore, one important area that must be standardized is the issue of what
variables are examined, particularly race, gender, education, seniority, and
unionization. If not examined explicitly, then reasons need to be given for why these
variables were not examined. Additionally, few studies, with the exception of Farber
(1997), examined the impact of part-time vs. full-time re-employment, or,
alternatively, hours worked. This variable may shed the most light on whether
displaced workers are being adequately re-employed in new jobs.
Finally, research over the past decade has addressed, for the most part, the decline
of manufacturing in the US. However, there is new evidence that service sector jobs,
Page 17
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
17 of 21
which many heralded as the antidote to manufacturing jobs, are now in jeopardy
(Bartel and Sicherman 1999). Future research must address this issue, particularly in
light of the decline of service sector jobs over just the past couple of years.
B. Policy Recommendations (Case Consensus)
Job training and education seem to be the most feasible policy solutions suggested
in light of the difficulty, both political and physical, of changing the impact of market
forces like deindustrialization and technological change. However, as Chevan and
Stokes (2000) note, such training may benefit individuals, but overall will not benefit
anyone if there is no demand for higher-skilled jobs.
Even if such training occurs, it will almost certainly be of little benefit to the
average low-wage worker, if the resulting jobs do not pay good wages, or offer
benefits. Similarly, job training will not help if the jobs lack stability, or if the hours
offered leave workers underemployed (Kletzer 1998).
Among the studies focused on low- to middle-income earners, there was some
support for policies that would increase job-search assistance. Andersson et al (2003)
found that any kind of search assistance, even temp agencies, can help workers find
higher-paying jobs, especially traditionally low-wage earners.
A redistributive policy option that has received some interest is the minimum
wage increase. Unfortunately, this option faces intense opposition from employers
and conservatives who believe that artificially increasing the minimum wage will
drive businesses out of the US. Similar opposition exists to changes to the tax code
that might benefit low-wage workers.
However, another redistributive policy, that of health care coverage aimed at low-
wage workers (or the unemployed) is constantly on the policy radar, and these
changes may prove less politically polarized than wages and taxes. In the same vein,
childcare and assistance for the elderly may soften the impacts of rising inequality.
C. Concluding Remarks
Examining studies from a comparative perspective can provide useful insight into
their shared research goals. However, as this paper shows, researchers who are
looking at wage inequality in the US are still somewhat piecemeal in their approach.
Standardization of approach could include changes as simple as agreement upon
certain definitions or datasets, as well as time periods or industries. Perhaps by
starting with the policy options generally addressed “in conclusion,” researchers can
then work backwards to identify methods and variables that will be useful in
examining the policy option. Whether the methods include measurement of existing
phenomena or identification of the effects of such events, only a cooperative
approach will make the insights gained through research valuable to policymakers.
Page 18
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
18 of 21
VI. REFERENCES
Abraham, K. and S. Houseman, Nov. 1993. “Earnings Inequality in Germany.” W.E.
Upjohn Institute for Employment Research Staff Working Paper No. 94-24.
Andersson, F., H. Holzer, and J. Lane. 2003. “Worker Advancement in the Low-Wage
Labor Market: The Importance of Good Jobs.” Washington, D.C.: The Brookings
Institution.
Bartel, A. and N. Sicherman, 1999. “Technological Change and Wages: an Inter-Industry
Analysis.” Journal of Political Economy 107 (2).
Bartik, T. September 2003. “Thoughts on American Manufacturing Decline and
Revitalization.” Upjohn Institute Staff Working Paper No. 03-96.
Berman, E., J. Bound and Z. Griliches. May 1994. “Changes in the Demand for Skilled
Labor within U.S. Manufacturing Industries: Evidence from the Annual Survey of
Manufacturing.” Quarterly Journal of Economics 109, 367-398.
Bluestone, B. and B. Harrison. May 1988. “The Growth of Low-Wage Employment:
1963-1986.” American Economic Review Proceedings 78, 124-128.
Borjas, G. and V. Ramey. Nov. 1995. “Foreign Competition, Market Power, and Wage
Inequality.” Quarterly Journal of Economics 110 (4), 1075-1110.
Carrington, W. and A. Zaman. 1994. “Interindustry Variation in the Costs of Job
Displacement.” Journal of Labor Economics 12 (2), 243-275.
Casselli, F. Mar. 1999. “Technological Revolutions.” American Economic Review, 89
(1), 78-102.
Chevan, A. and R. Stokes. August 2000. “Growth in Family Income Inequality, 1970-
1990: Industrial Restructuring and Demographic Change.” Demography 37 (3),
365-380.
Crandall, R. Mar. 2003. “The Decline in Manufacturing Jobs in the Syracuse
Metropolitan
Area.”
The
Brookings
Institution
online.
<http://www.brookings.edu>
Danziger and Gottschalk. 1993. Uneven Tides: Rising inequality in America. NY:
Russell Sage Foundation.
Devereux, P. 2002. “Do Employers Provide Insurance Against Low Frequency Shocks?:
Industry Employment and Industry Wages.” Paper presented at the August 2003
Page 19
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
19 of 21
conference entitled “Microeconomic Analyses of Labor Reallocation,” sponsored
by the W.E. Upjohn Institute for Employment Research.
Farber, H. 2001. “Job Loss in the United States, 1981-2001.” Industrial Relations
Section, Princeton University, January.
Goldin, C. and L. Katz. May 1996. “Technology, Skill, and the Wage Structure: Insights
from the Past.” American Economic Review, 86 (2), 252-257.
Gottschalk, P. and T. Smeeding. June 1997. “Cross-National Comparisons of Earnings
and Income Inequality.” Journal of Economic Literature 35, 633-387.
Hamermesh, D. 1987. “The Costs of Worker Displacement.” Quarterly Journal of
Economics 102 (1), 51-75.
Jacobson, L., R. LaLonde, and D. Sullivan. 1993. “The Costs of Worker Displacement.”
Kalamazoo, MI: W.E. Upjohn Institute for Employment Research.
Johnson, G. 1997. “Changes in Earnings Inequality.” Journal of Economic Perspectives
11 (2), 41-54.
Kingsley, G. and B. Bozeman. 1996. “Coker “Technology Transfer and Absorption: An
R&D Value Mapping Approach,” Research Policy 25: 977:995.
Kletzer, L, and R. Fairlie. July 2003. “The Long-Term Costs of Job Displacement for
Young Adult Workers.” Industrial and Labor Relations Review 56 (4).
Kletzer, L. 1998. “Job Displacement.” Journal of Economic Perspectives 12 (1), 115-
136.
Kletzer, L. 1996. “The Role of Sector-Specific Skills in Postdisplacement Earnings.”
Industrial Relations 35 (4).
Kuhn, P. and A. Sweetman. 2003. “Use It or Lose It? Human Capital Investment
Strategies and the Earnings of Displaced Workers.” Preliminary paper presented
at the August 2003 conference entitled “Microeconomic Analyses of Labor
Reallocation,” sponsored by the W.E. Upjohn Institute for Employment Research.
Lefranc, A. 2003. “Labor Market Dynamics and Wage Losses of Displaced Workers in
France and the United-States.” Paper presented at the August 2003 conference
entitled “Microeconomic Analyses of Labor Reallocation,” sponsored by the W.E.
Upjohn Institute for Employment Research.
Leonard, J. and L. Jacobson. May 1990. “Earnings Inequality and Job Turnover.” The
American Economic Review, Papers and Proceedings of the Hundred and Second
Annual Meeting of the American Economic Association 80 (2), 298-302.
Page 20
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
20 of 21
Levy, F. and R. Murnane, 1992. “U.S. earnings levels and earnings inequality: a review
of recent trends and proposed explanations. Journal of Economic Literature 30
(3), 1333-1381.
Levinsohn, J, and W. Petropoulos. 2001. “Creative destruction or Just Plain
Destruction?: The US Textile and Apparel Industries Since 1972.” NBER
Working Paper 8348.
Lorence, J. 1991. “Growth in Service Sector Employment and MSA Gender Earnings
Inequality.” Social Forces 69, 763-83.
Lloyd-Ellis, H. 1999. “Endogenous Technological Change and Wage Inequality.”
American Economic Review 89 (1): 47-77.
Machin, S., A. Ryan and J. van Reenen. “Technology and Changes in Skill Structure:
Evidence from an International Panel of Industries.” Centre for Economic
Performance, London School of Economics, L.M.C.T.S.C. Discussion Paper No.
4, May 1996
McCall, L. November 2000. “Explaining Levels of Within-Group Wage Inequality in
U.S. Labor Marketrs.” Demography 37 (4): 415-430.
Morris, M and B. Western. 1999. “Inequality in Earnings at the Close of the Twentieth
Century.” Annual Review of Sociology 25: 623-57.
Murphy, K. and F. Welch. 1993. “Industrial Change and the Rising Importance of Skill,”
in Danziger and Gottschalk, eds. 101-132.
Neal, D. 1995. “Industry-Specific Human Capital: Evidence from Displaced Workers.”
Journal of Labor Economics 13 (4).
Nelson, J. and J. Lorence. 1988. “Metropolitan Earnings Inequality and Service Sector
Employment.” Social Forces 67: 492-511.
Peter, K.S. 2003. “Skill-Biased Transition: The Role of Markets, Institutions, and
Technological Change. ” Paper presented at the August 2003 conference entitled
“Microeconomic Analyses of Labor Reallocation,” sponsored by the W.E. Upjohn
Institute for Employment Research.
Rodriguez-Planas, N. Feb. 2003. “Playing Hard to Get: New Evidence on Layoffs,
Recalls, and Unemployment.” Mathematica Policy Research. Paper presented at
the August 2003 conference entitled “Microeconomic Analyses of Labor
Reallocation,” sponsored by the W.E. Upjohn Institute for Employment Research.
Page 21
Professional Paper
Final Draft
Delia C. Elder
May 2, 2004
21 of 21
U.S. Bureau of Labor Statistics. 1994. “Technology and Its Effect in the Steel Industry,”
Washington, DC: U.S. GPO.
Wolff, E. 2002. “The impact of IT investment on income and wealth inequality in the
postwar U.S. economy.” Information Economics and Policy 14 (2).
Youtie, J. 1997. “Toward a Cross-Case Analysis of Outcomes of Exemplary
Engagements by MEP Centers.” Manufacturing Modernization: Learning from
Evaluation Practices and Results: Proceedings of Third Workshop on the
Evaluation of Industrial Modernization Programs, Aberdeen Woods, Atlanta,
1996. Philip Shapira and Jan Youtie, eds. Atlanta, GA: School of Public Policy
and Economic Development Institute, Georgia Institute of Technology: 97:167-
178.

IS IT GLOBALIZATION,

OR A GLOBAL HEGEMONY?

THE UNITED STATES

VERSUS THE WORLD

by Naseer Aruri

Globalization and the Search for a New Foreign Policy Consensus

American foreign policy elites are being challenged to embrace a new vision of a world order and determine the US role in that order now and beyond, into the new millennium. A search has been underway since 1989 for an intellectual rationale for a new American global role in post-cold war conditions. A national security doctrine based on anti-communism would have to be replaced, now that the pretext for the unprecedented level of a militarized U.S foreign policy has disappeared. This has been going on despite a decided shift in public opinion towards things domestic.

If there is a consensus emerging from the raging arguments, it is that the phenomenon of globalization is setting the pace, not only for America's global role, but for the entire world as well. The globalization thesis is being promoted as a new, inclusive and integrative force. It is becoming a powerful ideological tool to contain and suppress nationalist and oppositional movements around the world, in much the same way as these forces were kept under pressure during the Cold War. Today, however, the anti-Soviet, anti-nationalist weapon is being replaced by the seemingly benign tool of "free trade." Today's penetration targets not only the natural resources of what was called the third world, but also the markets, human resources and the ever-growing newly-created consumers. The term ascribed to this new phase of capitalist accumulation and colonization is the non-threatening and rather benign "globalization." In reality, globalization has never ceased to be integral to the process of capitalist development. As a process, it represents mobility of investment capital in pursuit of cheap, docile labor in stable environments. The state, in fact, has to some degree been reduced to the role of finding and assuring favorable business opportunities for its corporations.

The proponents of globalization posit that the most important post-cold war dichotomy is that of integration versus fragmentation. Thus again, the world is seen through the prism of the good versus the evil, with globalization representing the satisfaction of economic needs, removal of trade barriers in pursuit of upward movement and freedom form want.

The forces of integration are presumed to be those global institutions of management of the economy, environment and politics. Although these institutions have been in existence throughout most of the Cold War period, their functions have been revised and their missions broadened. They are the defacto tools of global governance in this unipolar world. They include GATT, the World Trade Organization(WTO), the Security Council, the World Bank and the IMF, among others. NAFTA, APEC, the EU, and the G-7 are among the regional instruments which perform their duties in synchronization with the global institutions. Together, they are touted as the instruments of "integration" and homogenization which can be counted upon to counter the global forces of "fragmentation." The latter are invariably defined as populist nationalists, Islamic fundamentalism, terrorism and ethnic rivalries.

Management of the economy and keeping the peace to assure stability are said to be entrusted to these global and regional institutions, depending on what is at stake. And herein lay our central question: How does the United States, as the lone surviving super power, define its new role in this process of globalization? How does it allocate its economic, diplomatic and military resources on behalf of this overarching goal of "integration?" By the same token, how does it allocate the same resources to combat the forces of "fragmentation?" When does it step into the quagmire and when does it overlook the infractions?

Business Leads the Way, Again

Just as in post-1945, when the US was in a position of military and economic ascendancy, today after the collapse of the USSR, the US is in a similar position, while Japan and Europe are struggling to come out of recession. The new hegemony is anchored in the world's largest economy which keeps growing, in an unprecedented military superiority and leadership in global information technology. US present hegemony is therefore based on military and economic power. Economic power derives from the acceleration of globalization, in which the US provides key leadership in:

a) promoting free trade,

b) setting standards for delivery of information.

The dominant role in that process has been carried out by business, which has been undergoing restructuring since the 1980's-consolidation through mergers and acquisitions, and downsizing in the name of efficiency, as efficiency is being recognized as a motor of growth. A US business model which had guided foreign policy in the post -1945 period was later sharpened and fine-tuned during the Reagan period, whose landmarks were deregulation, privatization, free trade. US free enterprise was setting the pace for US foreign policy, which itself began to assume the mission of internationalizing privatization. i,e. redistributing wealth in favor of the rich and mighty. The export of Reaganism was thus a significant step towards globalization.

The supermarket approach to business has widened the gap between rich and poor in the US by promoting centralization of resources and concentration of wealth through consolidation. Meanwhile the thrust of hyper-industrialization(the upsurge of the service sector), and proliferation of "thinking products," i.e., software, movies, books, music, have created de-industrialization. The process of deindustrialization has gone furthest in the U.S. Where by the early 1990s, industry accounted for only 29.2 per cent of GNP compared to 38.7 per cent and 41.8 per cent in Germany and Japan, respectively. Knowledge-based industries, requiring skills and training rather than formal education, have accelerated the process of deindustrialization.

It is the US business model of the 80's which is being advanced by U.S. foreign policy elites as the way to the future. It might be said that through globalization, this institutionalization of business values in public policy was extended abroad, going back to the early period of the cold war, when US foreign policy was promoting these values in the name of countering the Soviet "Threat." That "threat" amounted to nothing more than an ideological counter to spread these business values to the third world, which was somehow seen as up for grabs. The values, guided by US foreign policy, were used to under gird what was generally known as national security--realistically defined in economic (corporate) interests, but publicly proclaimed in military/strategic terms, i.e., the Soviet "Threat." In that regard, the phrase "free trade" was a euphemism for US economic penetration of the global south (the third world) which constituted the strategic goal of thwarting the movement of decolonization. Likewise, non-alignment, as a form of national self-assertion was opposed as a phenomenon favoring the USSR. The United Nations General Assembly was also portrayed by the U.S. pejoratively as a debating forum for anti-Western leaders, and was demonized as a hotbed of third world radicalism. Likewise, the 1970's drive for a new economic world order was ridiculed as a misguided attempt to impose an international welfare program for the poor, at variance with the sacred principles of free enterprise and market regulation. The impetus for the concept of a new information world order during the 1970s was also condemned as a form of censorship, at variance with the presumed Western tradition of free press and free expression.

Much of that US opposition to decolonization and its variants during the Cold War was actually prompted by the desire to promote these business values in international relations. Moving the world toward hard-edged capitalism was the goal of US foreign policy after 1945, but the term globalization, which sums up that process, had not been coined yet. Today, after the collapse of the Soviet Union, and as the US embarks on a second era of world hegemony, the goal remains the same, except that globalization has emerged as a benign label for the same goal. "Free trade" and the "free flow of oil" from the Middle East were expressions which provided a rationale for post-1945 penetration. They are still providing a similar rationale for post-cold war, and post-gulf war penetration. It is a change in vocabulary but not in the substance. The difference is that the post-cold war hegemony is more fierce and potentially more ruinous, yet seemingly gentle because it is couched in benevolent terms. It is also more damaging due to the absence of a counter-balance. The Soviet Union had been dissolved and the third world is immersed in debt and lies conquered for the time-being.

The Equation of Power and Purpose: A Benevolent Global Hegemony

The two elements which provided a rationale for post-1945 and post-Cold War U.S. penetration (free trade and free flow of oil from the Middle East) have been consistently portrayed by U.S. foreign policy elites as the equivalent of national security. The question however, is: Are they sufficient to maintain a new foreign policy consensus without being linked to some kind of a threat and a national purpose, that would rally the American public? The existence of a "threat" had galvanized a rather diverse American society. Today, the "national purpose" is not as clear, and the "threat" is hardly discernible; yet there is a preponderance of power and an overwhelming feeling of victory. Herein lay the paradox and the challenge to rearrange the equation of power and purpose. A number of scholars and strategic specialists have produced literature which yields some important conclusions about the new role and the relationship between power, threats, and purpose.

There is hardly anything new in the attempt to find the foreign devil ; in the 1950's it was communism, and the Soviet Union, but after the collapse of the Soviet Union, the new demons are Islam and Confucianism, which translate to the Arab/Muslim world and China. As is well-Known, that was the result of Samuel Huntington's initial search to pinpoint the enemy. Four years later, however, ( November 1997) Huntington revised his thesis. He now concludes that "Islamic Fundamentalism is too diffuse and too remote geographically," to constitute the principal threat to Western civilization. He turns inward and discovers that the culture of diversity in the United States has become the newest threat to American society.After all The Census Bureau of the U.S. Commerce Department has estimated that by the year 2050, the 'white' population in the U.S. will decrease from 75 to 53 percent of the total population.

At a conference co-sponsored by the Carnegie endowment and the World Policy Institute, comprising 25 scholars and foreign policy luminaries (April 11-12 1997), Huntington lamented that we are "in a position where, unlike in the cold war, when our major problem was to develop the power to support our purposes in the world, now our major problem is to develop the purpose to guide our power."

Other writers are not as uncertain as Huntington about what to do with American power. Somehow, the world needs America's "benevolent global hegemony," we are told by conservative commentators William Kristol and Robert Kagan (Foreign Affairs Summer 1997). The maintenance of American military strength has been a favorite theme of influential writers, scholars, and commentators, who would argue that, even if such a force is not used, the credible threat to use it would be sufficient in itself to compel good behavior on the part of would-be detractors. Joshua Muravchik writes in a new book that a more forceful America could generate "more accommodating behavior from other states who know that they have little to fear or distrust from a righteous state." Similarly, the Israeli-American strategist ,Edward Luttwak, told the Carnegie conference (April 11-12 1997); "the challenge lies in making others believe that the united States will actually use force when its interests are involved." He claimed that "the perception of American military superiority would prevent military buildups by other states and thus promote the cause of democracy."

What is really interesting about the current discourse is a curious connection between democracy, free markets, and the use of force. It almost seems that noblesse obligue is being revived at the close of the 20th Century, with the notion that opening markets for American investments constitutes a civilizing mission that also brings democratic ideas and a healthy respect for human rights. And just as the concept was sustained by military force during the 18th and 19th centuries, American force is now kept in reserve as guardian of globalized capital--to intimidate the defiant, to punish dissidents and to underwrite the profits made in the name of free trade. Just as the foreign policy theorists conceptualized the post-World War II policy of containment in terms of a mission, supposedly "not chosen" by the United States, but was somehow "incumbent" upon it; today's intellectuals emphasize a similar obligation. In their quest for military ascendancy and globalized hegemony, some have even found a rationale in the Federalist Papers, which served as the basis for the American constitutional union against those who preferred a looser confederation. Professor Daniel Deudney, of the University of Pennsylvania said that the first ten of the Federalist Papers provide a "clear intellectual and moral warrant for an American effort to bring all of the democracies into union with the United States." Charles William Maynes of the Carnegie Endowment, summed up what appears to be a consensus among the advocates of a "benevolent global hegemony," when he discerned an economic imperative for such hegemony, consisting of promoting democracy, free markets, and human rights. He counseled US policy-makers to "use the abundant American strength in ways that will benefit the international community." The fact that much of the world, including loyal allies, views this benevolent hegemony as a globalized arrogance seems to elude these theorists and opinion leaders.

This perspective of American power, which may be described as a form of "neo-reassertionism," is also propounded by Orientalist scholars such as Bernard Lewis, pro-Zionist scholars such as Daniel Pipes, Fouad Ajami, Connor Cruise O'Brien, and right-wing pro Zionist journalists such as William Safire and Judith Miller of the New York Times, and syndicated columnist Charles Krauthammer of the New Republic, among others. Perhaps it is not a coincidence that an exceedingly high proportion of the neo-reassertionists are also Zionists and advocate the ready use of American power against any type of Arab resistance to Israeli colonization and US hegemony, euphemistically described as fundamentalism and terrorism, hence "fragmentary."

Another group of scholars who seem to have almost no impact on present U.S. policy include Yale historian Paul Kennedy, who often makes comparisons between the US in the 1990's and Spain in the 1630's and 40's--two declining empires justifying intervention on grounds of "reputation." Emphasizing the non-military dimensions, Kennedy drew up a list of ills facing the Spanish empire and now facing the American Empire: less competitive industries, a torn social fabric, archaic tax structure, rural poverty, low productivity, national indebtedness, mediocre education systems, crumbling infrastructure, decayed inner cities, and yet despite all these things, massive global commitments. Kennedy commented that the decline of American power is illustrated in the kind of military interventions whose real purpose is merely the recovery of America's lost self esteem.

Other "declinists" include Christopher Layne of The Cato Institute, who perceived the display of military strength during the 1991 war against Iraq as a facade masking economic weakness and social decay. He wrote: The Cold War's end ought to have been a paradigm-shifting event that triggered a fundamental reappraisal of US foreign policy. The outbreak of the Gulf War aborted the new debate about US foreign policy. The danger now is that...the costs and consequences of America's world policeman role will be insulated from scrutiny.

Thus while declinists urge a long over-due debate as well as cost-cutting, the dominant group of neo-reassertionists argue for expanded global governance, not by multilateral arrangements and genuine international institutions, of course, but through multilateralized US intervention, and U.S.-dominated institutions such as the World Bank, the I.M.F., the G-7, and the World Trade Organization(WTO). But as America's ability to secure multilateralized cover for its military interventions declines, as demonstrated presently in Iraq , the hawkish intellectuals (known as neo-reassertionists) and Clinton's strategists will increasingly push unilateral intervention. A trigger happy U.S populace and mainstream press revealed their inclinations in 1997 when seven out of ten Americans favored US military action against Iraq. Editorialists and Op-Ed writers exercised no restraint when calling for the use of American force. Meanwhile, the ongoing debate seems to be limited to policy-oriented intellectuals, strategists,and governmental as well as corporate elites, leaving the wider public at the periphery in a society which professes to be pluralistic and democratic , which prides itself on free discussion. That debate clearly favors the militaristic approach and the arrogance of power.