M

a
n
p
o
w
e
r

H
u
m
a
n

R
e
s
o
u
r
c
e
s

L
a
b
The Theory and Practice
of Pay Setting
Alex Bryson & John Forth
MHRLdp001
December 2006

MANPOWER HUMAN RESOURCES LAB, CEP

As one of the worlds foremost employment agencies, Manpower is seeking to
contribute to the analysis of and policy debate around the issues of the
changing world of work. In May 2006 it therefore established with the Centre
for Economic Performance the Manpower Human Resources Lab at the
London School of Economics. The aim of the Lab is to become a leading
centre studying the impact of human resource decisions and labour market
trends on productivity at firm, national and global levels.



SENIOR STAFF

Prof. Stephen Machin, FBA Director MHRL & Research Director, CEP
Alex Bryson Manpower Research Fellow, CEP
Professor Richard Freeman Director, Labour Studies Programme NBER & Senior
Research Fellow CEP
Daniel Kasmir HR & Corporate Affairs Director, EMEA Manpower Inc
Prof. Alan Manning Director, Labour Markets Programme CEP
Prof. J ohn Van Reenen Director, Centre for Economic Performance

J o Cantlay Administrator



CONTACT US


Website: www.lse.ac.uk/manpower

Postal address: Manpower Human Resources Lab
Centre for Economic Performance
London School of Economics
Houghton Street
London WC2A 2AE

Email: j.m.cantlay@lse.ac.uk

Telephone: +44 (0) 20 7955 7285
Abstract

This review focuses on pay variance across workers, employers and across
time and illustrates how theories of pay determination can shed light on this
variance. We discuss the limitations of the orthodox economic approach to
pay setting and emphasise the importance of labour market imperfections and
the unique character of the labour contract in determining wage outcomes.
Two broad conclusions emerge: first that no single theory of pay setting has
an over-riding claim to virtue; and second that, in spite of the institutional
knowledge generated by industrial relations, obtaining a greater
understanding of the general pattern of wages remains one of the principal
challenges to the discipline at the present time.


Acknowledgments

Alex Bryson is the Manpower Fellow at the Manpower Human Resources
Lab, Centre for Economic Performance and Principal Research Fellow of the
Employment Group at the Policy Studies Institute. J ohn Forth is a Research
Fellow at the National Institute of Economic and Social Research (NIESR).

The authors gratefully acknowledge the contribution made by Neil Millward to
earlier work, on which some of this material is based. Alex Brysons
contribution was funded by the Wertheim Fellowship at the Harvard Law
School and the Manpower Human Resources Data Lab.


JEL Classification Numbers: J 31, J 33, J 38, J 42, J 45, J 52, J 71

1. Introduction

In modern economies employers purchase labour power from workers who, in
general, obtain their income from the sale of their labour, having little or no capital
from which to make a living. From the employers perspective, the wage is the price
of labour, that is, the cost of securing the workers productive capacity. However, as
management theorists (and efficiency wage theorists before them) tend to
emphasise, pay can also be used by employers as a motivator to elicit greater
worker effort. As Marx (1976) emphasized, employers generally purchase units of
workers time, that is, their potential to labour, rather than discrete, well-defined units
of effort. It is possible for employers to create incentives for workers to work harder,
or more intensively over a given period what Marx termed the real subsumption of
labour (ibid.) or more extensively, that is to increase the time they supply their
labour over a given day, or even over a career (through increased tenure). This is
one reason why the price of labour is not necessarily set mechanistically at the
intersection between the supply and demand curves.

From the employees perspective, pay is the reward for labour, that is, the actual
effort of producing goods or services. The precise nature of the payment varies
greatly across workers, and may include not only monetary income paid shortly
before or after the labour is supplied, but also deferred payments, such as pensions
and holiday pay, together with non-monetary rewards such as health insurance and
other fringe benefits which are often rated by employees as more valuable than their
monetary equivalents (Dale-Olsen, 2006).

Despite these complexities, this characterisation of pay nonetheless implies that
pay is the outcome of a relatively straightforward, private, economic transaction
between employers and employees in which workers receive pay in return for effort.
But this is not the case for at least three reasons. First, strong social customs and
norms govern what is considered to be the appropriate trade-off between effort and
reward. There is, for instance, a strong tradition relating to what constitutes a fair
days work for a fair days pay. Governments often feel impelled to intervene in
wage setting when market-set wages do not meet what appear to be the subsistence
needs of workers and their families. They may do so by requiring certain minimum
1
standards in terms of pay levels, or may supplement pay through public funds, for
instance through the payment of tax credits.

A second reason why pay is more than the outcome of a simple economic
transaction between employer and employee is that employers are often judged by
what they pay and employees by what they receive. Again, social norms play an
important role. For instance, employers may be cognisant of their image in the
public eye and with investors: they may gain plaudits for being a good employer
behaving in a socially responsible fashion, whereas their stock price may respond
more favourably by bearing down on labour costs. For employees, ones social
status is often bound up with ones wage and even how it is paid (hourly, weekly or
as an annual salary). It may have a direct bearing on the workers well-being, not
only in terms of what she can wear and eat, but in terms of what she can borrow,
and how she is perceived by work colleagues, friends and relatives. Above all else,
workers well-being is highly correlated with perceptions of their pay relative to their
peers (Brown et al., 2005).

Third, employers and employees sometimes choose to involve others in wage
setting. For instance, employees may join together to bargain with one or more
employers in setting wages, either through trades unions or other forms of collective
action. Similarly, employers may choose to act collectively if, for instance, they wish
to avoid outbidding one another for scarce labour, choosing instead to take wages
out of competition by setting single rates in particular industries, localities or
occupations.
2

This paper focuses on pay variance across workers, employers and across time and
illustrates how theories of pay determination can shed light on this variance. We
show the importance of labour market imperfections and the role of institutions in
influencing pay levels, the pay distribution and wage growth in the UK and
elsewhere. Yet, in spite of the time and effort academics and other analysts have

2
Adam Smith noted: What are the common wages of labour, depends everywhere upon the contract usually
made between those two parties, whose interests are by no means the same. The workmen desire to get as much,
the masters to give as little as possible. The former are disposed to combine in order to raise, the latter in order
to lower the wages of labour. (Wealth of Nations, Book 1, Chapter 8)
2
spent tackling the issue of pay determination, as will become apparent, much
remains to be understood.

The paper begins by examining the degree of variance in pay levels.


2. Patterns of Pay


Any cursory examination of variations in earnings indicates that there are systematic
differences in the pecuniary rewards received by different types of worker. Earnings
statistics illustrate, for example, how wages vary by occupation or skill level; they
point towards the role of collective bargaining in raising wages for unionised
employees; and they reveal the earnings differentials that exist between men and
women. Such statistics therefore begin to indicate some of the factors that play
either a direct or an indirect role in pay setting.

Comparisons between workers of different types are usually made on the basis of
gross earnings per hour worked, with the earnings figure being inclusive of bonus
payments or shift premia and the number of hours being inclusive of overtime.
Hourly earnings figures have the advantage of being unaffected by inequalities in
hours worked (for example, between men and women), whilst gross earnings have
not been altered by differences in the rates of taxation that might apply to low and
high earners.
3
The inclusion of both paid and unpaid overtime also permits a more
accurate indication of the relationship between wages and effort than measures
based purely on contracted terms. Unpaid overtime may be of lesser concern to an
employer, since only paid overtime hours contribute to the overall wage bill.
Nevertheless, obtaining greater numbers of unpaid hours from the workforce can
serve as one means of bearing down on unit labour costs.

Table 1 presents summary statistics on gross hourly earnings in the UK across a
selection of employee and job characteristics, and includes indicators both of central

3
Net earnings are, of course, the more appropriate indicators in considerations of income
3
tendency and wage dispersion. Whilst the mean utilises information from the whole
distribution including the lowest and highest earners, the 50
th
percentile, or median,
on the other hand, has the advantage of being unaffected by changes at the
extremes of the distribution, caused for example, by an increase in the minimum
wage or the payment of annual bonuses to city executives.
4
Thus the median might
be said to better capture the situation for the average employee. The mean and
median differ because, at the national level, there tends to be a long upper tail in the
earnings distribution, populated by a small number of highly-paid employees.

Table 1 Gross hourly earnings, Autumn 2005


Percentiles Percentile
ratios

Mean
10
th
50
th
90
th
50:10 90:10

All employees 10.90 5.00 8.84 19.22 1.77 2.17

Gender:
Men 12.21 5.40 10.32 21.92 1.91 2.12
Women 9.70 4.80 8.00 16.99 1.67 2.12

Hours worked per week:
Full-time (30 or more) 11.75 5.54 10.00 20.51 1.81 2.05
Part-time (1-29) 8.21 4.33 6.42 14.73 1.48 2.29

Highest qualification:
Degree or equivalent 16.16 7.00 14.58 27.03 2.08 1.85
Higher education 12.49 6.18 11.74 19.67 1.90 1.68
GCE A-level or equivalent 10.05 5.00 8.66 16.83 1.73 1.94
GCSE grades A-C or equivalent 8.67 4.69 7.56 14.16 1.61 1.87
Other qualifications 8.39 4.69 7.29 13.23 1.55 1.81
No qualifications 7.12 4.50 6.25 11.11 1.39 1.78

Unionisation:
Collective bargaining 11.72 5.78 10.35 19.22 1.79 1.86
No collective bargaining 10.65 4.80 8.21 19.24 1.71 2.34

Source: UK Labour Force Survey, Autumn 2005.

The table shows a gender wage differential of around 20 per cent in favour of male
employees, an advantage for full-time workers over part-time employees of around
30 per cent and extensive variation between employees with different levels of
academic achievement. There is also considerable variation in wages within each of

4
That is, unless the median wage earner seeks to maintain differentials with such groups.
4
these groupings. The employee at the 90
th
percentile of the earnings distribution
typically has gross hourly earnings that are around twice as large as those of the
median worker and around four times as large as the employee at the 10
th

percentile. This partly reflects variations in the way that the employment relationship
is formulated. One can think of variance in wages being generated by the
characteristics of individuals, such as their qualifications and experience, and by the
way in which different employers reward workers for the jobs they do and the
attributes workers have. If wages were set in a purely competitive market, workers
wages would reflect their own productivity so that wages would be independent of
the firm employing the worker. However, as discussed later, there are a variety of
reasons why workers pay may vary across firms, such as the acquisition of firm-
specific human capital, union bargaining or the desire of employers to pay above
market wages to capture the rents that good workers generate. Table 1 shows, for
example, that employees whose pay is negotiated by trade unions, for example,
have wages that are 10 per cent higher, on average, than non-unionised employees.

This raises the question: how much of the variance in wages in an economy occurs
within workplaces, and how much occurs across workplaces? The answer to this
question has important implications for labour mobility and productivity. If, for the
sake of argument, all workers in a firm are paid the same wage, but all firms pay a
different wage that is, all variance is accounted for by variance across workplaces
workers would have to move to obtain higher wages, and they may be particularly
concerned about which firm they join when first entering the labour market. At the
other extreme, say all firms are identical in the distribution of wages they pay their
workers that is to say, all firms are a microcosm of the whole economy such that all
pay variance is due to within-firm differences it would matter very little which firm
employs worker since the opportunities for advancement are equivalent in all firms.

Our own analysis of linked employer-employee data for Britain shows that workplace
characteristics can account for at least as much of the variation in hourly wages as
differences in employees educational achievements and work experience.
5


5
Full results available on request.
5
Figure 1a: Male Wage Inequality in Britain, 1972-2004
(Source: Kondylis and Wadsworth, 2006)

1
1
.
2
1
.
4
1
.
6
1
.
8
r
e
a
l

w
a
g
e

r
e
l
a
t
i
v
e

t
o

1
9
7
2
72 77 80 85 90 95 2000 2004
Men
90th pctile GHS 90th pctile LFS
50th pctile GHS 50th pctile LFS
10th pctile GHS 10th pctile LFS


The figures in Table 1 take no account of non-pecuniary rewards, such as subsidised
or free health care or childcare, or employers pension contributions. These tend to
increase the differential in rewards between higher and lower paid workers, for
example widening the gap between men and women (J oshi and Paci, 1998;
Anderson et al, 2001) and increasing the premium associated with collective
bargaining (Forth and Millward, 2000; Budd, 2004). From the employers
perspective, the picture is further complicated by the addition of indirect labour costs,
such as social security contributions. Indirect costs account for an average of around
13 per cent of total hourly labour costs in the UK, compared with an average of
around 25-30 per cent in the EU as a whole.

Returning to the pattern of direct earnings, there have been notable changes in the
degree of wage dispersion over time. Inequality in gross earnings grew in Britain in
the 1980s after a century of relative stability (Gregg and Machin, 1994). Figures 1a
and 1b show trends in real wage growth relative to 1972 for men and women
separately. During the 1980s, there was rising wage inequality between middle
6
earners and top earners (the 90:50 percentile ratio), and middle earners and low
earners (the 50:10 percentile ratio). The trend was apparent for men and women. In
the first half of the 1990s, the male wage distribution continued to widen, though at a
slower pace, driven by increases in the relative earnings of top earners. Women, on
the other hand, experienced a small compression in wages. Under the recent Labour
governments, male wages have compressed a little whereas, among women, there
has been wage compression at the top and some wage compression between
middle and low earners.
6

Figure 1b: Female Wage Inequality in Britain, 1972-2004
(Source: Kondylis and Wadsworth, 2006)

1
1
.
2
1
.
4
1
.
6
1
.
8
2
2
.
2
r
e
a
l

w
a
g
e

r
e
l
a
t
i
v
e

t
o

1
9
7
2
72 77 80 85 90 95 2000 2004
Women
90th pctile GHS 90th pctile LFS
50th pctile GHS 50th pctile LFS
10th pctile GHS 10th pctile LFS



Attempts to explain the causes of variations in wages across workers and over time,
are only partially successful. Statistical analyses that take into account differences in
individual, job and employer characteristics, for example, are typically able to explain
little more than 50 per cent of the cross-sectional variation in gross hourly earnings.
This is partly indicative of the difficulties that arise when trying to devise indicators

6
Widening wage inequality since the 1980s is not a universal phenomenon. Indeed, it was only
experienced in a few industrialised countries, notably the USA.
7
that adequately proxy important factors such as employee productivity or bargaining
power. But to some degree it also indicates the limitations of our ability to understand
the complexity of pay setting. The rest of the paper discusses what is generally
known about the nature of this complexity.


3. Neo-Classical Labour Market Theory


A conventional starting point in economics is the neo-classical model of the labour
market under which wages, like other prices, are determined by the forces of supply
and demand. Faced with a supply of workers willing to be employed (each with their
own preferences which determine how much labour they are willing to supply at
given wage levels), the employer (normally a firm) decides how many and which
types of workers to employ at wage levels at which it can trade profitably in its
chosen product market. For a given type of labour the firm will hire increasing
numbers of workers up to the point where their wage is equal to their marginal
product the additional added value gained by the firm from its least productive
worker. Workers, for their part, will offer themselves for employment until the next
potential recruit considers the wage an insufficient inducement to do the job. Under
perfect competition, and with freedom of entry and exit for firms and workers, these
market forces produce a single wage at which supply and demand are in equilibrium.
In this situation, employers and employees are each price takers, with none having
the power themselves to set wage rates. A state of disequilibrium may arise from
time-to-time, in which wages do not lie on the intersection of the supply and demand
curves. But this is expected only to be a short-run phenomenon whilst prices adjust
to changes in levels of demand or supply.

Neo-classical labour market theory predicts higher wages for workers who are more
productive (i.e. able to generate higher levels of output given a fixed amount of
capital). But it says little about what determines worker productivity, although there is
a broad notion of skill which implicitly equates with it in many instances. Adam
Smith noted that output and wages were related not only to innate abilities but also
8
to acquired skills (Smith, 1776), but human capital theory has more recently
formalised the conception of education and training as investments which could
generate returns, both to employers in the form of increased output and to
employees in the form of higher wages (Becker, 1964).
7
Estimates of the wage
returns to educational investments are widespread. Recent estimates for the UK
indicate that the standard academic route of five or more GCSEs, two or more A-
levels and a degree raises wages by around 70 per cent (McIntosh, 2006). However,
it is argued by some that educational attainment acts primarily as a signal of
unobserved ability, which is then rewarded appropriately in the labour market, rather
than serving as a productivity-enhancing investment (Arrow, 1973; Spence, 1973).
Estimates of the wage returns to skills are less widespread. Human capital theory
distinguishes between generic skills, which can be applied in a wide range of jobs or
industries, and firm-specific skills. Using a detailed questionnaire covering a wide
range of generic skills, Green and Dickerson (2004) showed that computing skills
and high-level communication skills attract higher wages, as do the ability to act
without close supervision and the capacity to manage varied tasks. Others have
identified returns to industry-specific skills, which become apparent through a drop in
wages when displaced workers find new jobs in a different sector (Neal, 1995). If
wages are related to skill, it is natural to expect that changes in the distribution of
skills within the economy will affect wage differentials. McIntosh actually finds no
change in the return to academic qualifications over the period 1996-2002, despite
the expansion of higher education (ibid). But others point to a wider process of skill-
biased technical change as a cause of rising wage inequality. The hypothesis is that
technological change has generated increasing demand for skill and led to the
automation of intermediate low-skilled jobs, resulting in congestion at the lower end
of the labour market, which, in turn, pushes down the wages in low-skilled jobs that
can not be automated. This is consistent with developments in the 1980s in the
United States (Autor, Levy and Murnane, 2003). However, in the 1990s the United
States experienced growing demand for both high-skilled jobs and lower skilled jobs
that can not be automated (Autor, Katz and Kearney, 2006). This growing demand
for lower-end jobs, coupled with the passing of the glut of displaced routine labour,
has resulted in wage increases at the bottom end of the labour market over this

7
Education was previously viewed primarily as a consumption good or status good (Machin and
Vignoles, 2005).
9
period. The net effect is stabilization in the wage distribution in the lower half of the
pay distribution as indicated by the 50:10 ratio and continuing growth in pay
disparities in the upper half of the pay distribution. Goos and Manning (2003) and
Spitz (2005) also point to a polarization of work over the last quarter century in the
UK and West Germany respectively. Although this story seems plausible, one might
wonder, since technological innovation is all-pervasive, why these forces have note
produced increasing wage inequality in more countries.

Criticisms of the neo-classical approach

Commentators have rejected the neo-classical model to different degrees. Almost all
would agree that it is an abstraction from reality, since few markets are perfectly
competitive and firms and workers are rarely free to enter and leave these markets
without delay. Some see it, then, as a simplification, but one which nonetheless
provides a good indication of general tendencies (Hicks, 1932: 5). Others are more
categorical in rejecting it as a meaningful approximation of the wage-setting process
which overplays the role of the invisible hand of the market (e.g. Rubery, 1997;
Manning, 2003).

Those who are critical of the neo-classical model typically point to a range of
empirical facts that appear to be at odds with its depiction of the labour market. First,
they point to the existence of long-term unemployment and wage rigidities as
evidence that markets do not, in fact, clear. So whilst real wages of the lowest decile
fell over the period 1979-2000 in North America and Australia this is very unusual
and is virtually unheard of elsewhere (Glyn, 2001). The combination of low
unemployment and low wage inflation in the United States and United Kingdom in
the late 1990s also cast doubt on the mechanics of the disequilibrium adjustment
process, depicted by Phillips (1958), whereby wage growth is expected to be
negatively associated with levels of unemployment. Furthermore, Blanchflower and
Oswald (2003) have demonstrated that, contrary to theoretical expectations in which
higher wage levels are associated with greater unemployment (all other things being
equal), it is common to observe the reverse. Dunlop (1988), for his part, asserts that
unemployment has little if any role to play in influencing wage reduction, except in
relatively small establishments.
10

A second empirical challenge comes from studies which indicate a range of
indeterminacy in wage-setting (Lester, 1952). The theory suggests that a perfectly
competitive labour market will be characterised by a single wage for a particular type
of labour. Instead, studies of similar occupations in specific localities suggest that
there may be considerable variation between firms (e.g. Dunlop, 1957; Mackay et al,
1971; Brown et al, 1980; Nolan and Brown, 1983; Gilman et al, 2002). Whilst the
theory allows for some short-term variance, as employers and employees adjust to
changes in the demand for labour, this should be only a temporary phenomenon with
any longer-term differences likely to be imperceptible (Hicks, 1932: 24-7). However,
the evidence suggests that differences may be more pervasive, and it is argued that
such differences do not simply reflect returns to firm-specific skills. Rather, it is
suggested that the reality of wage setting rejects the neo-classical depiction of wage
setting in favour of a much more complex account in which employers and
employees may variously play an active part in determining wages.

Imperfectly competitive markets

Relaxing the assumption of perfect competition leads to different predictions of the
outcomes of pay setting, and to some plausible explanations of variations in wages.
Departures from perfect competition in product markets and labour markets are both
relevant; in either arena, some level of concentration or collaboration among either
buyers or sellers can generate a degree of control over prices in this case, the
price of labour which may enable wages to sustain departures from employees
marginal product in the long run.

Monopoly or oligopoly

If a firm is a monopolist or oligopolist in its product market, the control it acquires in
relation to consumers over product pricing enables it to generate excess profits,
which in turn provide it with the freedom to pay wages that are above the level that
would be determined by the market in the case of perfect competition. Many of the
most conspicuous national monopolies in the UK were removed in the 1980s through
privatisation, and statistics show that industry concentration ratios fell during that
11
decade (Clarke, 1993). But it is not clear whether the degree of product market
competition has changed to any significant degree in aggregate since that time. The
picture is probably rather varied: in some industries such as banking and
accountancy, mergers have increased the level of concentration in recent years
(Mahajan, 2005). The degree of concentration among producers tends to be capped
by the regulatory activities of national and supra-national governments, but these
thresholds are set fairly high. In a wide range of industries in the UK, the top five
businesses account for at least one-third of gross output and in some this proportion
is currently above two-thirds (ibid.), suggesting at least the potential for some control
over prices.
8


Although such product market power means that the firm has less need or incentive
to bear down upon labour costs, there is no guarantee that any share of the
additional income that results from reduced competition in a particular product
market will be passed on to employees in the form of higher wages. However, the
evidence indicates that this does happen, at least to some degree, with a number of
empirical studies having demonstrated a positive association between wages and
past profitability in large samples of firms (see, for example, Blanchflower et al, 1996;
Hildreth and Oswald, 1997; Van Reenan, 1996). There may be a number of
alternative rationales for this behaviour. The monopolist or oligopolist may choose to
pay higher wages of its own volition, perhaps to demonstrate fairness, to reduce
turnover, or to boost recruitment. Alternatively, it may be induced to do so by a trade
union representing some or all of its employees.
9


Evidence over the last few decades points to a sizeable union wage premium in
many countries (Blanchflower and Bryson, 2003). Unions also raise non-pay fringe
benefits (Forth and Millward, 2000) so that the full labour cost increases associated
with unionisation are higher than those associated with pay alone. The size of this
premium is correlated with the ability of unions to monopolise the supply of labour,

8
The example of the rail industry since privatisation shows the limitation of using concentration
ratios to measure the degree of competition, however, since the individual train operating companies
still retain monopolies over many of their routes.
9
Although it is theoretically possible for other forms of labour organization to affect wages, empirical
evidence for Britain indicates that only unions actually affect wages (Millward, Forth and Bryson,
2001).
12
often proxied in empirical investigations by the presence of a closed shop (Stewart,
1987), union density (eg. Hirsch and Schumacher, 2001) or collective bargaining
coverage (Forth and Millward, 2002). Indeed, it is often the case that unions seem to
procure no premium where they have no power to monopolise the supply of labour.
For instance, Stewart (1987) found the workplace-level union recognition premium
for skilled manual workers in 1980 was confined to those working in establishments
with a closed shop.

Changes in unions monopoly power may also bring about changes in the level of the
union wage premium. Stewart (1995) attributes part of the decline in the union wage
premium in Britain to the demise of the closed shop over the period. One might have
anticipated a secular decline in the union premium arising from increased
opportunities for union employers to substitute non-union for union labour. These
opportunities arise with declining union density in developed countries and the ability
to out-source production to non-unionised labour in less developed countries. In fact,
there is little empirical evidence in support of a secular decline in the union wage
premium in Britain and the US. Instead, there has been counter-cyclical movement
in the premium arising from unions ability to resist the downward wage flexibility
non-union workers experience in recession (Blanchflower and Bryson, 2004). There
is also no evidence that increased competition induced by higher import penetration
has lowered union wage premia. In fact, quite the reverse: Bratsberg and Ragan
(2002) and Blanchflower and Bryson (2003) show higher import penetration raises
the premium, perhaps because organised workers are more able to resist downward
wage pressures that this induces in non-union firms, at least in the short-term.

Unions can exert power over wage setting in perfectly competitive markets, but neo-
classical theory would predict that this cannot be sustained in the long run because
these firms would have to raise product prices above market levels and would go out
of business. However, some firms have a greater ability to pay above-market rates
than others. One of the first to emphasise this was Slichter (1950) who illustrated
the point by drawing attention to the positive rank correlation in 1939 between
industries with high profit ratios to sales and those with high unskilled earnings.
There is conflicting empirical evidence as to whether unions affect the longevity of
firms (Bryson, 2004a) and no direct evidence that any negative effect on their
13
survival is attributable to union-induced labour costs. However, unions are
associated with lower employment growth (Bryson, 2004b).

Together, the persistence of a union wage premium and lower employment growth
rates in the union sector have been treated as evidence of a right-to-manage model
in which the employer unilaterally sets employment levels conditional on above-
market wages set in negotiation with unions. Although the right-to-manage model
has been treated as a stylised fact, imperfections in the product and labour markets
tend to be correlated (Booth, 1995: 59), leading to a situation of bi-lateral monopoly,
where the wage-employment outcome is a function of their relative bargaining power
rather than being precisely determined by the market (McDonald and Solow, 1980;
Manning, 2003).

Union involvement in wage setting can also be expected to have an effect on wage
inequality. A priori, the direction of this effect is ambiguous. Their pursuit of
standard rates for jobs is a force for pay equalization, as is the threat of unionization
which forces non-union employers to raise pay or benefits to keep unions out
(Rosen, 1969). On the other hand, they bargain for higher rates for their members
relative to non-members. Thus their impact on the wage distribution will depend, in
large part, on how many workers they organize, which workers they organize, and
variation in the wage premium across workers. Two decades ago Freeman (1980)
found that unions tended to reduce wage inequality among men because the
inequality-increasing effect of bargaining higher rates for members versus non-
members was smaller than the dispersion-reducing effect of standardizing pay within
the union sector. This remained the case for the last thirty years of the twentieth
century in the United States, Britain and Canada (Card, Lemieux and Riddell, 2003).
Because unionized men tend to be found in the middle of the skill distribution, the
union effect on compressing wages in this part of the distribution has a further
equalizing effect on the dispersion of wages. However, unions do not reduce wage
inequality for women in these countries because unionized women are concentrated
in the upper end of the wage distribution, the union wage premium tends to be larger
for women than it is for men and, whereas the premium is higher for lesser-skilled
than higher-skilled men, this is not the case for women.

14
There are also differences across gender in terms of the effect of unionization on
trends in pay dispersion trends. Declining unionization rates among men explained
a significant fraction of the growth in wage inequality in the US and UK whereas the
decline in unionization was much more modest for women and had little impact on
female wage inequality (Card, Lemieux and Riddell, 2003; Gosling and Lemieux
(2004).Of course, declining unionization is not a universal phenomenon: countries
such as Belgium, Denmark, Finland, Ireland, Norway and Sweden have experienced
increased union density in recent decades and declining pay dispersion.
10


Monopsony and oligopsony

Whilst the role of product or labour market monopolies is a common feature of the
literature on wage setting, the role of monopsony or oligopsony where significant
power is held by the buyer rather than the seller - is less extensively discussed.
Monopsony power in product markets, in particular, is given little attention. There are
examples in the industrial relations literature of how dependence relationships
between small firms and large retailers may affect work organisation and the
exercise of management authority (Moule, 1998), but there is little evidence of an
impact on wages, despite the obvious implications that dependent relationships
might have for cost structures in supplying firms. The evidence that is available
seems to be largely concerned with the role of multinationals in determining labour
standards among suppliers from developing countries. One would expect the effect
to be equivalent to that of monopsony in the labour market, where the dominance of
a single firm or small group of firms may drive down wages within a local labour
market.

The commonly-cited examples of monopsonistic labour markets tend to be of
company towns. These are a distant memory to most, but labour market monopsony
may equally provide an explanation for the present-day reduction in earnings that
tends to accompany the arrival of some large retailers into local labour markets (as

10
In many continental European countries most workers are covered by a collective bargaining
agreement. A recent review of union wage effects in these countries concludes: both the level at
which bargaining occurs and the degree of coordination between bargaining levels can affect wage
levels and wage dispersion. However, effects differ across studies (Bryson et al., 2006).
15
described by Neumark et al, 2005). It is also suggested that it may be a feature of
the labour markets for nurses (Sullivan, 1989) and university professors (Ransom,
1993), at least in the United States. Each of these examples serve to indicate that
restricted mobility on the part of employees is an important contributory factor in
consolidating the power of prospective employers in the wage setting process.
Employers may use this power to different degrees. The discriminating monopsonist
may take the opportunity to practice wage discrimination and pay each employee at
their individual reservation wage, whilst the simple monopsonist, on the other hand,
will instead pay a uniform wage that is just sufficient to fill all vacant posts. But there
are further implications, since the theory predicts that simple monopsony will result in
levels of wages and employment that are both lower than in a competitive market (a
further feature described in Neumark et als retail study), and so monopsony has
implications that extend beyond simply wages.

If monopsony power is depressing wages in a particular labour market, then one
would expect that any action that serves to reduce the power of the employer in the
labour market may thus serve to raise wages among employees. Illustrations in the
US literature tend to focus on the removal of the reserve clause in professional
baseball, which tied players indefinitely to the club for which they first signed; the
removal of the clause led to a significant transfer in income from owners to players in
the form of higher wages (Kahn, 1993). An equivalent example in the European
setting is the Bosman ruling which, since 1995, has allowed professional soccer
players in Europe to negotiate their own deals with a new employer once the
contract with their current club had expired.
11
In cases of simple monopsony, a
restriction on employer power may also result in an increase in the level of
employment without lowering wages, thereby raising economic efficiency. Indeed,
monopsonistic labour markets are put forward as one explanation as to why the
introduction of regulations governing wage setting, such as the Equal Pay Act and
the National Minimum Wage, do not appear to have brought about job losses
(Manning, 2003).


11
Whilst there is anecdotal evidence of a resulting rise in wages for the top players (Grant, 2005), the
paucity of public data on wages pre- and post-Bosman make quantification of the average effect of the
ruling difficult
16

4. The Uniqueness of the Labour Contract


Although the discussion thus far has focused on wages, the basic approach to wage
determination has differed little from the approach that might be applied in the
determination of prices for commodities. However, an acceptance of the uniqueness
of the labour contract is fundamental to a further understanding of the complexities of
wages. The open-ended nature of the employment contract is a key feature, bringing
with it uncertainty and thus potential controversy over the precise terms of the
exchange. Wages can then feature among the devices that employers may call upon
to raise the level of effort. The personal aspect of the exchange also means that
wages may be critically influenced by norms, social institutions and political
processes. An appreciation of these various factors which is one of the features
that sets the discipline of industrial relations apart from some parts of mainstream
economics means giving a lower priority to the process of market clearing,
accepting that the formal economic analysis of supply and demand is too narrow and
that the social aspect of wages must also be carefully considered (Brown and Nolan,
1988; Dunlop, 1988). This in turn, leads one to give greater attention to the context in
which wages are determined and also to downplay the degree of apparent rationality
that might be accorded to wage outcomes. In the following sections, we review a
number of approaches that view pay setting from this general perspective.

Efficiency wages

A separate line of development from neo-classical labour market theory is efficiency
wage theory. It abandons the conception of the spot market with a single wage for a
given type of labour and acknowledges that employers can choose different wage
levels to elicit different levels of effort. It therefore provides one potential means of
explaining variations in wages among like workers.

If wages fall short of what the worker considers a fair reference wage (such as the
rate set in other firms) the theory posits that a rise in wages will raise workers effort
(Akerlof, 1982, 1984; Akerlof and Yellen, 1990). The rise might also allow a firm to
17
recruit higher quality workers (Weiss, 1980), reduce turnover (Salop, 1979), and
improve employee morale, all of which can be productivity-enhancing.
12
But the
wage must not be set too high as there are diminishing returns since effort can only
rise so far. Accordingly, a wage exists the efficiency wage where the marginal
cost of increasing the wage equals the marginal gain in productivity. The notable
implication, from the perspective of this review, is that this wage is set independently
of labour market conditions outside the firm: the principal determinant is the influence
of wage changes on worker effort within the firm.

The incentive for firms to pay above the competitive market rate will differ according
to their production technologies which imply different gains to reducing turnover,
recruiting high-quality labour and so on. For instance, firms in industries reliant on
extensive firm-specific training will benefit from pay policies that lower workers
propensity to quit. Since the decision to pay above-market rates is driven, in part at
least, by production technologies which differ across industries, some point to
efficiency wages as a reason for inter-industry wage differentials. Kruger and
Summers (1988) and Gibbons and Katz (1992), for example, point to efficiency
wages as a reason for unequal pay for equally qualified workers. This is contested
by others who argue that such wage differentials are largely due to unmeasured
differences in worker quality (Murphy and Topel, 1990; Abowd, Kramarz and
Margolis, 1999). Reasons for inter-industry wage differentials may be disputed, but
Fehr, Gchter and Kirchsteigers (1996) laboratory experiments identify a clear
causal relationship between efficiency wages and effort.

Shapiro and Stiglitzs (1984) version of efficiency wage theory focuses on the use of
high wages to reduce work-avoidance (shirking) among employees. The problem
arises because employment contracts are incomplete giving employees some
discretion about the effort they put in. Where it is very costly to monitor worker
inputs firms may choose instead to pay higher wages, thus increasing the cost of job
loss to the worker. Krueger (1991) suggests that the higher rate of pay in company-
owned burger companies compared to franchised outlets in the same firms is
accounted for by local franchise owners monitoring their employees more easily, and

12
For a review of efficiency wage theory in its various guises see Akerlof and Yellen (1986).
18
thus having less need to buy the cooperation of their workers. Freeman et al.
(2006) have similarly shown that, in keeping with efficiency wage theory, US workers
are more likely to intervene when they feel co-workers are shirking if they believe
that their fixed pay is at or above the market level.
13


This leaves open the question of whether, as predicted by efficiency wage theory,
marginal wage increases raise productivity sufficiently to pay for themselves. Using
data from large North American firms spanning 1970-1985 Levine (1992) found that
they did.
14
Levines (1992) paper is also notable in finding lower productivity effects
of high wages in the presence of unions. This is consistent with the proposition that
unions push wages above the optimum efficiency wage level to a point where there
are diminishing productivity returns to further wage increases. Much earlier evidence
in favour of efficiency wages comes from Henry Ford who subsequently described
the doubling of wages to five dollars a day as one of the finest cost-cutting moves
we ever made because it dramatically reduced turnover and absenteeism and
raised labour discipline and output quality (Raff and Summers, 1987). In spite of this
evidence, some have questioned just how much time and effort managers devote to
devising incentive pay systems with the intention of eliciting greater productivity from
their workers. Others have questioned the whole basis of efficiency wage theory,
contending that there is no necessary relationship between pay and productivity
(Brown and Nolan, 1988: 352). This latter position raises substantial questions over
the continued efforts of employers to link the two.

Linking pay to performance

Paying higher fixed wages under an employment contract where the firm pays a
time-based wage is only one way in which firms might tackle work-avoidance in
situations where effort is difficult to monitor. An alternative is to offer a piece rate
which allows the worker to decide how much to work and thus how much to get paid.

13
In their study of a US multi-plant firm Cappelli and Chauvin (1991) attribute the link between
declining disciplinary layoffs and increasing plant wages relative to the local wage to the negative
effect of relative wages on the incentive to shirk.
14
For the UK see Wadhwani and Wall (1991). Both Levine (1992) and Wadhwani and Wall (1991)
note that their finding would also be consistent with rent-sharing theories in which higher
productivity leads to higher wages. Wadhwani and Wall try to account for this endogeneity using
instrumental variables.
19
Economists view piecework as a means of inducing greater effort by equating the
marginal value of an extra unit of output with the marginal cost of producing it
(Weitzman and Kruse, 1990). Both economists and sociologists point to the usage
of piecework when monitoring the workers effort is difficult. Economists couch this in
terms of the imperfect information firms have about worker productivity when
monitoring effort is difficult or costly. Sociologists, on the other hand, tend to view
piecework and other forms of performance-based pay - as a mechanism for
managerial control when management can not provide adequate supervision (Gallie
et al., 1998: chapter 3). Thus Hobsbawm gives the example of deep pits being sunk
on time-based wages for the first few hundred feet and on piecework thereafter
(Hobsbawm, 1964). Either way, one might anticipate a lesser need for monitoring
when piecework is in place since, as Marx remarked, the superintendence of labour
becom[es] to a great extent superfluous (Marx, 1976: 695).

The originators of the British Workplace Industrial Relations Surveys reported their
findings on payments-by-results (PBR) under the heading Systems of payment and
control alongside methods for controlling time keeping and payments while sick
(Daniel and Millward, 1983: 200). They went on to argue (1983: 205): Traditionally
the purpose of PBR systems of pay has been to encourage workers to increase
effort and output.In practice.there has been a tendency for PBR to become more
an instrument of management control designed to ensure consistency of output. In
the Donovan tradition, PBR was treated as part of the problem of shop floor
bargaining and a cause of industrial unrest (Daniel and Millward, 1983: 292). As
Blanchflower and Cubbin (1986: 26) note, PBR has traditionally been included in
analyses of strike propensities since, through the need for regular adjustments to
pay, it increases opportunities for disagreement.

Marx (1976: 698-99) reports Factory Inspectorate figures showing that, in the mid-
1800s four-fifths of those employed in British factories were paid by the piece. Marx
notes that it is the form of wage most appropriate to the capitalist mode of
production. Yet today pure piece-rate pay is rare, confined to occupations where
outputs can easily be observed, such as sales or fruit picking.

20
In the United States shared forms of compensation (share ownership plans, stock
options, profit and gainsharing) have been rising since the 1980s, a trend that seems
to have accompanied greater employee involvement in decision-making (Dube and
Freeman, 2006). The trends are very different in the UK. Task discretion declined in
Britain in the 1990s (Gallie et al., 2004).
15
Both profit-related pay and share-
ownership schemes grew in the 1980s, with government tax incentives playing some
part, but their incidence has remained fairly static since then (Bryson, 2006). There
also seems to have been little change in the incidence of payments-by-results since
1990 (ibid). Nevertheless, these forms of payment were widespread in Britain in
2004, when one-third of workplaces use individual payments-by-results, 16 per cent
used subjective merit pay and one-quarter used group-based payments-by-results
(Bryson and Freeman, 2006). One-quarter had profit-related pay schemes and one-
fifth had employee share-ownership schemes (ibid.).

In theory, performance-based pay will generate inequalities in output and thus
earnings where workers are heterogeneous in effort and ability, but it can only do so
if there is a direct link at individual level between output and earnings; this link is not
always present as some forms of performance-related pay measure output at group
or even firm level. Using the Workplace Employment Relations Survey 1998 Belfield
and Heywood (2001) found incentive pay increased wage dispersion at workplace
level whereas profit-sharing and share-ownership did not. Using WERS 2004
Bryson and Freeman (2006) find the only performance-based pay associated with
employees gross earnings was individual performance-based pay, which was
associated with higher earnings. It was also the only form of pay affecting
workplace-level pay dispersion.

The link between variable pay and productivity and performance is unclear a priori.
First, the employer incurs costs through the introduction and maintenance of variable
pay which might outweigh its potential benefits (Freeman and Kleiner, 2005; Levine
and Tyson, 1990). Second, variable pay has the potential to demotivate workers
(Brown and Nolan, 1988). Employees may perceive the pay/performance link to be
unfair if, for instance, performance is measured with error or employees have not

15
In any event, Bryson and Freeman (2006) find shared compensation schemes are not strongly
associated with employee decision-making in Britain.
21
been consulted about the criteria governing the scheme (Marsden, 2004). Third, it
may not be optimal to involve employees in the decision-making that might come
with shared compensation (J ones, 1987), perhaps because employees may not be
best placed to make decisions (Loveridge, 1980).

In large firms employers may encourage effort through promotion as a reward, rather
than directly through pay for performance. In these settings with internal labour
markets (Doeringer and Piore 1971) pay varies little within grades but it increases
with job changes. Where promotion slots are limited, workers are thus motivated to
supply effort by virtue of the wage increases they would earn if promoted, with the
competition for promotions then resembling a form of tournament where winner
takes all (Lazear and Rosen 1981).

The advantage of tournaments to an employer is that it is often easier to observe
relative performance than absolute performance. Additionally, it may be in the
interests of the company to structure pay so that the winner makes very large sums
as a way of spurring on those lower in the hierarchy as well as giving the CEO
herself the incentive to perform well. Tournament theory therefore provides one
possible explanation for the high wages of CEOs and, more generally, wage
inequality within firms. Tournaments might be viewed as one form of deferred
compensation whereby worker and firm commit to each other. Under schemes of
deferred compensation, workers are paid above their marginal product when old and
below it when young. This may be because firms want to limit costly labour turnover
(Salop and Salop, 1976) or because distinguishing good from poor workers takes
time. Lazear (1979) shows how the common upward-sloping age-earnings profile
can discourage workers from shirking.

However, tournaments may make workers reluctant to help one another. Freeman
and Gelbers (2006) laboratory experiment shows that total tournament output
depends on pay inequality according to an inverse-U shaped function. They find
productivity is lowest when payments are independent of the participants
performance; it rises to a maximum at a medium level of inequality, but then it falls at
the highest level of inequality. Thus inequality can be too high as well as too low for
efficiency. In one case study, a fruit farm registered a 50 per cent increase in worker
22
productivity after moving away from relative incentives to piece rates, an increase
that the authors of the study explained by the shift away from a system in which
workers knew that increased individual effort could have negative effects on co-
workers earnings (Bandiera, Barankay and Rasul, 2005). One implication is,
however, that where a system generates positive externalities as in the case of
group incentive schemes this should generate still greater productivity. A further
implication is that notions of fairness are clearly important in the labour market.

Fairness, norms and reciprocity

In standard neo-classical economics actors are driven by self-interest and decisions
are made to maximise financial gain. However, it is commonly recognised that
exchanges between people are often conducted according to shared social norms of
fairness and reciprocity which are anchored in reference points that are amenable to
change.
16
These notions are formalised in equity theory where the target
relationship sought by individuals is equality between their own reward per unit of
input (effort, investment) and their cognition of others rewards per unit of input.
Employee perceptions of what they contribute to the organization and what they get
in return, and how this ratio compares to others inside and outside the organisation,
determines how fair they perceive their employment relationship to be (Adams,
1963). Naturally, only some others are seen as salient in these comparisons and
these are most likely to be in the same work-group, workplace or firm (Brown et al.,
1998). This may explain why employees pay satisfaction is so strongly associated
with their wage rank within the workplace (Brown et al., 2005). However, employers
have little influence over which individuals their employees choose as salient others
in their equity assessments, so they may include similar workers in the external
labour market. Such comparisons are particularly likely to be brought to bear in union
bargaining. And when the union is dominated by less skilled employees, as with
general rather than craft unions, the pressure from the union is generally to favour
the lower paid and to compress wage differentials among covered workers.

16
Concerns about the changing conceptions of fairness under capitalism led Engels to express
concern about the continued use of the rallying cry a fair days wages for a fair days work, a
phrase he readily admitted had been the motto of the English working-class movement for the last
fifty years (Engels, 1881).
23

The inputs that form components of the equity calculation are variously defined in
different occupations. In manual occupations it is common to prioritise physical effort,
dexterity and skill in using tools and machines, whereas in non-manual occupations
greater priority is accorded to literacy, communication, information processing,
responsibility and so on. There is a broad correspondence between the inputs
forming the foundation of social norms of equity and the investments that increase
human capital. The pay premium for supervisory and managerial jobs, compared
with workers they supervise, also rests upon a widely-held norm that responsibility
for other peoples work should be rewarded.

These concepts first came to the fore among economists with Akerlofs (1982) partial
gift exchange model of the labour contract in which he argues that worker effort
depends on work norms of a relevant reference group and that the firm can alter
these norms and thus effort by paying workers a wages gift in excess of the
minimum required in return for above-minimum effort. Akerlof and Yellens (1990)
fair wage-effort hypothesis suggests that workers form a notion of the fair wage
and, if the actual wage is lower than this reference point, they will withdraw their
effort in proportion. The fair wage thus plays a role in wage bargaining where
entitlements are fashioned by reference points. More broadly, notions of fairness can
underpin the role of custom and practice in wage determination (Brown and Nolan,
1988; Rubery, 1997) though, as Hicks (1932) points out, forces of supply and
demand can occasion changes in customary structures.

Greenbergs (1990) field experiment presents evidence in support of the fair wage-
effort model by comparing worker rates of theft from a company in response to
proposed pay cuts. Where management provided significant information on reasons
for pay cuts and expressed remorse subsequent employee theft rates were
significantly lower than in the scenario in which management provided inadequate
information and showed no remorse. The author suggests that perceptions of
inequity caused employees to take actions to restore equity. Often what matters is
employee perceptions of distributive justice. Thus Cowherd and Levine (1992) link a
negative association between the quality of service and goods as perceived by
24
customers and the size of pay differentials between hourly paid employees and top
management to a diminution in citizenship-type behaviour.

Laboratory experiments repeatedly show individuals and firms behaving in
accordance with norms of fairness and reciprocity, leading to outcomes that would
not otherwise be predicted by neo-classical theory. For instance, experiments
support Solows (1990) contention that wage rigidity may be due, at least in part, to
conformity with social norms. Falk, Fehr and Zehnder (2006) show that the
introduction of a minimum wage law changes what is considered a fair wage by
workers, thereby forcing firms to increase wages above the level required by the
minimum wage and hold them there even when the minimum wage is rescinded.
This helps to explain why fast food restaurants in the USA increased wages for
workers by more than was necessary following increases in the minimum wage
(Card and Krueger, 1995; Katz and Krueger, 1992) and why the opportunity to pay
sub-minimum wages to youth workers has no discernible effect on teenage workers
wages (Katz and Krueger, 1992).

Questions of fairness and equity are also salient in debates regarding CEO pay. In
their study of over 120 firms tracked over a five year period Wade et al. (2006) show
that norms of fairness are salient to top decision makers and that over or
underpayment of the CEO cascades down to lower organizational levels, partly due
to CEOs using their own power not only to increase their own salaries but also those
of their subordinates. Further, CEOs serve as a key referent for employees in
determining whether their own situation is fair, influencing their reactions to their
own compensation. They find that when lower-level managers are underpaid relative
to the CEO (that is, underpaid more than the CEO or overpaid less), they are more
likely to leave the organization.

This discussion of fairness has focused on the ways in which levels of pay are
influenced by the earnings of co-workers in the firm or wider labour market. But pay
may also be influenced by relativities in the conditions of work.



25
Compensating wage differentials

The importance of balancing the physical and mental demands of work with the
environment in which that work is undertaken finds formal acknowledgment in the
principle of compensating wage differentials or equalizing differences. This
approach reinforces the idea that wages are not simply the reward for effort or skill,
but that the context in which that effort is delivered is important in determining the
price of labour. Thus one should not necessarily expect wage equality between jobs
of equal value, but rather equality between the overall job package, which takes into
account not only money wages but also non-pecuniary benefits and the whole range
of working conditions. The principle also departs from the traditional framework of
supply and demand by according workers some measure of preference over issues
other than the monetary rewards.
17

The theory is often expressed in terms of the wage compensation for dangerous
work that brings the risk of injury or even death. Most workers can be expected to
value both higher wages and greater levels of safety, but some are presumed willing
to accept some additional risk in exchange for a higher wage rate that is sufficient to
maintain the same overall level of utility. Equally, a specific firm can choose to invest
in mechanisms and procedures that offer workers greater safety, or they can obtain
the same level of profitability by economizing on safety and distributing the savings
to workers in the form of higher wages. Yet the principle of compensating
differentials can be extended beyond the issue of workplace risk. This was clearly
acknowledged in the writings of Adam Smith (1776):

the wages of labour vary with the ease or hardship, the cleanliness or dirtiness, the
honourableness or dishonourableness, of the employment. A journeyman
blacksmith, though an artificer, seldom earns so much in twelve hours, as a collier,
who is only a labourer, does in eight. His work is not quite so dirty,
is less dangerous, and is carried on in day-light, and above ground.


17
That is to say the metric of monetary compensation is insufficient in determining a worker's utility from work.

26
One might therefore expect higher wages to be explained, in at least some
instances, by the presence of any one of a number of negative job attributes that
might include dirt, unsociable hours and insecurity as well as danger. Similarly,
under the theory, one might expect lower wages to result from the availability of
positive job attributes such as job security, hours flexibility, good opportunities for
career progression, or meaningful work.

To properly examine the theory, one must compare wages and working conditions
within like jobs and for like workers. The payment of London allowances are
perhaps one of the most obvious examples that wage variations can, on some
occasions, be explained by the principle of compensating differentials (the
compensation here reflecting the higher cost of living for those employees living and
working in the capital). The payment of premiums for shift-working or unsociable
hours (weekends or nights) are another. Of course, there could be reasons other
than compensating differentials for pay premia of this sort such as differences in
the characteristics of those who are paid the premia and those who are not. But
Lanfranchi et al (2002) and Kostiuk (1990), among others, show that the higher
wages typically observed for shift-workers in aggregate wage data do not wholly
reflect differences in worker characteristics, indicating that the premium does indeed
represent some compensation for accepting shift work over regular hours.

Other studies have indicated the apparent presence of wage differentials to
compensate for greater levels of job stress (French and Dunlap, 1998) and job
insecurity (Moretti, 2000). But in these and other areas it is common to see different
results depending upon the data and methodology employed. McNabb (1989) finds a
wage premium for unfavourable working conditions among manual workers in
Britain. But as he willingly accepts (pp.328-330), and a number of other authors also
point out (e.g. Purse, 2004; Gunderson and Hyatt, 2001), reliable estimation of the
reward for engaging in dangerous work is difficult because workers are not randomly
allocated across jobs of varying degrees of risk. One therefore needs to account for
any biases arising from worker selection into good or bad jobs when evaluating the
compensating differentials hypothesis. It might be, for instance, that the
compensating differentials are underestimated because employees with high
earnings potential are able to secure safer jobs (Viscusi, 1978). Gunderson and
27
Hyatt (2001) use a variety of approaches to show why past estimates may have
been either negative or too small.

Alternatively, workers may sort into jobs on the basis of who can be most productive
under dangerous conditions (Garen, 1988) in which case the analyst may overstate
the size of compensating differentials. It is difficult to separate compensating
differentials from returns to skill where the employee is being rewarded for being
able to manage the risk inherent within a particular job. This skill might result from
prior training, experience or simply the possession of a cool-head which enables the
employee to be productive in an environment where others might fear to work or be
prone to panic. Clearly this cannot account for all of the possible examples of
compensating differentials cited above; context clearly is important in some
instances, independent of skill (viz. London Allowances). But a number of wage
compensations might arguably be seen in this light: for example, the ability to
remain alert at night or the ability to complete tasks under extreme stress. One
necessary implication is that a person in a non-risky job may not necessarily be able
to attain the higher wage of person in a risky job simply by choosing to accept a
higher level of risk; equally, it does not necessarily follow that seemingly under-paid
employees undertaking meaningful work in the voluntary sector could necessarily
raise their earnings by moving to equivalent jobs in the business sector.

A further area of empirical investigation has been the extent to which lower wages
accompany the provision of fringe benefits such as private health insurance and the
like. For instance, Olson (2002) finds that married women in the United States
accept a wage penalty of around 20 per cent in return for employer-provided health
insurance. We are not aware of any equivalent studies in the UK. Brown and
Crossmans research in advance of the introduction of the National Minimum Wage
suggested that employers would cut fringe benefits as a means of accommodating
the increase in money wages (Brown and Crossman, 2000), but there has been little
evidence of this in practice (see, for example, Lucas and Langlois, 2003), probably
because such benefits were already comparatively rare in low-paying sectors such
as hospitality and textiles.

28
This helps to highlight the fact that higher wages are often found in conjunction with
favourable levels of non-wage benefits. Dale-Olsen (2006) finds such a positive
correlation in a recent and thorough examination of data on the Norwegian labour
market. Similarly, various studies find that the degree of wage inequality between
low and high-wage groups tends to increase when one considers the complete
compensation package including fringe benefits (J oshi and Paci, 1998; Pierce,
2001). The co-existence of high wages and good conditions might arise through the
bargaining power of trade unions (see Fairris, 1989). Similarly, the co-existence of
low wages and poor conditions could be indicative of employers monopsony power
in the labour market. However, a positive correlation between wages and working
conditions might also imply some segmentation of the labour market.

Segmentation and discrimination

At it simplest, the theory of dual or segmented labour markets proposes a primary
sector where a workers effort is hard to measure and monitoring is costly, plus a
secondary sector where output is measurable and supervision easy. In the primary
sector, firms offer high wages, good working conditions, secure employment and
promotion opportunities and in return obtain higher levels of labour productivity. In
the secondary sector firms offer low wages, poor conditions, insecure employment
and little chance of promotion and they obtain lower levels of productivity. Whilst this
echoes efficiency wage theories, the key distinction of the segmentation approach is
to argue that structural features of the labour market limit workers ability to transfer
from the secondary to the primary sector. Such limits on mobility might arise
because closed internal labour markets in the primary sector offer limited points of
entry or because employers place a low (or even negative) valuation on work
experience gained in the secondary sector. The effect is that the segment in which
an employee is located plays a part in determining their wage, both now and in the
future, quite apart from their marginal productivity

The identification of labour market segments is, however, not straightforward. In their
study of industrial organisation and low pay, Craig et al (1982) followed Piore (1979)
in distinguishing between primary industrial sectors characterised by large firms,
using modern technology, facing a stable product market and enjoying market power
29
and secondary sectors composed of small firms, using traditional technology,
facing competitive and often declining product markets and frequently acting as sub-
contractors to the primary sector. They found that wages in primary sectors were
typically higher and were more likely to be governed by formal rules and supported
by union organisation. But they also observed that primary industrial characteristics
provided no guarantee of primary employment conditions, particularly where
institutional constraints prevented worker organisation. They found examples of two-
tier workforces within primary firms, where the bulk of the workforce was paid good
wages and enjoyed good working conditions, but in which a minority of marginalised
occupations (in their case directly-employed canteen workers) were employed on
relatively low rates of pay more characteristic of the secondary sector. Other
researchers have focused on occupational distinctions in the application of a dual
labour market approach, including Theodossiou (1995) who examined differences in
wage determinants among career and non-career workers, and Blackaby et al
(1995) who distinguished between wages of regular and marginal workers. The work
of Atkinson (1984) is also relevant. Alternative conceptions have focused on the
limited mobility of immigrant guest workers in the German economy (e.g. Constant
and Massey, 2003).

As ever, practice is not as neat as some readings of the theory might suggest. But
the notion that employees mobility (and thus their wage) is restricted by factors other
than abilities and preferences is valuable. It may be useful, for example, in
attempting to understand the part-time pay penalty among female employees in
Britain, which remains around 10 per cent after accounting for various other factors
that determine wages (Manning and Petrongolo, 2005). Part-time work is
concentrated in particular occupations, such as customer service, care work or
cleaning, and within such occupations the penalty for part-time work is generally
found to be low (3 per cent on average, according to Manning and Petrongolo). This
suggests the presence of distinct segments of the labour market, identified by the
form of hours that employers are willing to consider, with employees needing to offer
full-time hours in order to gain access to the better-paid segments.

An alternative view point would see the lower pay of part-timers as part of an overall
compensation package whereby wages are traded for flexibility, following the
30
principles of compensating wage differentials (or similarly, Hakims preference
theory (Hakim, 2000)). But there is clear evidence of limited mobility between part-
time and full-time work: White and Forths study of movements from unemployment
found that less than one in four of those entering part-time jobs were in full-time jobs
after 5 years, despite substantial proportions stating a preference for full-time work
(White and Forth, 1998). A variety of studies also show that part-time work
experience is poorly rewarded by employers in comparison with experience in full-
time work (Myck and Paull, 2004; Harkness, 1996; Swaffield, 2000).

Seen in this light, the lower rewards for part-time work might be seen as a form of
discrimination, since workers are seemingly receiving a lower wage because of
reasons other than their personal abilities. There are a number of possible
explanations for discriminatory behaviour. The most obvious is commonly couched in
terms of tastes (Becker, 1957), with an employer paying lower wages to women
(say) in order to compensate for the additional psychological cost of employing them
over equally-qualified men. The impetus might equally come from other workers with
a preference for working alongside employees of their own gender, or even
consumers with a preference for purchasing goods or services from someone of their
own ethnic group. Wage discrimination might also come about in the absence of
prejudice if a particular characteristic (e.g. gender) is thought to provide a signal
about a persons productivity: so-called statistical discrimination.

The wage effects of discrimination are not easy to discern, not least because some
of the impact on wages may arise indirectly through discrimination in access to
employment, training opportunities or promotions. The advent of the Equal Pay Act
had a dramatic impact in reducing the gender pay gap in Britain in the 1970s, with
the wages of women in manufacturing rising by 21 per cent relative to male wages,
having moved marginally in the opposite direction in the previous decade (Manning,
2003: 211). This partly came about through the re-negotiation of collective
agreements that had previously specified lower wages for women than men within
the same jobs (ibid). There has been a more gradual closing of the gap since that
time, but the gap between male and female full-time wages nonetheless stood at 17
per cent in 2006. Direct wage discrimination still plays some part in explaining wage
variations between some workers: a fact that is evident from the number of
31
successful claims brought each year under equal pay legislation (3,722 in 2005/6)
(Employment Tribunal Service, 2006). But it is difficult to judge how prevalent the
practice might currently be in the wider population. Statistical analyses enable one to
identify differences in the characteristics of men and women (such as their skill
levels) which can together explain around 50 per cent of the current gap. But one
needs to have considerable faith in the comprehensiveness of the statistical
specifications if one is to attribute the whole of the unexplained portion to wage
discrimination.

Since some part of the current gender pay gap is accounted for by differences in
skills and educational attainments, the elimination of the gap is not realistic in the
short-term. But one of the most likely means of reducing the gap further is perhaps
through the more extensive use of equal pay audits within individual firms. There has
been some progress on this point in Britain, under the prompting of the Equal
Opportunities Commission, but progress has been most extensive in the public
sector where the state has either seen fit to employ coercion (as in the case of
central government) or has taken pre-emptive action on a national scale in response
to high-value equal pay claims (as in the NHS). At the same time, the ability of
current approaches to tackle the issue is becoming further comprised by the
increasing fragmentation of enterprises through sub-contracting (Forth and Metcalf,
2005). And more broadly, it is recognised that the notion of comparable worth is
limited in the extent to which it can tackle the social valuation of skills, whereby jobs
held by either men or women may be considered unskilled because they are
dominated by female employees (Crompton and Sanderson, 1990). Such social
processes can, of course, serve to reinforce the segmentation tendencies noted
above.

The state as employer

Many of the determinants of wages in the private sector necessarily also have an
influence in the public sector, not least the concern to appropriately reward skills and
abilities and to facilitate adequate levels of recruitment and retention. But it is clear
that the wider political context of pay setting takes on a more prominent role in wage
determination for public sector employees. Whilst pay has a political dimension in the
32
private sector, this is generally the result of pressures internal to the firm (see the
earlier discussion of fairness). In the public sector, however, the wider public has an
interest because of their role in funding state expenditure via taxation and their ability
to choose the identity of the ultimate paymaster the government.
18


Managing this public interest involves balancing a number of tensions. First, it
imposes an additional emphasis on value for money, since the burden of funding pay
increases falls to the general public unless such rises pay for themselves through
equivalent gains in efficiency. But any attempts to constrain wages must also
consider the potential impact on recruitment and retention, and thus service quality,
since a large percentage of public sector employment is in key areas such as
education and health which impinge directly on peoples lives and livelihoods. The
critical nature of such services, along with the high degree of unionisation and the
tendency towards national pay scales, also mean that battles over wage settlements
tend to be fought out on the public stage, further accentuating the politicisation of
public sector pay.

Governments, for their part, have a further interest in the stability of public finances,
which promotes prudence when setting the parameters for wage settlements in the
public sector. Most public services are labour intensive, with Brooks (2004)
estimating that 50 per cent of revenue costs in the UK public sector are accounted
for by wages and salaries. Governments also have a broader economic interest to
maintain price stability and, for this reason, incomes policies have tended to bite
more in the public sector in those periods when inflation was a major issue. Wage
setting is also informed to some degree by the public sector ethos towards being a
good employer, which encourages it to do more than simply meet statutory
requirements in areas such as equal pay.
19
However, there is considerable variation
in the extent to which central government can ensure that each of these principles
are brought to bear across the full range of public sector organisations, some of

18
Wage costs are clearly also of relevance to the consumers of private sector services but, in product
markets with some degree of competition, they can exercise pressure on wages by seeking out lower-
cost producers. In contrast, the general public must continue to pay for the wages of public sector
employees even if they do not personally use the services that they provide.
19
There is considerable variation in the extent to which this principle is upheld, however (see Bewley,
2006, for example).
33
which may operate at some considerable distance from the centre (see Bewley,
2006, for a discussion of the variation in practice in respect of equal pay).

A number of different arrangements exist for setting wages in the UK public sector,
including centralised collective bargaining, index-linking and Independent Pay
Review Bodies. The fact that each tends to yield national settlements has important
implications for wage differentials, with Grimshaw (2000) suggesting that the greater
use of centralised pay determination (when compared with the private sector) may
have both curbed the growth of wage inequality in Britain and limited the gender pay
gap. Despite the commonly-held view that employees in the UK public sector are
underpaid, statistical analyses generally find a pay premium for public servants in the
order of around 5 per cent, on average, after accounting for differences in personal
and job characteristics (see Lucifora and Meurs, 2004, for a short review of recent
studies). However, this average premium varies considerably by earnings decile,
being considerably larger at the lower end of the earnings distribution and smaller (or
non-existent) at the higher end. The premium is also generally found to be larger for
women than for men. Lucifora and Meurs (ibid.) find similar patterns in the public-
sector pay premium in France and Germany, despite obvious differences in wage-
setting machinery, although they do find that the overall magnitude of the premium is
somewhat smaller than in Britain.
20


The states interest to maintain fair rates of pay for lower-skilled workers and the
strength of union organisation throughout the public sector as a whole no doubt play
their respective roles in reducing wage inequality, alongside the historic preference
for national settlements. A further dimension to the public-private pay differential is
that changes in the public-sector premium in various countries appear to be counter-
cyclical (Holmlund and Ohlsson, 1992; Elliott and Duffus, 1996; Brooks, 2004), which
may either reflect the fact that the sector is sheltered from changing economic

20
The non-random selection of employees into public service activities and the limited overlap
between some public and private sector activities make the estimation of the pay premium difficult.
Disney and Gosling (2003) take advantage of employees movements between the two sectors to
control for otherwise unobserved differences. Their findings suggest that, for men, any public sector
premium actually exists at the higher-end of the skill distribution, although their results for women
are in line with the stylised facts.
34
fortunes or serve as a further indication of the prominent role of trade unions.
21

Finally, it is important to recognise that all of the findings cited here relate solely to
wages. The availability of fringe benefits also tends to be favourable in the public
sector, a fact that is evident from unions recent attempts to protect pension
arrangements in sectors such as local government.

The tradition of national settlements has also sheltered pay setting in the British
public sector from local factors. The Conservative governments of the 1980s and
1990s were critical of national bargaining and sought to decentralise pay setting so
that it might take more account of local labour market conditions, a move that was
encouraged by the devolution of management responsibility for budgets. Some
change did occur, with localised payments being made possible for health service
workers and schoolteachers in order to address recruitment difficulties. But the
degree of decentralisation was limited by central government concerns over cost
control, by management and trade union resistance to radical change, and by the
general preference for national settlements among the Pay Review Bodies (Bach
and Winchester, 2003). At the end of the 1990s, regional variations in wages
remained somewhat smaller than those found among private sector employees
(Frontier Economics, 2003). However, the pressure to localise public sector pay has
recently been renewed by HM Treasury (2003) and so national settlements continue
to be under strain.

At the same time, pay negotiations have been used as a tool for restructuring in
search of efficiency gains, evident in recent attempts to reform working practices in
the Fire Service (Charlwood, 2004: 389). Attempts to introduce performance-related
pay have also been prominent in areas such as the Civil Service, teaching and
health (see, for example, Bach, 2000), thereby introducing another avenue for pay
decentralisation.
22
In either case, there is a clear motivation to do away with the
historic notion of fair comparability, whereby pay levels were to be determined

21
Research shows a similar pattern in the union wage premium, with the argument being that, in the
private sector, unions protect their members wages from downward pressure in times of recession
(Blanchflower and Bryson, 2003).
22
Within Europe, such developments have been most prominent in Britain. But Bach and Della Rocca
(2000) also document moves towards performance-related pay for public sector workers in Denmark
and Italy.
35
through direct comparison to the private sector. Instead, this has been replaced by a
closer alignment to the principle of affordability, whereby pay increases must be
earned, either as a reward for greater flexibility in working arrangements or as the
result of enhanced performance. These developments can be expected to lead to
greater variance in pay levels within the public sector.

Privatisation and contracting-out have also played their part in changing the nature of
pay-setting within public services more broadly. Kondylis and Wadsworth (2006)
note that the shrinking of the public sector has reduced its share of male
employment from one-third to one-quarter since the early 1980s, thereby contributing
to rising wage inequality. The Transfer of Undertakings Regulations (TUPE) have
limited the impact on terms and conditions for those transferring from the public to
the private sector, but the regulations have not prohibited suppliers from offering
inferior terms to new recruits (Unison, 2000). Nevertheless, in a significant victory for
the trade unions, the government has recently signaled its commitment to end this
so-called two-tier workforce. Indeed, some movement has already been made, with
the Department of Health agreeing in 2005 that ancillary staff employed by private
contactors would receive levels of pay on a par with those available to directly-
employed workers (Bewley, 2006).

The state as a pay regulator

In addition to its role as the public sector employer, the State regulates pay levels
and changes in pay in the economy at large both directly through minimum wages,
equal pay legislation, and incomes policies and so on, and indirectly through policies
such as welfare-to-work programmes and statutes affecting union organising and
union bargaining power. Policies relating to taxation, employers social security
contributions and pensions affect wage and non-wage labour costs as well as
earned income.

Among these various activities, the introduction of minimum wages has typically
attracted the most attention. The rationales espoused for minimum wages are
numerous but chief among them are the alleviation of household poverty and the
guarantee of a fair days work for a fair days pay. The former arises from a
36
concern that even if market-set wages reflect workers marginal product, these
wages may be inadequate to meet what society deems the nutritional and material
needs of a workers family. The latter implies either that workers are not being paid
their marginal product because employers possess monopsony power, or else
labour market segmentation permits wage discrimination against those with low
bargaining power or else, even if they are paid their marginal product, society
deems that wage unfair according to commonly-held precepts of fairness and
equity.

In eighteenth century England, J ustices of the Peace were empowered to fix local
minimum wages to alleviate poverty. However, concerns about State interference
with market wage setting led policy makers to support parish-funded in-work wage
supplementation in preference to minimum wages at the beginning of the 19
th

Century. The Speenhamland system, named after the place in which it was first
conceived, led to a decline in wages variously attributed to employers downwardly
adjusting wages in the knowledge that the parish would relieve ensuing poverty, an
increase in labour supply induced by higher returns to paid work, and downward
pressure on wages as unsubsidised labourers faced competition from subsidised
workers (Hobsbawm and Rude, 1993; Hammond and Hammond, 1947). Similar
concerns have been expressed in relation to modern-day wage supplements which
have burgeoned in the last few decades in the UK and the USA (Blundell and
Hoynes, 2004). Wage supplements can also limit earnings progression, trapping
employees in long-term reliance on State assistance while in work by creating
disincentives to train, seek promotion or work longer hours due to the withdrawal of
the wage supplement as earnings rise (Bryson, 1998; Blundell and Hoynes, 2004:
455-456). Others criticise wage supplements on the grounds of equity when equally
productive workers doing identical jobs take home different net income when one
qualifies for the supplement and the other does not.

If low pay results from a demand side imperfection such as monopsony then a wage
subsidy may compound those imperfections by further lowering the price of labour
below its marginal productivity. As Wilkinson notes (2001: 21): In this case, the
appropriate action is to raise wages by establishing a minimum wage or the
encouragement of the counterveiling power of trade unions. Yet there was no
37
national system of legal minimum wage setting in Britain until 1909 when Winston
Churchills Board of Trade set up Wages Councils to set pay rates in some of the
lowest paying industries. In the 1980s, concerned that above-market rates damaged
jobs growth and might even create substantial unemployment, governments sought
to abolish the Wages Councils on two occasions, only to find that, on balance,
employers felt they benefited from them since they prevented competition through
wage undercutting (Bryson, 1989). However, they were finally abolished in 1993,
leaving the Agricultural Wages Board as the only minimum wage fixing mechanism.
In 1999 a national minimum wage was introduced for the first time in British history.
It seems to have halted the decline in the wages of the lowest paid men relative to
median earners that had been taking place
23
, and increased the rate of real wage
growth among the lowest paid women (Kondylis and Wadsworth, 2006).

Evidence from other countries confirms that minimum wages compress the wage
distribution. A raft of studies in the United States suggest that the decline in the real
value of the minimum wage in the 1980s accounts for most, or at least a large part,
of the rise in wage inequality between middle and low earners for both men and
women (DiNardo et al., 1996; Fortin and Lemieux, 1997; Lee, 1999; Card and
Krueger, 1995). Card and Krueger (1995) also compared changes in the wage
distribution at the state level for 1989-1992 according to the percentage of workers
who were directly affected by the 1990 and 1991 minimum wage increases and
found that these increases measurably lowered earnings inequality. In contrast, the
value of the minimum wage rose in the 1980s and 1990s in countries such as
France, the Netherlands and Spain (Dolado et al, 1996; Abowd et al, 2000).
Increases in France's minimum wage appear to have prevented erosion in real
wages at the low end of the French wage distribution (Katz et al, 1995: Abowd et al,
2000). As noted earlier, employer responses to minimum wages are often at odds
with the expectations of neo-classical theory. This is because employer responses
are partly conditioned by norms and customs and perceptions of fairness (which are
themselves affected by minimum wage setting), by substantial indeterminacy of pay,
and by what often appear to be monopsonistic market conditions (Gilman et al.,
2002).

23
There was little restoration of wage differentials among those a little above the minimum (Dickens
and Manning, 2004).
38

There are a number of other public policies that affect the wage distribution but,
because they are difficult to pin down, require a more qualitative than quantitative
assessment of their impact. First there are prices and incomes policies. These were
central to macro-economic management of the economy in the 1970s and sought to
tackle wage-push inflation. These were used by governments of the Left and Right to
restrict wage increases through wage freezes, and ceilings to both relative and
absolute pay increases, all of which tended to compress the pay distribution
(Blackaby, 1978; Flanagan, Soskice and Ulman, 1983). Second, there was public
support for industry-level wage setting and the extension of collectively bargained
terms and conditions through provisions such as the Fair Wages Resolution and
Schedule 11 of the Employment Protection Act. Again, these contributed to relative
stability in the wage distribution (Flanagan, Soskice and Ulman, 1983). The reversal
of these policies in the 1980s through the promotion of decentralized pay bargaining,
the removal of auxiliary legislation extending collectively agreed terms, the switch
away from incomes policies towards monetary policy targets, and increased
emphasis of managements right to manage, have created conditions in which the
wage dispersion was likely to grow (Kessler and Bayliss, 1998). It appears that
moves to reduce the size of unemployment-related benefits relative to earnings, the
reduction in the number of individuals entitled to benefit, and the creation of a more
punitive regime for those refusing job offers were also motivated by a desire to
create more jobs through the lowering of wages and an increase in labour supply at
the bottom end of the labour market (Bryson and J acobs, 1992; Kessler and Bayliss,
1998: 222-223). Yet, in spite of these policies - which in many ways resembled
those pursued avidly in the United States although wage dispersion grew, real
wages did not fall. This wage-stickiness is a further indication of the limitations of
neo-classical wage theory.

The State has also played a vital role in breaking social customs determining pay for
generations. Perhaps the most notable example was the introduction of the Equal
Pay and Sex Discrimination Acts of 1975 which challenged the custom, entrenched
in many industries and occupations, of paying women less than men for work of
equal value. As noted earlier, these Acts and subsequent amendments are credited
with the closing of the gender pay gap, but not its elimination. Strong anti-
39
discriminatory policies and other policies geared to supporting women in the labour
market can also affect differentials between part-time and full-time workers. This is
the reason for the absence of a part-time pay gap among women in Norway (Hardoy
and Schne, 2006).

State policy has targeted not only pay levels and changes in pay, but also the way in
which employers pay their workers. In particular, governments have sought to
encourage the spread of performance-based pay, such as share ownership schemes
and profit-related pay, through tax breaks. Evidence from the Workplace Industrial
Relations Surveys indicates that these tax incentives were important factors
explaining the growth of such schemes in the 1980s (Bryson, 2006).


5. Conclusion


This paper revisits the issue of pay determination from a theoretical perspective and
presents evidence for and against alternative theories. This is not the traditional
approach adopted in most industrial relations texts. These tend to be strong on
description, the main thrust of which is to emphasise the centrality of institutions and
human relations in pay determination. This is valuable but, in a sense, our paper is
written as a corrective to a purely descriptive approach. This, we think, is excusable,
partly because it is difficult to know what to make of pay patterns differentials
between workers, the wage distribution, earnings progression and so on if one
does not start with testable theoretical propositions about the underlying
determinants. Without theory it is also difficult to make predictions about what might
happen in the future and the effects of policy on pay outcomes. We have also
focused on micro-theory emanating from economics and industrial relations, rather
than macro-economic and macro-political theory, although we recognise the
contribution that other theoretical frameworks can make to understanding pay
setting.

Two broad conclusions emerge from this exercise. The first, hardly surprising,
conclusion is that neo-classical economic theory of supply and demand can only
40
take us so far in understanding patterns of pay. There are clear and systematic
departures in pay differentials and pay trends from those we might expect through
the application of standard micro-economic theory, which focuses on the dominance
of market mechanisms and underplays the variance generated by institutional
settings, norms and customs, and idiosyncratic arrangements for the determination
of pay. This has been recognised by industrial relations academics for some time
(see, for example, Dunlop, 1988; Rubery, 1997; Brown et al, 2003), and labour
economists have been saying much the same thing for over half a century. As long
ago as 1950 Slichter (1950: 91) remarked: the models used in accepted wage
theory are too simple and need to be supplemented. This is hardly news.. This is
evident, not least, in the range of indeterminacy that seemingly allows similar
workers, with similar skill-sets located in the same area to be paid very different
wages. But at the same time, it is equally evident that other theoretical perspectives,
whilst each having something to offer, are inevitably limited in their ability to explain
the full variety of wage outcomes. In truth, a reliance on any one theoretical position
is likely to overstate the coherence and functionality of wages, and those working in
the field of industrial relations have long recognised that no single theory of pay
setting has an over-riding claim to virtue.

The second conclusion is that, in spite of the institutional knowledge generated by
industrial relations, the discipline has contributed comparatively little theory about
pay determination which allows the analyst to test alternative propositions regarding
pay patterns. Others have made this point before (see, for instance, Brown and
Nolan, 1988). It might be argued that, given the obvious complexity of wage
determination, gaining a greater understanding of the general pattern of wages need
not be one of the primary challenges facing industrial relations at the present time.
We disagree.



41
References


Abowd J , Kramarz F, and Margolis D (1999) High Wage Workers and High Wage
Firms, Econometrica, 67, 2: 251-333

Abowd J , Kramarz F, Lemieux T and Margolis D (2000), Minimum wages and youth
employment in France and the United States in Youth employment and
joblessness in advanced countries edited by D.G. Blanchflower and R.B.
Freeman, University of Chicago Press and NBER

Adams, J . S. (1963) Toward an Understanding of Inequity, Journal of Abnormal
Psychology, 67: 422-436

Akerlof, G. (1982) Labor Contracts as Partial Gift Exchange, Quarterly Journal of
Economics 97: 543 569

Akerlof, G. (1984) Gift Exchange and Efficiency-Wage Theory: Four Views,
American Economic Review 74: 79 83

Akerlof, G. A. and Yellen, J . L. (1986) Efficiency Wage Models of the Labour Market,
Cambridge: Cambridge University Press

Akerlof, G. and Yellen, J . (1990) The Fair Wage-Effort Hypothesis and
Unemployment, Quarterly Journal of Economics 105: 255 - 283.

Anderson T, Forth J , Metcalf H and Kirby S (2001) The Gender Pay Gap, London:
Women and Equality Unit, Cabinet Office.

Arrow K (1973) Higher education as a filter , Journal of Public Economics, 2, 3:
193-216.

Atkinson J (1984) Manpower strategies for flexible organisations, Personnel
Management, 16, 2: 32-5.

Autor, D. H., Katz, L. F. and Kearney, M. S. (2006) The Polarization of the U.S.
Labor Market, NBER Working Paper No. 11986

Autor, D. H., Levy, F. and Murnane, R. J . (2003) "The Skill Content Of Recent
Technological Change: An Empirical Exploration," The Quarterly Journal of
Economics, 118, 4: 1279-1333

Bach S (2000) Annual review article 2001. Public-sector employment relations
reform under Labour: muddling through on modernization?, British Journal of
Industrial Relations, 40, 2: 319-339.

Bach S and Della Rocca G (2000) The management strategies of public sector
employers in Europe, Industrial Relations Journal, 31, 2: 82-96.

42
Bach S and Winchester D (2003) Industrial relations in the public sector, in P
Edwards (ed.) Industrial Relations: Theory and Practice in Britain, Oxford:
Blackwell.

Bandiera, O., Barankay, I. and Rasul, I. (2005) Social Preferences and the
Response to Incentives: Evidence from Personnel Data, Quarterly Journal of
Economics, 130, 3: 917-962

Becker G (1957) The Economics of Discrimination, Chicago: University of Chicago
Press.

Becker G (1964) Human Capital, Chicago: University of Chicago Press.

Belfield, C. R. and Heywood, J . S. (2001) Unionization and the pattern of nonunion
wages: evidence for the UK, Oxford Bulletin of Economics and Statistics, 63, 5:
577-598

Bewley H (2006) Annual review article 2005. Raising the standard? The regulation
of employment, and public sector employment policy, British Journal of Industrial
Relations, 44, 2: 351-372.

Blackaby, F. (1978) Incomes Policy, in F. Blackaby (ed.) British Economic Policy
1960-1974, Cambridge University Press, Cambridge

Blackaby D, Clark K and Leslie D (1995) Dual labor markets and the potential
earnings of the unemployed, Scottish Journal of Political Economy, 42, 1: 3752.

Blanchflower, D. and Bryson, A. (2003) Changes over time in union relative wage
effects in the UK and the US Revisited, chapter 7 in International Handbook of
Trade Unions, J ohn T. Addison and Claus Schnabel (eds.), Edward Elgar,
Cheltenham England and Northampton Mass., USA

Blanchflower, D. and Bryson, A. (2004) Union Relative Wage Effects in the United
States and the United Kingdom, Proceedings of the 56
th
Annual Meeting of the
Industrial Relations Research Association, 133-140

Blanchflower, D.G. and Cubbin, J . (1986) Strike Propensities at the British
Workplace, Oxford Bulletin of Economics and Statistics, 48, 1: 19-39

Blanchflower D and Oswald A (2003) The Wage Curve, Cambridge, MA: The MIT
Press.

Blanchflower, D., Oswald, A. and Sanfey, P. (1996) Wages, Profits and Rent-
Sharing, Quarterly Journal of Economics, 111: 227-252.

Blundell, R. and Hoynes, H. (2004) Has In-Work Benefit Reform Helped the
Labour Market?, in D. Card, R. Blundell and R. B. Freeman (eds.) Seeking a
Premier Economy: The Economic Effects of British Economic Reforms, 1980-
2000, University of Chicago Press: Chicago and London

43

Booth, A. (1995) The Economics of the Trade Union, Cambridge University Press

Bratsberg, B. and Ragan, J . F. (2002) Changes in the Union Wage Premium by
Industry, Industrial and Labor Relations Review, 56, 1: 65-83

Brooks K (2004) Pay and the public service workforce, London: Institute for Public
Policy Research.

Brown D and Crossman A (2000) Employer strategies in the face of a National
Minimum Wage: an analysis of the hotel sector, Industrial Relations Journal, 31,
3: 206-219.

Brown, D. A., Gardner, J ., Oswald, A. and Qian, J . (2005) Does Wage Rank Affect
Employees Wellbeing?, IZA Discussion Paper No. 1505

Brown W, Deakin S, Hudson M, Pratten C and Ryan P (1998) The Individualisation
of Employment Contracts in Britain, London: Department of Trade and Industry,
Research Series No. 4.

Brown W, Hayles J , Hughes B and Rowe L (1980) Occupational pay structures
under different wage-fixing arrangements: a comparison of intra-occupational pay
dispersion in Australia, Britain and the United States, British Journal of Industrial
Relations, 18, 2: 217-30.

Brown,W. and Nolan, P. (1988) Wages and labour productivity, British Journal of
Industrial Relations, 26, 3: 339-361.

Bryson, A. (1989) Undervalued, Underpaid and Undercut, Low Pay Unit, London

Bryson, A. (1998) Lone Mothers Earnings, in Private Lives, Public Responses:
Lone Parenthood and Future Policy in the UK, edited by Reuben Ford and J ane
Miller, Policy Studies Institute, London

Bryson, A. (2004a) Union effects on workplace closure, 1990-1998, British Journal
of Industrial Relations, Vol. 42, 2: 283-302

Bryson, A. (2004b) Unions and Employment Growth in British Workplaces During
the 1990s: A Panel Analysis, Scottish Journal of Political Economy, Vol. 51, No.
4, 477-506

Bryson, A. (2006) WIRS Through the Ages, PSI Mimeo

Bryson, A. and J acobs, J . (1992) Policing the Workshy: Benefit Control, Labour
Markets and the Unemployed, Avebury, Aldershot

Bryson, A. and Freeman, R. (2006) Doing the Right Thing? Does Fair Share
Capitalism Improve Workplace Performance?, mimeo

44
Bryson A., Martins P., Rycx F., and Salverda W. (2006), "Rent-sharing and
Profitability", Interim Report (Task 2) for the European Commission


Budd J (2004) Non-wage forms of compensation, Journal of Labor Research, 25, 4:
597-622.

Cappelli, P. and Chauvin, K. (1991) An interplant test of the efficiency wage
hypothesis, Quarterly Journal of Economics, 106, 869-84

Card, D. and Krueger, A. (1995) Myth and Measurement: The New Economics of the
Minimum Wage, Princeton University Press: Princeton, New J ersey

Card, D., Lemieux, T. and Riddell, W. C. (2003) Unionisation and Wage Inequality: A
Comparative Study of the US, the UK and Canada, NBER Working Paper #9473

Charlwood A (2004) The New Generation of Trade Union Leaders and Prospects
for Union Revitalization, British Journal of Industrial Relations, 42, 2: 379-397.

Clarke, R. (1993) Trends in concentration in UK manufacturing, 1980-9. In M
Casson and J Creedy (eds) Industrial Concentration and Economic Inequality,
Cambridge: Edward Elgar, pp. 121-42

Constant A and Massey D (2003) Labor market segmentation and the earnings of
German guestworkers, Institute for the Study of Labor (IZA) Discussion Paper
No. 774

Cowherd, D.M. and Levine, D. I. (1992) Product quality and pay equity between
lower-level employees and top management: An investigation of distributive
justice theory, Administrative Science Quarterly, 37: 302-320

Craig C, Rubery J , Tarling R and Wilkinson F (1982) Labour Market Structure,
Industrial Organisation and Low Pay, Cambridge: Cambridge University Press.

Crompton R and Sanderson K (1990) Gendered Jobs and Social Change, London:
Unwin Hyman.

Dale-Olsen H (2006) Wages, fringe benefits and worker turnover, Labour
Economics, 13, 1: 87-105.

Daniel, W.W. and Millward, N. (1983) Workplace Industrial Relations in Britain,
London: Heinemann

Dickens, R. and Manning, A. (2004) Has the National Minimum Wage Reduced UK
Wage Inequality?, Journal of Royal Statistical Society, Series A, Vol. 167: 623-
626

DiNardo, J .E., N.M. Fortin, and T. Lemieux (1996), "Labor market institutions and the
distribution of wages, 1973-1992: a semi-parametric approach", Econometrica,
pp. 1001-1044.
45

Disney R and Gosling A (2003) A new method for estimating public sector pay
premia: evidence from Britain in the 1990s, Centre for Economic Policy
Research Discusion Paper No. 3787.

Doeringer, P. B. and M. J . Piore. (1971) Internal Labor Markets and Manpower
Analysis. Lexington, Massachusetts, Heath Lexington Books.

Dolado, J ., F. Kramarz, S. Machin, A. Manning, D. Margolis and C. Toelings (1996),
The economic impact of minimum wages in Europe, Economic Policy, 23,
October, pp. 319=372

Dube, A. and Freeman, R. (2006) Complementarity of Shared Compensation and
Decision-making Systems: Evidence from the American Labor Market, NBER
mimeo

Dunlop J (1957) The task of contemporary wage theory, in J Dunlop (ed) The
Theory of Wage Determination, London: Macmillan.

Dunlop J (1988) Labor markets and wage determination: then and now, Chapter 2
in B Kaufman (ed) How Labor Markets Work: Reflections on Theory and Practice
by John Dunlop, Clark Kerr, Richard Lester, and Lloyd Reynolds, Lexington
Books.

Elliot and Duffus (1996) What has been happening to pay in the public-service
sector of the British economy? Developments over the period 1970-1992, British
Journal of Industrial Relations, 34, 1: 51-86.

Employment Tribunal Service (2006) Annual Report and Accounts 2005-06, London:
The Stationery Office.

Engels, F. (1881) A Fair Days Wages for a Fair Days Work, The Labour Standard,
May 7
th
, http://www.marxists.org/archive/marx/works/1881/05/07.htm

Fairris D (1989) Compensating wage differentials in the union and non-union
sectors, Industrial Relations, 28, 3: 356-372.

Falk, A., Fehr, E. and Zehnder, C. (2006) The Behavioural Effects of Minimum
Wages, forthcoming in the Quarterly Journal of Economics

Fehr, E., Gchter, S., and Kirchsteiger, G. (1996) Reciprocal Fairness and
Noncompensating Wage Differentials, Journal of Institutional and Theoretical
Economics 152, 608 - 640.

Flanagan, R., Soskice, D. and Ulman, L. (1983) Unionism, Economic Stabilization
and Incomes Policies: European Experience, The Brookings Institution,
Washington DC

46
Forth J and Metcalf H (2005) Comment on How to close the gender pay gap in
Europe, Industrial Relations Journal, 36, 3:,214-216.

Forth J and Millward N (2000) The Determinants of Pay Levels and Fringe Benefit
Provision in Britain, Discussion Paper No. 171, London: National Institute of
Economic and Social Research.

Forth, J . and Millward, N. (2002) Union effects on pay levels in Britain, Labour
Economics, 9: 547-561

Fortin, N.M. and T. Lemieux (1997), "Institutional changes and rising wage
inequality: is there a linkage?", Journal of Economic Perspectives, 11, 2: 75-96.

Freeman, R. B. (1980) Unionism and the Dispersion of Wages, Industrial and Labor
Relations Review, 34, 1: 3-23

Freeman, R. B. and Gelber, A. M. (2006) Optimal Inequality: Evidence from a
Tournament, Harvard mimeo

Freeman, R. B., Kruse, D., Blasi, J . (2006) Worker responses to shirking, NBER
mimeo

Freeman, R. B. and Kleiner, M. M. (2005) The Last American Shoe Manufacturers:
Decreasing Productivity and Increasing Profits in the Shift from Piece Rates to
Continuous Flow Production, Industrial Relations, 44, 2: 307- 330

French M and Dunlap L (1998) Compensating wage differentials for job stress,
Applied Economics, 30, 8: 1067-1075.

Frontier Economics (2003) Southern discomfort: setting pay in the public sector,
Frontier Economics Bulletin, April.

Gallie, D., Felstead, A. and Green, F. (2004) Changing Patterns of Task Discretion
in Britain, Work, Employment and Society, 18, 2: 243-266

Gallie, D., White, M., Cheng, Y. and Tomlinson, M. (1998) Restructuring the
Employment Relationship, Oxford: Oxford University Press

Garen J (1988) Compensating wage differentials and the endogeneity of job
riskiness, Review of Economics and Statistics, 70, 1: 9-16.

Gibbons, R. and Katz, L. (1992) Does Unmeasured Ability Explain Inter-Industry
Wage Differentials?, Review of Economic Studies, 59, 3: 515-35

Gilman M, Edwards P, Ram M and Arrowsmith J (2002) Pay determination in small
firms in the UK: the case of the response to the National Minimum Wage,
Industrial Relations Journal, 33, 1: 52-67.

Glyn, A. (2001) Inequalities of Employment and Wages in OECD Countries, Oxford
Bulletin of Economics and Statistics, 63: 697-713
47

Goos, M. and Manning, A. (2003). Lousy and Lovely Jobs: the Rising Polarization of
Work in Britain, CEP Discussion Paper No. 604, London School of Economics

Gosling, A. and Lemieux, T., (2004), Labor Market Reforms and Changes in Wage
Inequality in the United Kingdom and the United States in D. Card, R. Blundell
and R. Freeman (eds). Seeking a Premier League Economy. The Economic
Effects of British Economic Reforms 1980-2000, University of Chicago Press.

Grant M (2005) Unchained: 10 years after Bosman, The Sunday Herald, 4
th

December.

Green F and Dickerson A (2004) The growth and valuation of computing and other
generic skills, Oxford Economic Papers, 56, 3: 371-406.

Greenberg, J . (1990) Employee theft as a reaction to underpayment or inequity: The
hidden cost of pay cuts, Journal of Applied Psychology, 75, 561-568.

Gregg P and Machin S (1994) Is the UK Rise in Inequality Different? in R Barrell
(ed.) The UK Labour Market, Cambridge: Cambridge University Press.

Grimshaw D (2000) Public sector employment, wage inequality and the gender pay
ratio in the UK International Review of Applied Economics, 14, 4: 427-448.

Gunderson M and Hyatt D (2001) Workplace risks and wages: Canadian evidence
from alternative models, Canadian Journal of Economics, 34, 2: 377-395.

Hakim C (2000) Work-Lifestyle Choices in the 21
st
Century: Preference Theory,
Oxford: Oxford University Press.

Hammond, J . L. and Hammond, B. (1947) The Village Labourer, Vols. 1 and 11,
London: Longman

Hardoy, I. and Schne, P. (2006) The Part-time Wage Gap in Norway: How Large is
it Really?, British Journal of Industrial Relations, 44, 2: 263-282

Harkness S (1996) The gender earnings gap: evidence for the UK, Fiscal Studies,
17, 2: 1-36.

Hicks J (1932) The Theory of Wages, London: Macmillan.

Hildreth, A. and Oswald, A. (1997), Rent-Sharing and Wages: Evidence from
Company and Establishment Panels, Journal of Labour Economics, 15, 2:
318-337.

Hirsch, B. T. and Schumacher, E. J . (2001) Private Sector Union Density and the
Wage Premium: Past, Present and Future. Journal of Labor Research 22: 487-
518.

48
HM Treasury (2003) Guidance Note on Progressing Local Pay, London: HM
Treasury.

Hobsbawm, E. J . and Rude, G. (1993) Captain Swing, London: Pimlico.

Hobsbawm, E. J . (1964) Custom, Wages and Workload in Nineteenth Century
Industry, Chapter 17 in Labouring Men: Studies in the History of Labour, London:
Weidenfeld and Nicolson

Holmlund B and Ohlsson H (1992) Wage linkages between private and public
sectors in Sweden, Labour, 6, 2: 3-17. Hwang H, Reed W and Hubbard C (1992)
Compensating wage differentials and unobserved productivity, Journal of
Political Economy, 100, 4: 835-858.

J ones, D.C. (1987) `The Productivity Effects of Worker Directors and Financial
Participation by Employees in the Firm: The Case of British Retail Cooperatives',
Industrial and Labor Relations Review, 41, 1: 79-92.

J oshi H and Paci P (1998) Unequal Pay for Women and Men: Evidence from the
British Birth Cohort Studies, Cambridge, MA: The MIT Press.

Kahn L (1993) "Free Agency, Long-Term Contracts and Compensation in Major
League Baseball: Estimates from Panel Data" Review of Economics and
Statistics, 75, 1: 157-64.

Katz, L. and Krueger, A. (1992) The Effect of the Minimum Wage on the Fast Food
Industry, Industrial and Labor Relations Review, 46, 1: 6-21

Katz, L., G. Loveman and D. Blanchflower (1995), A comparison of changes in the
structure of wages in four OECD countries, in Differences and changes in wage
structures edited by R.B. Freeman and L.F. Katz and published by University of
Chicago Press and NBER, pp. 25-65.

Kessler, S. and Bayliss, F. (1998) Contemporary British Industrial Relations,
Macmillan, Basingstoke

Kondylis F and Wadsworth J (2006) Wages and wage inequality since the 1970s,
Royal Holloway College, University of London, mimeo

Kostiuk P (1990) Compensating differentials for shift work, Journal of Political
Economy, 98, 5: 1055-1075.

Krueger, A. B. (1991) Ownership, Agency and Wages: An Examination of
Franchising in the Fast-Food Industry, Quarterly Journal of Economics, 106: 75-
101

Lanfranchi J , Ohlsson H and Skalli A (2002) Compensating wage differentials and
shift-work preferences, Economics Letters, 74, 3: 393-398.

49
Lazear, E. P. (1979) Why Is There Mandatory Retirement?, Journal of Political
Economy, 87, 1261-1264

Lazear, E. P. and Rosen, S. (1981) Rank-order tournaments as optimum labor
contracts Journal of Political Economy, 89, 5: 841-64

Lee, D. (1999), Wage inequality in the US during the 1980s: rising dispersion or
falling minimum wage?, Quarterly Journal of Economics, 114: 977-1023.

Lester, R. A. (1952) A range theory of wage differentials, Industrial and Labor
Relations Review, 5: 483-500

Levine, D. (1992) Can Wage Increases Pay For Themselves? Tests with a
Production Function, The Economic Journal, 102, 414: 1102-1115.

Levine, D.I., and Tyson, L.D. (1990), Participation, productivity and the firms
environment, in Blinder, A. S. (ed.), Paying for Productivity: Looking at the
Evidence, Washington D.C.: Brookings Institute

Loveridge, R. (1980) `What is Participation? A Review of the Literature and Some
Methodological Problems', British Journal of Industrial Relations, 18, 3: 297-317.

Lucas R and Langlois M (2003) Anticipating and adjusting to the introduction of the
National Minimum Wage in the hospitality and clothing industries, Policy Studies,
24, 1: 33-50.

Lucifora C and Meurs D (2004) The public sector pay gap in France, Great Britain
and Italy, IZA Discussion Paper No. 1041.

Machin, S.J . and Vignoles, A. (2005) What's the Good of Education?
The Economics of Education in the UK, Princeton, NJ : Princeton University
Press.

Mackay D, Boddy D, Brack J , Diack J and J ones N (1971) Labour Markets Under
Different Employment Conditions, London: Allen and Unwin.

Mahajan S (2005) Input-output: concentration ratios for businesses by industry in
2005, Economic Trends, 624: 52-62.

Manning A (2003) Monopsony in Motion, Princeton, NJ : Princeton University Press.

Manning A and Petrongolo B (2005) The Part-Time Pay Penalty, London: Women
and Equality Unit.

Marsden, D. (2004) 'The Role of Performance Related Pay in Renegotiating the
"Effort Bargain": The Case of the British Public Service.' Industrial and Labor
Relations Review 57, 3: 350-370.

Marx, K. (1976) Capital Volumn One, Harmondsworth, Penguin

50
McDonald, I. M. and Solow, R. M. (1981). Wage bargaining and employment.
American Economic Review, 71: 896908.

McIntosh S (2006) Further analysis of the returns to academic and vocational
qualifications, Oxford Bulletin of Economics and Statistics, 68, 2: 225-252.

McNabb R (1989) Compensating wage differentials: some evidence for Britain,
Oxford Economics Papers, 41, 1: 327-338.

Millward N, Forth J and Bryson A (2001) Who calls the Tune at Work? The Impact of
Trade Unions on Jobs and Pay, York: J oseph Rowntree Foundation

Moretti (2000) Do wages compensate for risk of unemployment? Parametric and
semi-parametric evidence from seasonal jobs, Journal of Risk and Uncertainty,
20, 1: 45-66.

Moule C (1998) The regulation of work in small firms: a view from the inside, Work,
Employment and Society, 12, 4: 635-54.

Murphy, K. M. and Topel, R. H. (1990) Efficiency Wages Reconsidered: Theory and
Evidence in Y. Weiss and G. Fishelson (eds.) Advances in the Theory and
Measurement of Unemployment, New York: Macmillan, pp. 204-240

Myck M and Paull G (2004) The role of employment experience in explaining the
gender wage gap, Institute for Fiscal Studies Working Paper 04/116.

Neal D (1995) Industry-specific human capital: evidence from displaced workers,
Journal of Labor Economics, 13 ,4: 653-77.

Neumark D, Zhang J and Ciccarella S (2005) The effects of Wal-Mart on local
labour markets, Working Paper No. 11782, National Bureau of Economic
Research.

Nolan P and Brown W (1983) Competition and workplace wage determination,
Oxford Bulletin of Economics and Statistics, 45, 3: 269-287.

Olson C (2002) Do Workers Accept Lower Wages in Exchange for Health
Benefits?, Journal of Labor Economics, 20, 2 (part 2): S91-S114.

Phillips A (1958) The relation between unemployment and the rate of change of
money wage rates in the United Kingdom, 1861-1957, Economica, 25: 283-299.

Pierce B (2001) Compensation inequality, Quarterly Journal of Economics, 116, 4:
1493-1525.

Piore M (1979) Birds of Passage: Migrant Labor in Industrial Societies, Cambridge:
Cambridge University Press.

Purse K (2004) Work-related fatality risks and neoclassical compensating wage
differentials, Cambridge Journal of Economics, 28, 4: 597-617.
51

Raff, D. and Summers, L. (1987) Did Henry Ford pay efficiency wages?, Journal of
Labor Economics, 5, 2: 557-86

Ransom M (1993) "Seniority and Monopsony in the Academic Labor Market."
American Economic Review 83, 1: 221-231.

Rosen, S. (1969) Trade Union Power, Threat Effects and the Extent of
Organization, Review of Economic Studies, 36: 185-196

Rubery, J . (1997) Wages and the Labour Market, British Journal of Industrial
Relations, 35, 3: 337-366

Salop, S. (1979) A Model of the Natural Rate of Unemployment, American
Economic Review, 69: 117-25

Salop, J . and Salop, S. (1976) Self-selection and Turnover in the Labor Market,
Quarterly Journal of Economics, 90, 4: 619-27

Shapiro, C. and Stiglitz, J . (1984) Equilibrium Unemployment as a Worker Discipline
Device, American Economic Review 74: 433 - 44.

Slichter, S. H. (1950) Notes on the Structure of Wages, Review of Economics and
Statistics, 32, 80-91

Smith A (1776) An Inquiry into the Nature and Causes of the Wealth of Nations, ???

Solow, R. (1990) The Labour Market as a Social Institution, Oxford: Blackwell

Spence M (1973) J ob market signalling, Quarterly Journal of Economics, 87, 3:
355-74.

Spitz, A. (2005) Technical Change, J ob Tasks and Rising Educational Demands:
Looking Outside the Wage Structure. Mimeo, Centre for European Economic
Research (ZEW), Mannheim

Stewart, M. B. (1987), Collective Bargaining Arrangements, Closed Shops and
Relative Pay, Economic Journal, 97 (385), 140-156.

Stewart, M. B. (1995) Union wage differentials in an era of declining unionisation,
Oxford Bulletin of Economics and Statistics, 57: 143-66

Sullivan D (1989) Monopsony Power in the Market for Nurses." Journal of Law and
Economics 32, no. 2 part 2 (1989): S135-S178.

Swaffield J (2000) Gender, motivation, experience and wages, Centre for
Economic Performance, London School of Economics, Discussion Paper No.
457.

52
Theodossiou I (1995) Wage Determination for Career and Non-Career Workers in
the UK: Is There Labour Market Segmentation? Economica, 62: 195-211.

Van Reenen, J . (1996), The Creation and Capture of Rents: Wages and Innovation
in a Panel of UK Companies, Quarterly Journal of Economics, 111 (1), pp. 195-
226.

Viscusi W (1978) Wealth effects and earnings premiums for job hazards, Review of
Economics and Statistics, 60, 3: 408-416.

Wade, J . B., OReilly, C. A., and Pollock, T. G. (2006) Overpaid CEOs and
Underpaid Managers: Fairness and Executive Compensation, Organization
Science, forthcoming

Wadhwani, S. B. and Wall, M. (1991) A Direct Test of the Efficiency Wage Model
using UK Micro-Data, Oxford Economic Papers, 43, 4: 529-548

Weiss, A. (1980) J ob Queues and Layoffs in Labor Markets with Flexible Wages,
Journal of Political Economy, 88: 526-8

Weitzman, M. L. and Kruse, D. L. (1990) Profit sharing and productivity in A. S.
Blinder (ed.) Paying for Productivity. A Look at the Evidence, Washington DC:
Brookings Institute

White M and Forth J (1998) Pathways through Unemployment: The Effects of a
Flexible Labour Market, York: J oseph Rowntree Foundation.

Wilkinson, F. (2001) The Theory and Practice of Wage Subsidisation: Some
Historical Reflections, Centre for Business Research Discussion Paper No. 201,
University of Cambridge




53
Alex Bryson and John Forth, November 2006
For more information on the Manpower Human Resources Laboratory, please call
020 7955 7285, or email j.m.cantlay@lse.ac.uk
Manpower Human Resources Laboratory
Centre for Economic Performance
London School of Economics
Houghton Street
London WC2A 2AE
Web: www.lse.ac.uk/manpower