Mcnabb 2007

This article was downloaded by: [New York University]
On: 28 May 2015, At: 02:21

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
The International Journal of Human Resource

Management
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/rijh20
The impact of varying types of performance-related

pay and employee participation on earnings
Robert McNabb & Keith Whitfield
Published online: 22 Jun 2007.
To cite this article: Robert McNabb & Keith Whitfield (2007) The impact of varying types of performance-related pay and
employee participation on earnings, The International Journal of Human Resource Management, 18:6, 1004-1025, DOI:
10.1080/09585190701321542
To link to this article: http://dx.doi.org/10.1080/09585190701321542
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
Int. J. of Human Resource Management 18:6 June 2007 1004 –1025
The impact of varying types of

performance-related pay and employee
participation on earnings
The International Journal of Human Resource Management 2007.18:1004-1025.
Robert McNabb and Keith Whitfield
Abstract Using matched employer-employee data from a nationally representative

sample of British establishments, the paper examines the impact of different types of
performance-related pay (PRP) on earnings and whether this is moderated by the presence
of employee involvement and financial participation schemes. It supports previous work
that has suggested that there are strong and complex interactions between employee
involvement schemes and compensation mechanisms. In particular, the results indicate
that employees’ earnings are highest where both individual and work-group PRP schemes
are present at a workplace, and where there are congruent PRP and employee participation
schemes.
Keywords Performance-related pay; employee participation; employee earnings.
Introduction
In recent years, performance-related pay (PRP) has been much more widely used as part
of the human resource policies of organizations. It has typically been introduced
alongside other HR practices as part of a package of measures, the aim of which has been
to increase the input (both quantitative and qualitative) of workers to the production
process (Whitfield and Poole, 1997).
Previous work has indicated that there is a positive relationship between PRP and
earnings (Booth and Frank, 1999), and the suggestion is that this is due to the greater
work effort and/or ability of those on incentive pay (Barkume, 2004). The tendency in
such studies has been to see PRP as both relatively homogeneous and limited in its
design. Typically, it is viewed as synonymous with individual PRP, especially piecework
or commission pay (see, e.g., Lazear, 2000; Seiler, 1984). However, PRP, as currently
practiced, is much broader than this, and many of its forms involve group- or team-based
compensation, which are potentially very different in orientation and design from the
more individualistic forms, especially piecework. Most empirical studies ignore this, and
either use a catch-all definition of PRP (e.g., Booth and Frank, 1999), or combine
individual and group forms together (Brown, 1992).
A further complication is that some workplaces adopt more than one type of PRP.
The impact of PRP on earnings could, therefore, be different in those with just one type
of scheme rather than a number. Moreover, these differences could involve interactions
Robert McNabb, Cardiff Business School, Aberconway Building, Cardiff University, Cardiff CF10
3EU, UK (e-mail: mcnabb@cardiff.ac.uk); Keith Whitfield, Cardiff Business School, Aberconway
Building, Cardiff University, Cardiff CF10 3EU, UK (e-mail: whitfield@cardiff.ac.uk).
The International Journal of Human Resource Management
ISSN 0958-5192 print/ISSN 1466-4399 online q 2007 Taylor & Francis
http://www.tandf.co.uk/journals
DOI: 10.1080/09585190701321542
McNabb and Whitfield: The impact of varying types of . . . on earnings 1005
that make the PRP –earnings relationship vary from the situation when each is used in
isolation. Such interactions could be either negative, whereby one type of scheme blunts
the impact of another, or positive, whereby differing types of scheme are mutually
reinforcing.
There is also a tendency in the existing literature to view PRP in a vacuum,
independently of other aspects of the human resource management and production
system of which it is a part. This is despite a growing acknowledgement that PRP is an
integral part of holistic approaches to management that are aimed at improving
organizational performance. Of particular note are the relationships between PRP and
attempts by firms to devolve decision-making power to workers (employee participation
schemes), and those that tie some aspect of remuneration to the firm’s performance
(financial participation schemes). To date, these interrelationships have not been directly
investigated. It is possible, for example, that individual forms of PRP work best where
employee decision-making processes are narrowly focused on jobs, whereas group-based
forms of PRP work best in environments where workers are involved in making decisions
about their work-roles in their broader context (see, e.g., Delbridge and Whitfield, 2001).
The objective of this paper is to examine both the impact of different types of PRP on
earnings, and whether these relationships are affected by the use of employee
involvement and financial participation schemes. It extends previous analysis in two
significant respects. First, it addresses the question of whether the earnings effects of PRP
schemes are sensitive to the particular measure of PRP used. Second, it provides
evidence as to whether these are conditional on the use of other forms of work practices
and/or financial participation schemes.
PRP, employee participation and earnings

PRP, participation and performance
The development of PRP schemes has generally been viewed as a response by employers
to the principal –agent problem inherent in the employment relationship. Many workers
have a wide range of discretion in how they undertake their work. They can vary their
effort, their cooperation with colleagues and management, and their contribution to
improving the effectiveness of both their and related inputs. In such a situation,
mechanisms are needed to ensure that workers (the agents) undertake their work in a
manner that is congruent with the interests of their employers (the principals).
The principal– agent problem can be most directly addressed by the close monitoring
of workers’ performance and the disciplining of workers deemed to be shirking.
However, this can prove costly, not only because of the need to pay supervisors, but also
because the explicit demonstration by employers of a lack of trust in their workers
can encourage behaviour based on narrow opportunistic motives (Drago and Perlman,
1989).
An alternative is to design incentive schemes to reward individual or group output.
Such schemes have the advantage to the firm in that they not only induce greater effort
from workers, but they may also encourage the commitment of workers to the goals and
values of the organization. Linking pay to performance also potentially introduces equity
and consistency in the pay structure and enables the firm to attract more able workers
(Holmstrom, 1982; Lazear, 2000).
The standard models, therefore, view PRP as a technique that can monitor workers’
contribution to the production process, thereby more closely aligning workers’ pay and
productivity (Lazear, 2000). Some also view it as a mechanism for increasing worker
effort (Brown, 1990). The key problem herein is seen to be the cost of monitoring
1006 The International Journal of Human Resource Management
performance. Where this is less than the gain from aligning pay and productivity and/or
eliciting greater effort, the firm will introduce a PRP scheme; otherwise it will not.
Such models take a very narrow view of PRP and its potential role in improving firm
performance. In particular, they focus on individual forms of PRP and thereby ignore the
role of group PRP and the potential for PRP to improve the performance of teams
rather than individuals. Indeed, they also ignore a possible inherent contradiction with
individual PRP schemes, that they may improve individual performance, but reduce that
of the work-group as a whole, by giving a disincentive for workers to work cooperatively
with others.
Furthermore, such models also ignore that PRP is only one type of payment system
aimed at increasing the contribution of employees to the production process. Also
important are schemes for sharing profits and employee share-ownership. These could be
seen as broad forms of group PRP, though their link to performance is much less direct
than those group schemes that are typically given this title. Such schemes, being potential
substitutes for the more non-individualistic types of PRP, are likely to reduce the impact
of group PRP on earnings.
The development of work systems offering workers a greater role in decision-making
and in increasing functional flexibility in recent years has focused enhanced attention on
the need to develop new pay systems. Innovations such as quality circles, briefing groups,
total quality management and team-working have resulted in a marked change in the way
in which work is done (Appelbaum and Batt, 1994). They involve workers taking more
responsibility for key aspects of the production process and, in particular, being
more responsive to product market pressures. Such changes not only empower workers in
key areas, but also intensify the potential for principal –agent divergences. There is, for
example, evidence that workplace changes aimed at increasing employee participation in
decision-making are typically only successful in raising organizational performance
where they are associated with compatible gain-sharing mechanisms (Levine and Tyson,
1990; McNabb and Whitfield, 1998). It is, therefore, likely that the presence or absence
of such schemes at a workplace will have an impact on the PRP–earnings relationship,
a factor ignored in existing models.
PRP, participation and earnings

Once the complexities of the nature of PRP and the context within which they are
situated are acknowledged, the possible impacts of PRP, particularly group PRP, on
employee earnings are not as clear-cut as the standard models suggest. Most of the
analyses based on this model suggest that workers covered by (individual) PRP will be
paid more than those who are not. This is seen to reflect that more able workers chose to
work within PRP regimes and work harder within this regime. However, they ignore
potential variations for differing types of PRP and the presence or absence of employee
and financial participation schemes.
In order to be effective, PRP needs to offer workers a significant return on their higher
effort and commitment that would be missed in its absence. In short, workplaces are
prepared to incur higher wage costs for PRP workers in the hope of gaining a greater than
commensurate increase in performance. Furthermore, to the extent that PRP yields
higher work effort and commitment, it might be expected that PRP workplaces would
show better financial performance. This, in turn, would allow such workplaces to pay its
employees more than those not operating PRP.
Alternatively, it may be that PRP operates simply as a distributive mechanism,
rewarding the high-achieving more and the low-achieving correspondingly less than
in non-PRP workplaces. In such a situation, the overall average effect for the two types of
workplaces will not be significantly different. However, previous empirical work has
suggested that PRP does more than simply re-distribute rewards.
Therefore, assuming that workers prefer less work effort to more, and that PRP evokes
greater worker inputs into the production process, the former postulate can be expected to
predominate. This yields our first hypothesis:
Hypothesis 1: Earnings will be higher, all other things equal, in workplaces operating
PRP schemes than in workplaces not operating such a scheme.
The relative contributions of individual, group and organizational PRP schemes to

earnings are less clear a priori. To some extent, they depend on the importance attached
to each by managers in the workplaces concerned. The more that they are wedded to a
‘hard’ HRM approach based on individual incentives, the greater the contribution to
earnings of individual PRP schemes can be expected to be. Conversely, the more the
emphasis is on the ‘softer’ HRM approach of building team-working to raise
performance, the more important will be the contribution of group schemes.
The role of group and organizational PRP in enhancing performance is likely to be
different from that of individual PRP. In parallel to individual PRP, the objective of the
scheme will be to ensure that output is both aligned to its payment and that those covered
by the scheme produce a high level of input into the production process. However, in
contrast to individual PRP, gains may be made by reducing the importance attached to
individual incentives, which may encourage behaviour that works against the interests of
the team as a whole.
In recent years, a large number of organizations have recognized the need to
encourage teamwork in the search for competitive advantage, and have put in place a
range of practices with this in mind. Allied to this has been an increased interest in
group forms of PRP. The introduction of such schemes, however, is not unproblematic.
The two main problems are the measurement of team output and free riding. In parallel
with the situation for individual PRP, where team output cannot be easily measured, it
can be expected that there will be less pay that is determined by group PRP. A great deal
also depends on the potential magnitude of the free rider problem, whereby members of
the team rely on others to make the extra effort to increase group performance related
pay. The greater the concern about some workers within teams free-riding, thereby
reducing the total team input into the production process, the less will pay be linked to
group PRP.
The widespread concerns about free riding and the problems associated with
measuring team output, yield hypothesis 2:
Hypothesis 2: Group PRP schemes have made a lesser contribution to earnings than
individual schemes.
It is, however, possible that firms have over-ridden such concerns in the search for a
compensation system that can support their attempts to capture team efficiencies.
The tying of PRP schemes to attempts to increase employee participation in decision-
making raises the further issue of whether such efforts have been accompanied by greater
increases in earnings for the workers concerned as opposed to where they have been
introduced in a more stand-alone way. One might expect the return to be greater where
there is such a linkage for two main reasons. First, one might expect that the increase in
earnings in such integrated situations would be greater, given that it suggests that the
intent to transform the organization is much stronger. Second, such a situation might
be expected to yield a greater return and, therefore, more leeway for firms to reward their
employees.
Additionally, there is the issue of whether this varies for different types of scheme.
A priori, one would expect some degree of congruence between the nature of the
participation scheme and the type of PRP scheme yielding the greatest increase in
earnings. More job-focused participation schemes, such as quality circles, can be
expected to be more strongly linked to individual PRP schemes and broader forms of
participation, such as briefing groups, to group PRP schemes. This gives us Hypothesis 3:
Hypothesis 3: The increase in earnings due to PRP will be greater where there are
attempts to increase the participation of employees in decision-making.
Following on from this, hypothesis 4 is that:
Hypothesis 4: The magnitude of such an increase will be greater the more the
congruence between the participation scheme and the PRP scheme.
The presence or absence of financial participation schemes is also likely to affect the
increase in earnings due to a PRP scheme. Most likely, this will reduce the increase in
earnings, due to the two types of scheme being substitutes. This is especially true for
organizational/workplace-based schemes, which are not dissimilar in design from many
financial participation schemes. Hence hypothesis 5 states that:
Hypothesis 5: The increase in earnings related to PRP schemes will be smaller where
there are no financial participation schemes in place, and that this
downward impact will be greatest for organizational/workplace
schemes than for individual and group schemes.
Estimation
To measure the single and joint impacts of employee involvement and performance-
related pay schemes on individual earnings, the following model is estimated. This takes
the form,
log wij ¼ a0 þ X ij b þ a1 EPj þ a2 PRPj þ a3 Z j þ 1ij ð1Þ
where, EPj measures whether different employee participation schemes and work
practices are used at the establishment and PRPj measures whether the establishment has
adopted individual, group or an establishment level performance related pay scheme, Zi
is a vector of establishment characteristics and Xij is a vector of employee characteristics.
Here wij is worker i’s hourly wage at the jth establishment. This is derived by dividing
data on each person’s weekly wage by the number of hours worked per week. While
the data on hours worked contained in the 1998 Workplace Employee Relations Survey
(WERS98) is the actual number of hours usually worked per week, the information for
weekly earnings is grouped into 12 bands. To construct an hourly wage, the lower and
upper bounds of the pay band in which an individual is located are divided by that
person’s hours of work. Following Stewart (1983, 1987), the method of interval
regression is then used to derive consistent parameter estimates for Equation (1).
Because the data used in the analysis is multi-level, the error term can be written as,
1ij ¼ wij þ uj ð2Þ
where wij represents that part of the error term that varies independently across
individuals both within and between establishments and uj measures that part that varies
across establishments but which is constant for workers within establishments. In short,
there is a possibility of establishment-specific fixed effects yielding inefficient estimates
of parameter values. The error structure describes a random effects model and the
efficient estimator is feasible generalized least squares.
Five equations are estimated. The first includes a variable indicating whether a
workplace has a PRP scheme in place; the second sub-divides this variable into
individual, group and establishment/organization PRP and the third focuses on the
relative importance of having single types of PRP scheme in place or combinations
thereof; the fourth and fifth equations examine interactions of the PRP variables with
employee participation variables and are based on the definitions of PRP in Equations (1)
and (2) respectively.
It is possible that the relationship between PRP and earnings is more complex than this
model implies. In particular, there may be endogeneity resulting from the selection
(particularly self-selection) of more productive workers into workplaces operating PRP
schemes. To some extent, this is mitigated by the use of such a comprehensive set of
controls for employee, workplace and organizational variables, although even these
(most notably the human capital variables) may also be endogenous to the use of PRP.
However, work by Booth and Frank (1999) would suggest that this is not a serious
problem. In so far as there might be unmeasured factors that influence the decision to
work in a PRP workplace, these are likely to inflate the coefficient estimates on the PRP
variables, especially that for individual PRP.
Data
The analysis uses the 1998 Workplace Employee Relations Survey (WERS98). It differs
from its predecessors in collecting matched employer–employee information. It has
comprehensive information on a nationally representative sample of 2,191 establish-
ments employing ten or more employees and on 28,323 employees who work in these
establishments. (Cully et al., 1998). The data on employees include weekly pay,
occupational group, qualifications, marital status, age, gender, union membership, hours
worked and tenure. Data on the establishments include information on their structural
characteristics, management employment practices, product markets, labour force
composition and the nature of their collective employment relations.
A major advantage of using the WERS data for the current study is that it is based on
information that relates both to employees and establishments, rather than just the latter.
This has two important advantages. The first is that it mitigates the endogeneity problems
that have bedeviled much work on PRP (Booth and Frank, 1999). Second, it allows the
deployment of a wide range of both employee and establishment controls for factors
affecting earnings other than PRP.
Performance-related pay
Performance-related pay schemes are measured in two ways. The first is based on
whether employees at the establishment receive payments from individual or group PRP
schemes. Three measures of performance are used: those based on individual output or
performance; those based on group or team performance; and those that tie performance-
related payments to the performance of the establishment or organization. Since many
establishments use a combination of measures, a number of interactions between the
schemes are also considered. Specifically, the combination of schemes that reward
individual performance with schemes based on the other two measures of performance
are examined.
Defining employee participation

In this paper, we define three groups of employee participation:
1 Quality circles – groups at the workplace that solve specific problems or discuss
aspects of performance or quality and which involve more than 60 per cent of non-
managerial employees.
2 Briefing groups – a system of briefings for any section or sections of the workforce,
the meetings of which give over 25 per cent or more of their time to questions from
employees or for employees to offer their views.
3 Representative participation – committees of managers and employees primarily
concerned with consultation rather than negotiation.
Following Delbridge and Whitfield (2001), the first can be regarded as a scheme focused
at the job level, whereas the second and third are broader schemes allowing workers a say
over work organization extending beyond their immediate job.
Defining financial participation

Two groups of financial participation scheme are defined:
. Profit-related pay – employees receive payments or dividends from either profit-

related pay or deferred profit sharing scheme.
. Employee share ownership – employees receive payments from an employee share
ownership scheme.
Control variables
Clearly, employee earnings are influenced by a wide range of variables other than the
presence of PRP and employee participation schemes. To control for these, an
augmented human capital equation is estimated, including variables relating to the
employee, his or her job and the establishment in which he/she works. These are listed in
Appendix 2.
Results
The results which are summarized in column one of Table 1 provide corroboration for the
findings of earlier studies that have reported a positive impact of individual PRP on
earnings and thereby support Hypothesis 1. The magnitude of the positive impact of PRP
on earnings is approximately 7 per cent – which is in line with previous estimates (Booth
and Frank, 1999). This should, however, be placed in the context of the returns to other
employee and workplace characteristics. For example, the return to holding a degree is
more than four times larger than that to working in a PRP workplace. It is almost double,
however, the return to working in a unionized workplace.
The results in column two of Table 1 also indicate that workers in establishments that
use group- or team-based PRP and those that link earnings to individual performance
earn more, other things equal. Moreover, this relationship is very similar for the two
types of scheme, thereby questioning Hypothesis 2, which suggested that the increase in
earnings would be less for those in workplaces with group-based schemes than for those
Table 1 Impact of performance related pay, employee and financial participation on individual earnings
General PRP variable Specific PRP variables PRP workplace regimes
McNabb and Whitfield: The impact of varying types of . . . on earnings

Constant 1.674 (74.93) 1.676 (75.86) 1.681 (74.57)
Gender 0.153 (20.63) 0.147 (20.33) 0.156 (20.71)
Age
25– 9 0.173 (13.84) 0.167 (13.86) 0.176 (13.85)
30– 9 0.273 (21.86) 0.268 (22.28) 0.276 (21.76)
40– 9 0.299 (21.58) 0.296 (22.17) 0.300 (21.30)

50– 9 0.296 (20.00) 0.294 (20.68) 0.295 (19.66)
60 or over 0.23 (10.11) 0.230 (10.53) 0.229 (9.92)
Degree 0.304 (25.67) 0.298 (26.05) 0.307 (25.53)
A-level or equivalent 0.178 (15.05) 0.174 (15.33) 0.179 (14.90)
O-level or equivalent 0.116 (11.45) 0.114 (11.70) 0.117 (11.39)
CSE or equivalent 0.038 (3.22) 0.040 (3.52) 0.038 (3.16)
Vocational qualification 2 0.026 (2 4.00) 2 0.024 (23.73) 20.027 (24.09)
Ethnic group
Black 2 0.086 (2 2.95) 2 0.093 (23.30) 20.083 (22.81)
Asian 2 0.094 (2 3.59) 2 0.098 (23.88) 20.088 (23.28)
Other 2 0.008 (2 0.24) 2 0.015 (20.48) 20.007 (20.21)
Long-standing health problem or disability 2 0.037 (2 2.76) 2 0.036 (22.81) 20.039 (22.83)
Marital status
Single 2 0.08 (28.96) 2 0.076 (28.87) 20.082 (29.02)
Divorced 2 0.043 (2 3.57) 2 0.039 (23.36) 20.045 (23.66)
Widowed 2 0.099 (2 2.89) 2 0.092 (22.80) 20.098 (22.82)
Children
Aged 0 –4 0.038 (3.85) 0.039 (4.12) 0.037 (3.75)
Aged 5 –11 0.006 (0.69) 0.007 (0.79) 0.006 (0.63)
Aged 12– 18 0.001 (0.13) 0.002 (0.26) 0.001 (0.05)
Current job tenure
Less than 1 year 2 0.117 (2 10.50) 2 0.117 (210.87) 20.116 (210.28)
1 year to less than 2 years 2 0.119 (2 10.07) 2 0.117 (210.27) 20.119 (29.93)
1011
2 years to less than 5 years 2 0.064 (2 6.83) 2 0.062 (26.93) 20.063 (26.62)
Table 1 (Continued)

5 years to less than 10 years 2 0.039 (2 4.40) 2 0.037 (2 4.33) 2 0.041 (24.51)
Occupation
Technical 2 0.139 (2 12.26) 2 0.141 (2 12.87) 2 0.137 (211.85)
Clerical/secretarial 2 0.278 (2 25.93) 2 0.284 (2 27.36) 2 0.274 (225.22)
Craft related 2 0.334 (2 27.55) 2 0.332 (2 28.26) 2 0.334 (227.24)
Protective/personal service 2 0.325 (2 18.43) 2 0.333 (2 18.95) 2 0.325 (218.31)

Sales 2 0.355 (2 22.74) 2 0.353 (2 23.24) 2 0.356 (222.49)
Plant/machine operatives 2 0.504 (2 42.00) 2 0.497 (2 42.46) 2 0.506 (241.65)
Routine unskilled 2 0.543 (2 35.63) 2 0.537 (2 36.39) 2 0.541 (234.96)
Union member 0.06 (7.27) 0.057 (7.10) 0.062 (7.44)
Type of job
Temporary 2 0.107 (2 4.42) 2 0.103 (2 4.41) 2 0.107 (24.37)
Fixed term 2 0.123 (2 5.59) 2 0.120 (2 5.64) 2 0.128 (25.75)
Size of organization
51– 200 employees 0.028 (2.61) 0.028 (2.52) 0.022 (2.07)
201–10,000 employees 0.062 (7.51) 0.063 (7.40) 0.062 (7.56)
Establishment is more than 20 years old 2 0.027 (2 3.97) 2 0.026 (2 3.75) 2 0.026 (23.82)
Market grew in last year 0.019 (2.94) 0.019 (2.84) 0.017 (2.67)
Union presence 0.046 (5.30) 0.048 (5.49) 0.041 (4.81)
Percentage of workforce who are routine unskilled workers 2 0.205 (2 11.44) 2 0.209 (2 11.43) 2 0.213 (211.81)
Market share greater than 25% 0.019 (2.84) 0.022 (3.16) 0.016 (2.31)
Briefing groups 2 0.002 (2 0.29) 2 0.002 (2 0.27) 2 0.001 (20.17)
Joint consultation committees 0.02 (3.03) 0.021 (2.98) 0.018 (2.73)
Quality circles 2 0.014 (2 1.61) 2 0.012 (2 1.28) 2 0.012 (21.38)
Employee share-ownership scheme 0.03 (3.62) 0.030 (3.48) 0.031 (3.78)
Profit-related payments 0.016 (2.25) 0.019 (2.52) 0.022 (3.15)
Performance-related payments (PRP) 0.073 (9.98)
Individual PRP 0.041 (3.58)
Workgroup PRP 0.047 (3.59)
Table 1 (Continued)

Establishment/organization PRP 0.011 (0.98)
Individual PRP only 0.032 (2.25)
Workgroup PRP only 0.078 (2.76)
Establishment/organization PRP only 0.033 (2.18)
Individual and workgroup PRP only 0.119 (7.84)
Workgroup and establishment/ organization PRP only 0.071 (4.00)
Individual, workgroup and establishment/organization PRP only 0.118 (2.72)
N 9970 9970 9970
Log likelihood function 2 18210.845 218003.464 218288.816
1013
with individual schemes. This would suggest that firms have not been reluctant to
develop more costly group-based PRP schemes because of problems measuring
group output and the potential for free-riding. Linking pay to broad measures of
performance at the establishment or organization level does not have direct impact on
employee pay (see column two).
Earnings are highest in those establishments where there is a combination of
individual and workgroup-based PRP schemes and that the combined effect of these is
greater than their single effects, see column three of Table 1. The largest premia
associated with incentive pay schemes are for employees in establishments that adopt
just group or team-based schemes and in those that combine these with incentive
schemes that reward individual performance. When all the combinations of the
different schemes are distinguished, earnings premia are also found in establishments
that operate establishment/organization-level PRP schemes. A key finding here is that,
without exception, workgroup PRP offers higher earnings premia than the other types.
In short, it pays to work in workplaces that have PRP, especially those with group forms.
The results summarized in Table 1 also indicate that workers who are able to
participate in employee share ownership or profit related pay schemes and those in
establishments where there are joint consultative schemes have higher earnings, other
things equal. However, there is no evidence that employee participation per se leads to
compensation gains for workers, which is broadly in line with results reported by
Osterman (2000) for the US. In fact, there is some evidence that employees in
establishments that have adopted quality circles actually face lower earnings, on average.
The results summarized in Table 2 also provide a number of further insights into the
complex way compensation systems, employment practices and financial participation
interact. Employee participation appears to confer little advantage for workers even
when tied to PRP. However, there is evidence that the use of briefing groups does lead to
higher earnings where they are associated with workgroup-based compensation systems,
providing some support for Hypothesis 4, which states that earnings will be higher where
workplaces introduce congruent PRP and EP schemes. In contrast, there is little gain in
the form of higher earnings for employees in establishments that have introduced briefing
groups in conjunction with individual PRP, whereas briefing groups used in conjunction
with a reward system that is based on establishment or organizational performance has a
clear negative impact on employee earnings.
Earnings are found to be lower in establishments that have introduced quality circles.
However, where these are used alongside an appropriate incentive mechanism,
employees share in the mutual gains associated with productivity and quality
improvement. Interestingly, the important link is between quality circles and pay linked
to individual, rather than group, performance. Employees do not benefit from working in
quality circles where their pay is more loosely tied to broader measures of output. This
may suggest that the use of quality circles only gives rise to organizational improvements
in which workers can benefit where there is a very clear link between the compensation
gains they face and the use of quality circles.
In short, there is only mixed support for Hypothesis 3. The increase in earnings due to
PRP does not seem to be significantly higher where there is employee participation in
place, but it does seem that the increase in earnings is greater where there is congruence
between PRP and the participation scheme.
The results summarized in Table 2 also highlight a complicated pattern of interactions
between PRP and financial participation. Profit-related schemes are associated with
higher employee earnings when introduced either in isolation or in conjunction with a
pay scheme tied to establishment or organizational performance. Employees gain when
Table 2 Interaction effects
Interactions with general PRP variable Interactions with specific PRP variables

Constant 1.677 (74.48) 1.676 (73.77)
Gender 0.153 (20.57) 0.154 (20.52)
Age
25– 9 0.173 (13.80) 0.172 (13.59)
30– 9 0.272 (21.74) 0.271 (21.44)
40– 9 0.298 (21.53) 0.297 (21.19)

50– 9 0.295 (19.96) 0.295 (19.68)
60 or over 0.229 (10.10) 0.227 (9.86)
Qualifications
Degree 0.303 (25.62) 0.306 (25.51)
A-level or equivalent 0.178 (15.05) 0.179 (14.92)
O-level or equivalent 0.115 (11.41) 0.115 (11.22)
CSE or equivalent 0.038 (3.20) 0.036 (3.02)
Vocational qualification 20.026 (23.86) 2 0.026 (2 3.89)
Ethnic group
Black 20.086 (22.95) 2 0.091 (2 3.09)
Asian 20.095 (23.63) 2 0.092 (2 3.44)
Other 20.009 (20.28) 2 0.015 (2 0.45)
Long-standing health problem or disability 20.037 (22.75) 2 0.036 (2 2.67)
Marital status
Single 20.08 (29.02) 2 0.082 (2 9.05)
Divorced 20.044 (23.62) 2 0.044 (2 3.57)
Widowed 20.101 (22.93) 2 0.099 (2 2.85)
Children
Aged 0 –4 0.038 (3.87) 0.039 (3.91)
Aged 5 –11 0.007 (0.79) 0.007 (0.73)
Aged 12– 18 0.001 (0.14) 0.001 (0.16)
Current job tenure
Less than 1 year 20.118 (210.60) 2 0.122 (2 10.84)
1015
1 year to less than 2 years 20.12 (210.18) 2 0.125 (2 10.47)
Table 2 (Continued)
2 years to less than 5 years 20.065 (26.96) 20.068 (27.18)
5 years to less than 10 years 20.04 (2 4.46) 20.042 (24.64)
Occupation
Technical 20.14 (2 12.31) 20.137 (211.91)
Clerical/secretarial 20.279 (226.05) 20.277 (225.47)
Craft related 20.334 (227.58) 20.335 (227.31)

Protective/personal service 20.325 (218.38) 20.322 (218.22)
Sales 20.356 (222.80) 20.356 (222.56)
Plant/machine operatives 20.504 (241.92) 20.504 (241.56)
Routine unskilled 20.541 (235.51) 20.546 (235.34)
Union member 0.061 (7.38) 0.061 (7.32)
Job type
Temporary 20.107 (24.42) 20.109 (24.47)
Size of organization 20.125 (25.66) 20.124 (25.57)
51 –200 0.027 (2.48) 0.032 (2.97)
201– 10,000 employees 0.061 (7.49) 0.068 (8.25)
Establishment is more than 20 years old 20.028 (24.05) 20.025 (23.76)
Market grew in last year 0.017 (2.56) 0.017 (2.62)
Union presence 0.046 (5.29) 0.044 (5.13)
Percentage of workforce who are routine unskilled workers 20.206 (211.43) 20.211 (211.72)
Market share greater than 25% 0.02 (2.92) 0.023 (3.32)
Briefing groups 20.004 (20.52) 20.005 (20.66)
Joint consultation committees 0.029 (3.69) 0.030 (4.01)
Quality circles 20.031 (22.94) 20.024 (22.37)
Employee share ownership scheme 0.013 (1.29) 0.019 (1.94)
Profit-related payments 0.025 (2.96) 0.021 (2.66)
Performance-related pay (PRP) scheme 0.078 (5.23)
PRP and briefing groups 0.01 (0.66)
PRP and joint consultation committees 20.021 (21.46)
Table 2 (Continued)
PRP and quality circles 0.058 (2.99)
PRP and employee share-ownership schemes 0.047 (2.82)

PRP and profit-related pay 20.038 (22.31)
Individual PRP 0.017 (0.75)
Workgroup PRP 0.051 (1.98)
Establishment/organization PRP 0.038 (1.70)
Individual PRP and briefing groups 0.038 (1.61)
Individual PRP and joint consultation committees 20.024 (21.06)
Individual PRP and quality circles 0.133 (4.25)
Individual PRP and employee share ownership schemes 0.087 (3.22)
Individual PRP and profit-related pay 20.059 (22.09)
Workgroup PRP and briefing groups 0.077 (2.82)
Workgroup PRP and joint consultation committees 0.011 (0.41)
Workgroup PRP and quality circles 20.047 (21.25)
Workgroup PRP and employee share ownership schemes 20.088 (22.80)
Workgroup PRP and profit-related pay 0.002 (0.07)
Establishment/organization PRP and briefing groups 20.111 (25.04)
Establishment/organization PRP and joint consultation 20.022 (21.03)
Establishment/organization PRP and quality circles 20.078 (22.88)
Establishment/organization PRP and employee share ownership 0.025 (1.08)
Establishment/organization PRP and profit-related pay 0.046 (1.85)
N 9970 9970
Log likelihood function 218201.794 218248.118
1017
the establishment/organization is more profitable and this is especially marked where
there is a clear link between earnings and organizational performance.
There is also evidence that employee share ownership schemes provide an incentive
effect that leads to higher employee compensation. This is reinforced if workers are paid
according to an individual PRP scheme. However, employees do not share in any gains
that arise from employee share ownership if their earnings are also linked to
group performance incentive pay. The support for Hypothesis 5 is, therefore, also mixed.
The increase in earnings is greater where individual PRP is in place, but not where
group PRP is in place.
Conclusions
The main conclusion of this study is that workplaces that adopt any type of PRP pay
higher levels of earnings than those that do not. This impact is particularly large where
workplaces have both individual and group forms of PRP. It is least where workplaces
have only individual or establishment/organizational PRP. The impact of PRP on
earnings is also strongly affected by the presence of financial participation and employee
participation schemes. In particular, pay is lower where PRP co-exists with a profit-
related pay scheme and higher where there is employee share-ownership. The presence
of an employee participation scheme also influences the PRP– pay relationship. This is
highest when there is a congruent form of PRP and EP in place, for example, individual
PRP and quality circles, and workgroup PRP and briefing groups.
The results indicate that PRP schemes are costly for employers. They may
help increase the input of workers, but whether this yields better financial performance
depends on the relative magnitudes of the costs and benefits.
The main implication for researchers is that great care must be taken in modelling the
interactions between PRP and earnings so as to avoid possible mis-specification and
thereby the drawing of invalid inferences about the links between them. In particular,
researchers must take account of the heterogeneity of PRP schemes and the contexts with
in which they are situated.
The main limitation of the analysis is that the PRP variable relates to the workplace
rather than the worker. Therefore, our analysis is strictly one of whether those working in
workplaces with PRP schemes earn more than those in workplaces that do not, rather
than whether those working under PRP earn more. This clearly lends a degree of
imprecision to the analysis.
Appendix 1 Table of means

Mean
Hourly rate lower 1.79
Hourly rate upper 2.19
Gender 0.64
Age
25—9 0.16
30—9 0.29
40—9 0.24
50—9 0.17
60 or over 0.03
Appendix 1 (Continued)
Mean
Qualifications
Degree 0.23
A-level or equivalent 0.16
O-level or equivalent 0.27
CSE or equivalent 0.12
Vocational qualification 0.41

Ethnic group
Black 0.01
Asian 0.01
Other 0.01
Long-standing health problem or disability 0.06
Marital status
Single 0.24
Divorced 0.07
Widowed 0.01
Children
Aged 0 –4 0.14
Aged 5 –11 0.17
Aged 12– 18 0.18
Current job tenure
Less than one year 0.15
1 year to less than 2 years 0.11
2 years to less than 5 years 0.22
5 years to less than 10 years 0.22
Occupation
Technical 0.11
Clerical/secretarial 0.19
Craft related 0.12
Protective/personal service 0.04
Sales 0.05
Plant/machine operatives 0.15
Routine unskilled 0.07
Union member 0.39
Job type
Temporary 0.02
Fixed term 0.02
Size of organization
51– 200 employees 0.27
201– 10,000 employees 0.47
Establishment is more than 20 years old 0.48
Market grew in last year 0.44
Union presence 0.55
Percentage of workforce who are routine unskilled workers 0.10
Market share greater than 25% 0.34
Briefing groups 0.37
Joint consultation committees 0.46
Quality circles 0.16
Employee share-ownership scheme 0.28
Profit-related payments 0.51
Performance related pay (PRP) scheme 0.27
Mean
Individual PRP only 0.05
Workgroup PRP only 0.01
Establishment/organization PRP only 0.04
Individual and workgroup PRP only 0.05
Workgroup and establishment/organization PRP only 0.03
Individual, workgroup and establishment/organization PRP only 0.01

PRP and briefing groups 0.10
PRP and joint consultation committees 0.13
PRP and quality circles 0.05
PRP and employee share-ownership schemes 0.1
PRP and profit-related pay 0.18
Individual PRP 0.2
Workgroup PRP 0.14
Establishment/organization PRP 0.16
Individual PRP and briefing groups 0.08
Individual PRP and joint consultation committees 0.09
Individual PRP and quality circles 0.04
Individual PRP and employee share-ownership schemes 0.08
Individual PRP and profit-related pay 0.14
Workgroup PRP and briefing groups 0.05
Workgroup PRP and joint consultation committees 0.07
Workgroup PRP and quality circles 0.02
Workgroup PRP and employee share-ownership schemes 0.05
Workgroup PRP and profit-related pay 0.09
Establishment/organization PRP and briefing groups 0.06
Establishment/organization PRP and joint consultation 0.07
Establishment/organization PRP and quality circles 0.03
Establishment/organization PRP and employee share ownership 0.05
Establishment/organization PRP and profit-related pay 0.11
Appendix 2 Definition of variables
Employee variables

Gender ‘1’ if the individual is male, ‘0’ if the individual is female
Current job tenure
Less than 1 year. ‘1’ if the employee had been employed in their current job for less than 1 year, ‘0’ otherwise
1 year to less than 2 years ‘1’ if the employee had been employed in their current job for more than 1 year but less than 2 years, ‘0’
otherwise
2 years to less than 5 years ‘1’ if the employee had been employed in their current job for more than 2 years but less than 5 years,
‘0’ otherwise
5 years to less than 10 years ‘1’ if the employee had been employed in their current job for more than 5 years but less than 10 years,
‘0’ otherwise
If all the variables above take the value ‘0’, the employee has been employed for more than 10 years
Type of job
Temporary ‘1’ if the employee’s job was temporary, ‘0’ otherwise
Fixed-term ‘1’ if the employee was employed for a fixed-term, ‘0’ otherwise
If both these variables above take the value ‘0’, the individual was employed on a permanent basis
Age
20 –4 ‘1’ if the employee was aged between 20 and 24, ‘0’ otherwise
60 or over ‘1’ if the employee was aged over 60, ‘0’ otherwise
If all the age variables take the value ‘0’, then the employee was below the age of 20
Marital status
Single ‘1’ if the employee was single, ‘0’ otherwise
Divorced ‘1’ if the employee was divorced, ‘0’ otherwise
Widowed ‘1’ if the employee was widowed, ‘0’ otherwise
1021
Employee variables
If all the marital status variables take the value ‘0’, then the employee was living with spouse or partner
Children
Aged 0 –4 ‘1’ if the employee had dependent children aged between 0 and 4, ‘0’ otherwise
Aged 5 –11 ‘1’ if the employee had dependent children aged between 5 and 11, ‘0’ otherwise
Aged 12– 18 ‘1’ if the employee had dependent children aged between 12 and 18, ‘0’ otherwise
If all the children variables take the value ‘0’, then the employee has no children
Qualifications
Degree ‘1’ if the highest educational level attained by the employee was a degree or equivalent, ‘0’ otherwise
A-level or equivalent ‘1’ if the highest educational level attained by the employee was ‘A’ level or equivalent, ‘0’ otherwise
O-level or equivalent ‘1’ if the highest educational level attained by the employee was ‘O’ level or equivalent/GCSE (grades A-C),
‘0’ otherwise
CSE or equivalent ‘1’ if the highest educational level attained by the employee was CSE level or equivalent/GCSE (grades D-G),
‘0’ otherwise
If all the qualification variables take the value ‘0’, the employee has not attained any educational qualification
Vocational qualification ‘1’ if the employee has a attained a vocational qualification, ‘0’ otherwise
Long-standing health ‘1’ if the employee has a long standing health problem
problem or disability or disability, ‘0’ otherwise
Union member ‘1’ if the employee was member of a trade union, ‘0’, otherwise
Ethnic group
Black ‘1’ if the employee considers that they belong to the black ethnic groups, ‘0’ otherwise
Asian ‘1’ if the employee considers that they belong to the Asian ethnic group, ‘0’ otherwise
Establishment variables
Log of employment size The logarithm of the total number of employees on the payroll of the establishment
Organization employs The number of people employed by the establishment
Less than 50 employees ‘1’ if the establishment employs less than fifty people, ‘0’ otherwise. (Omitted category)
51–200 ‘1’ if the establishment employs between 51 and 200 people, ‘0’ otherwise
Employee variables

201– 500 ‘1’ if the establishment employs between 201 and 500 people, ‘0’ otherwise
10,001 or more ‘1’ if the establishment employs more than 10,000 people, ‘0’ otherwise
Percentage of workforce: The percentage of the people on the payroll described as being in each of the respective occupational groups
Managers and administrators. (Omitted category)

Professional
Technical
Clerical and secretarial
Craft and skilled services
Protective and personal services
Sales
Operative and assembly
Routine unskilled
Union presence ‘1’ if at least one union present at the establishment; ‘0’ otherwise
Foreign ‘1’ if the establishment was predominantly or completely foreign owned’ 51% or more, ‘0’ otherwise
No/few competitors ‘1’ if the establishment had few or no competitors in the market for its main product or service, ‘0’ otherwise
High degree of competition ‘1’ if the establishment assesses the degree of competition in the product market for its main product or service,
‘0’ otherwise
National/International Product Market ‘1’ if the market for the establishment’s main product was national market or international, ‘0’ otherwise
Share of UK product market ‘1’ if the establishment’s share of the UK market for its main product is greater than 26%, ‘0’ otherwise
Market growing ‘1’ if the market for the main product or service of the establishment is growing, ‘0’ otherwise
Sales grew ‘1’ if the value of sales for the main product or service of the establishment grew in the last year, ‘0’ otherwise
Establishment is more than 20 years old ‘1’ if the establishment is more than 20 years old, ‘0’ otherwise
Employee participation variables
Quality circles ‘1’ if the establishment has groups that solve specific problems or discuss aspects of performance or quality and
more than 60% of non-managerial employees are involved, ‘0’ otherwise
Briefing groups ‘1’ if the establishment has a system of briefings for any section or sections of the workforce and more than 25%
of the time at meetings is given over to questions from employees, or for employees to offer their views
1023
Employee variables
Consultation committees ‘1’ if the establishment has any committees of managers and employees primarily concerned with consultation
rather than negotiation, ‘0’ otherwise
Performance related pay variables

Employee share-ownership schemes ‘1’ if the establishment has an employee share-ownership scheme, ‘0’ otherwise
Profit-related payments ‘1’ if the establishment has a profit-related payment scheme, ‘0’ otherwise
Performance-related pay ‘1’ if the establishment has a performance-related payment, ‘0’otherwise

Individual PRP ‘1’ if the establishment has individual performance measures only,’0’otherwise
Workgroup PRP ‘1’ if the establishment has workgroup performance measures,’0’otherwise
Establishment/organization PRP ‘1’ if the establishment has organization and/or establishment performance measures, ‘0’ otherwise
Individual and workgroup PRP ‘1’ if the establishment has individual and workplace performance measures but not establishment/organization;
‘0’ otherwise
Individual and establishment/organization ‘1’ if the establishment has individual and establishment/organization performance measures but not workgroup,
PRP ‘0’ otherwise
Workgroup and establishment/organization ‘1’ if the establishment has workgroup and establishment/organization performance measures but not individual,
PRP ‘0’ otherwise
Individual, workgroup and establishment/ ‘1’ if the establishment has individual, workgroup and establishment/organization performance measures,
organization PRP ‘0’ otherwise
References
Appelbaum, E. and Batt, R. (1994) The New American Workplace. Cornell, Ithaca, NY: ILR Press.
Barkume, A. (2004) ‘Using Incentive Pay and Providing Pay Supplements in US Job Markets’,
Industrial Relations, 43: 618 –33.
Booth, A.L. and Frank, J. (1999) ‘Earnings, Productivity and Performance Related Pay’,
Journal of Labor Economics, 17(3): 447 – 64.
Brown, C. (1990) ‘Firms’ Choice of Method of Pay’, Industrial and Labour Relations Review,
43 (Special Issue): 165S –82S.
Brown, C. (1992) ‘Wage Levels and Methods of Pay’, Rand Journal of Economics, 23: 366 –75.
Cully, M., O’Reilly, A., Millward, N., Forth, J., Woodland, S., Dix, G. and Bryson, A. (1998)
The 1998 Workplace Employee Relations Survey: First Findings. HMSO.
Delbridge, R. and Whitfield, K. (2001) ‘Employee Perceptions of Job Influence and Organizational
Participation’, Industrial Relations, 40: 472 – 89.
Drago, R. and Perlman, R. (1989) ‘Supervision and High Wages as Competing Incentives: A Basis
for Labour Segmentation Theory’. In Drago, R. and Perlman, R. (eds) Microeconomic Issues in
Labour Economics: New Approaches. New York: Harvester Wheatsheaf.
Holmstrom, B. (1982) ‘Moral Hazard in Teams’, Bell Journal of Economics, 13: 324– 40.
Lazear, E.P. (1989) ‘Pay Equality and Industrial Politics’, Journal of Political Economy,
97: 561–80.
Lazear, E.P. (2000) ‘Performance Pay and Productivity’, American Economic Review, 90:
1346 – 91.
Levine, D. and Tyson, L. (1990) ‘Participation, Productivity and the Firm’s Environment’.
In Blinder, A. (ed.) Paying for Productivity. Washington, DC: Brookings Institute.
McNabb, R. and Whitfield, K. (1998) ‘The Impact of Financial Participation and Employee
Involvement on Financial Performance’, Scottish Journal of Political Economy, 45: 171– 88.
Osterman, P. (2000) ‘Work Reorganization in an Era of Restructuring: Trends in Diffusion and
Effects on Employee Welfare’, Industrial and Labor Relations Review, 53(2): 179– 96.
Seiler, E. (1984) ‘Piece Rate versus Time Rate: The Effect of Incentives on Earnings’, Review of
Economics and Statistics, 66: 363 –76.
Stewart, M. (1983) ‘On Least Squares Estimation When the Dependent Variable is Grouped’,
Review of Economic Studies, 50(4): 737 – 53.
Stewart, M. (1987) ‘Collective Bargaining Arrangements, Closed Shops and Relative Pay’,
Economic Journal, 97(1): 140 – 56.
Whitfield, K. and Poole, M. (1997) ‘Organizing Employment for High Performance’, Organization
Studies, 18(5): 745 –64.

Mcnabb 2007

Uploaded by

Copyright:

Available Formats

Mcnabb 2007

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mcnabb 2007

Uploaded by

Copyright:

Available Formats

This article was downloaded by: [New York University]

On: 28 May 2015, At: 02:21

The International Journal of Human Resource

The impact of varying types of performance-related

To link to this article: http://dx.doi.org/10.1080/09585190701321542

PLEASE SCROLL DOWN FOR ARTICLE

The impact of varying types of

Robert McNabb and Keith Whitfield

Abstract Using matched employer-employee data from a nationally representative

Keywords Performance-related pay; employee participation; employee earnings.

PRP, employee participation and earnings

PRP, participation and earnings

PRP schemes than in workplaces not operating such a scheme.

The relative contributions of individual, group and organizational PRP schemes to

Following on from this, hypothesis 4 is that:

Defining employee participation

Defining financial participation

. Profit-related pay – employees receive payments or dividends from either profit-

McNabb and Whitfield: The impact of varying types of . . . on earnings

40– 9 0.299 (21.58) 0.296 (22.17) 0.300 (21.30)

General PRP variable Specific PRP variables PRP workplace regimes

Protective/personal service 2 0.325 (2 18.43) 2 0.333 (2 18.95) 2 0.325 (218.31)

General PRP variable Specific PRP variables PRP workplace regimes

McNabb and Whitfield: The impact of varying types of . . . on earnings

40– 9 0.298 (21.53) 0.297 (21.19)

Craft related 20.334 (227.58) 20.335 (227.31)

PRP and employee share-ownership schemes 0.047 (2.82)

group PRP is in place.

Appendix 1 Table of means

Vocational qualification 0.41

Individual, workgroup and establishment/organization PRP only 0.01

McNabb and Whitfield: The impact of varying types of . . . on earnings

McNabb and Whitfield: The impact of varying types of . . . on earnings

Managers and administrators. (Omitted category)

Performance related pay variables

Performance-related pay ‘1’ if the establishment has a performance-related payment, ‘0’otherwise

You might also like