Leeway for the Loyal: A Model of Employee Discretion
| Author | Francis Green |
| DOI | http://doi.org/10.1111/j.1467-8543.2007.00666.x |
| Date | 01 March 2008 |
| Published date | 01 March 2008 |
Leeway for the Loyal: A Model of
Employee Discretion
Francis Green
Abstract
This article examines the factors underlying task discretion from an economist’s
perspective. It argues that the key axis for understanding discretion is the
trade-off between the positive effects of discretion on potential output per
employee and the negative effects of greater leeway on work effort. In empirical
analysis using matched employer–employee data, it is shown that discretion is
strongly affected by the level of employee commitment. In addition, discretion is
generally greater in high-skilled jobs, although not without exceptions, and
lower where employees are under-skilled. Homeworking and flexitime policies
raise employee discretion. The impact of teamworking is mixed. In about half of
cases team members do not jointly decide about work matters, and the net effect
of teams on task discretion in these cases is negative. In other cases, where team
members do decide matters jointly, the impact is found to be neutral according
to employees’ perceptions, or positive according to managers’ perceptions.
There are also significant and substantial unobserved establishment-level
factors which affect task discretion.
1. Introduction
This article examines the design of jobs from an economic perspective, with
a focus on employees’ autonomy. While worker autonomy has been central
both to sociology’s class-based analyses of work (Braverman 1974), and to
psychology’s demand-control model of job satisfaction and stress (Karasek
1979), neoclassical economics traditionally placed no emphasis on the
subject. Economics preferred to leave issues of job design and work organi-
zation to the ‘black box’ of technical processes taking place inside the work-
place. Yet, in recent decades, the extent to which employees are free to decide
matters concerning their job tasks has come to be seen as important in several
strands of modern economic theory, which sees the worker–boss relationship
as an example of a principal–agent problem. The location of influence over
Francis Green is at the Department of Economics, Keynes College, University of Kent.
British Journal of Industrial Relations doi: 10.1111/j.1467-8543.2007.00666.x
46:1 March 2008 0007–1080 pp. 1–32
© Blackwell Publishing Ltd/London School of Economics 2008. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
production decisions is an issue for the design of optimal incentive structures
(e.g. Athey and Roberts 2001) or for the distribution of power (e.g. Guy and
Skott 2005), depending on one’s perspective; it is an implicit or explicit
ingredient in efficiency-wage models. Workplace autonomy is also a highly
valued job feature in itself, being a major factor underpinning workers’
intrinsic job satisfaction as proposed both in an earlier literature on job
design (e.g. Cooper 1973; Davis 1966) and in more recent studies of job
satisfaction (Green and Tsitsianis 2005; Harley 2001). There is, moreover,
indirect evidence of a high demand for autonomy, in observed studies of the
demand for self-employment status. Although the latter is constrained by
credit and cost restrictions, the desire for autonomy is frequently given as a
major reason for wanting self-employment (Clinton et al. 2006; Taylor 1996).
The particular aspect of autonomy upon which this article concentrates is
the level of personal influence or discretion over the tasks that employees do
in their own jobs, referred to as ‘task discretion’. This aspect is distinct from
any influence that might be afforded to the teams in which employees might
work, and from broader forms of participation in organization-wide decision
making, whether through works councils, trade unions or other communi-
cation channels.
The need for an understanding of task discretion is heightened by the
discovery that its average level fell considerably in Britain throughout the
1990s (and was probably declining since at least the mid-1980s). By contrast,
participation in decision making at establishment level, which is affected by
different factors and is arguably less important than task discretion for
employee well-being, appears to have grown somewhat more widespread
among British establishments over this period (Felstead et al. 2002: 67–8;
Gallie et al. 2004; Millward et al. 2000). The fall in task discretion in Britain
in the 1990s hardly conforms to the paradigm of post-Fordism and has yet
to be adequately explained. The fall contrasts with modest rises in task
autonomy in Finland and in Norway during both the 1980s and 1990s (Lehto
and Sutela 1999, 2005; Leiulfsrud and Dahl 2005), while in Sweden there were
increases from at least 1975 until the 1990s, at the end of which some declines
of decision latitude are reported (Theorell 2004; Vogel and Theorell 2006).
While comparisons of autonomy levels between nations are subject to pos-
sible biases, in 2006, workers in the UK are reported to have an average
amount of autonomy, according to the Fourth European Working Conditions
Survey, while all the Scandinavian countries rank much more highly (Euro-
pean Foundation for the Improvement of Living and Working Conditions
2007).
Task discretion is only implicitly incorporated within economists’ empiri-
cal investigations of efficiency wage models, which have focused on the
behavioural implications of the principal–agent nexus for productivity,
wages or work effort. Authors in this tradition have only rarely attempted to
weave direct measures of workplace autonomy into their tests of model
predictions. Bryson and Freeman (2007) is a rare exception, in which it is
found that the impact on productivity of certain combinations of ‘fair share
2British Journal of Industrial Relations
© Blackwell Publishing Ltd/London School of Economics 2008.
capitalism’ is somewhat larger in workplaces that devolve greater autonomy
in decision making to their employees. Empirical research on job design, and
attendant studies of autonomy in jobs, have been carried out largely on
small-scale samples or in single case studies, by occupational psychologists
with rather different objectives to those of economists. Consequently, rela-
tively little is known about the characteristics of firms, jobs and workers that
are associated with high or low levels of autonomy — factors which might be
thought relevant to a better understanding of the formation of labour con-
tracts. Moreover, the determination of autonomy has only rarely been con-
sidered in the context of a profit-maximizing model of the firm.
What, then, determines the amount of autonomy afforded to workers? The
answer, I propose in this article, is derived from post-Fordism, and is linked
to the related issue of workers’ commitment to the organization. I use the
term ‘post-Fordism’ here as a shorthand for the proposition that, in contrast
to the technologies and power structures of earlier generations, prevailing
modern technologies are such that relatively dense local decision making (by
workers themselves, individually or in teams) is efficient up to a point: it
therefore profits employers to allow employees some leeway in their jobs. But
employers also face the fundamental post-Fordist dilemma that leeway
permits self-interested employees to take advantage and work less hard.1
Autonomy can be more effectively granted, therefore, to those workers who
are less likely to behave opportunistically — that is, those whose preferences
are to identify with the company’s objectives, share its values and to show
loyalty. Leeway is thus for the loyal, not for a purely self-interested homo
economicus. The centrality of a key role for affective commitment is this
article’s core hypothesis.
Recent large-scale empirical studies of worker autonomy have examined
the roles of collective bargaining, skill, technology, work organization
and sector. Contrary to the expectations of high-performance management
theorists, Harley (2001) found that team production techniques had little or
no effect on personal discretion. Batt (2004) reports US evidence that self-
managed teams enhance discretion somewhat for workers, but decrease it for
supervisors. Gallie et al. (2004) also report a mixed impact of teamworking,
but they also found that neither technological/organizational nor composi-
tional changes in Britain’s labour market account for observed movements in
task discretion. They speculate that the decline may have been linked to a
growing culture of management by target-setting, and closer monitoring of
work as a response to greater international competition or to an increasingly
controlling regulatory framework. None of these studies have attempted to
bring out a key role for commitment in determining job design.
This article proposes several advances in our understanding of task discre-
tion. As a preliminary, the determination of discretion is first set in a simpli-
fied theoretical framework designed to bring out the role of affective
commitment alongside skill. To set the empirical scene, it sheds some pre-
liminary light on whether the decline in British workers’ task discretion in the
1990s has continued in the present decade. I then make use of matched
Leeway for the Loyal 3
© Blackwell Publishing Ltd/London School of Economics 2008.
To continue reading
Request your trial