1

HR discretion: understanding line managers’ role in

Human Resource Management
Juan López-Cotarelo, Industrial Relations Research Unit, Warwick Business School

Introduction
Line managers play a central role in human resource management. In many
organisations, they are charged with myriad HR-related tasks, such as filling out
performance appraisal forms, interviewing candidates for employment, making salary
increase recommendations and breaking employment-related news –good and bad- to
employees. This paper investigates how managers carry out their HR activities.
In the literature so far, the prevailing view has been that managers act primarily on behalf
of the organisation, applying HR policy in ways by-and-large consistent with
organisational procedures. My findings from a number of store visits at a leading UK
fashion retailer, give support to a different view: that managers carry out HR activities as
organisational actors who pursue a combination of organisational, departmental and
individual goals.
Crucial in my discussion is the concept of HR discretion. I define HR discretion as a
manager’s capacity to influence his (her) team members’ HR outcomes, and their
perceptions of those HR outcomes. HR discretion must be understood as a subset of
overall managerial discretion 1 . Managerial discretion has been an important concept in
the economics and management literatures, from very different perspectives. In
economics, managerial discretion refers to the freedom managers have to pursue their
own objectives (pay, power, status and prestige) rather than shareholder interests: high
discretion means managers are able to pursue personal objectives without being caught or
punished (Williamson, 1963). In the management literature by contrast managerial
discretion refers to the choice of actions available to managers in the pursuit of
organisational goals: high discretion means managers have low environmental and
organisational constraints on their choice of actions. (Hambrick and Finkelstein, 1987).
Shen and Cho (2005) usefully integrated both points of view into a single framework
which differentiates two aspects of managerial discretion: latitude of objectives and
latitude of actions. Latitude of objectives refers to a manager’s freedom to pursue their
own objectives rather than those of other stakeholders. Latitude of actions, on the other
hand, refers to differing ranges of actions available for managers to pursue organisational
objectives. These two aspects are useful in framing my discussion below.

1
Although virtually all discussion of managerial discretion has been limited to top management –mostly
CEO- actions, the concept is clearly applicable to all levels of management in organisations (Scott et al.,
2009).
 2

Line managers in the HRM literature

Treatment of line managers in the human resource management literature has been at best
patchy. The ‘functional’ or ‘micro’ HRM subfield (Wright and Boswell, 2002) has
produced knowledge about the role of line managers in the separate HR processes, such
as personnel selection and performance appraisals. Most of the work in this subfield
however has focused on describing the various ways in which managers can be subject to
biases in their decision making. For instance, the personnel selection literature has
shown that the behaviour of interviewers influences the performance of applicants (Liden
et al., 1993), and that interviewer similarity and affect towards the interviewee is linked
to perceived job suitability of the applicant (Howard and Ferris, 1996). Likewise, the
performance appraisal literature has devoted much effort to determining the effects of
rater affect and rater similarity on performance evaluations (Levy and Williams, 2004)
On the ‘macro’ or ‘strategic’ side of the literature (Wright and Boswell, 2002), research
has almost universally espoused a research design which has limited attention to line
manager actions. Virtually all empirical studies on the relationship between HR and
organisational performance have looked at establishment-level data on the use of HR
policies (Boselie et al., 2005). This dominant approach assumes uniformity in the use of
policy within organisations, and neglects any differences arising from the actions of line
managers. A notable exception here is Bartel’s (2004) study of a retail bank. Her
quantitative analysis measured the impact of differences in employee satisfaction with
three areas of HR activity (skills, communication and rewards) on branch performance.
Through further qualitative research at a number of the branches, she was then able to
attribute these differences in employee satisfaction to differences in policy
implementation by branch managers. For instance, she described how a manager
enhanced communication within her branch through holding regular staff meetings, a
practice that was absent at another branch where communications seemed to be an issue.
In her analysis Bartel concluded that “because branch managers had considerable
discretion in managing their operations and employees, the HR environment could vary
greatly across branches and over time” (2004: 181, emphasis added).
Beyond this work, some qualitative research, mostly UK based, has examined the role of
managers in HRM. Early contributions were part of the wider critical debates on the
applicability of HRM to the British context (Guest, 1987), its differentiation from
personnel management (Legge, 1989, Legge, 1995) and its implications for industrial
relations and the labour movement (Guest, 1989, Martínez Lucio and Weston, 1992).
One stream within this debate centred on the issue of ‘HR devolution’ to the line
(Brewster and Soderstrom, 1994, McGovern et al., 1997). In a representative study,
McGovern and colleagues (1997) analysed interview and survey data from seven UK
organisations and found that there was a wide variation within organisations in the
consistency of implementation and the quality of HR practices, due to line manager
involvement. The prototypical manager was described as low skilled and likely to
identify with workers’ interests rather than the objectives of top management, thus
unwilling and unprepared for their enhanced role in HRM, compared with ‘traditional’
personnel management. Within the debates of its time, this finding was portrayed as a
serious impediment for the success of HRM in the British context.
 3

A second group of contributions aimed to provide support to the critique of human

resource management as managerial rhetoric (Legge, 1995). In a case study of the British
subsidiary of an American multinational, Truss (2001) found “a strong disconnect
between the ‘rhetoric’ of human resource management […], and the ‘reality’ experienced
by employees”. Although she mostly attributed the gap to pressure from the external
context and the inertia of the informal organisation, she also acknowledged “the role of
each individual manager as agent, choosing to focus his or her attention in varying ways.”
(Truss, 2001: 1145).
These debates have by now run their course (Legge, 2005: 2-3). Arguably, their biggest
legacy in the current literature is the distinction between intended and actual HR
practices, first proposed by Truss and Gratton (1994) as an extension of Mintzberg’s
(1978) intended and realised business strategy. This distinction remains a key element
in recent models of the HR-performance causal link (Nishii and Wright, 2008, Purcell
and Kinnie, 2007) in which the gap between intended and actual HR practices is mostly
attributed to line managers’ actions. Managers are characterised as implementers,
enactors or translators of HR policy, but what exactly is meant by that remains unclear.
Often authors seem to refer to managers’ inconsistent approach to corporate procedures.
For instance, managers at HP were found to pre-write performance appraisals, rather than
draft them jointly with the employee over the appraisal interview, as company policy
stated (Truss, 2001: 1137). Similarly, managers at Selfridges conducted appraisals with
less frequency than required by HR policy (Purcell and Hutchinson, 2007). These are
instances of managers following procedures that are different from those set by policy
makers, with negative implications for employee satisfaction with HRM. In a recent
theoretical work, Nishii and Wright (2008) take a similar view when they note that
managers influence employee perceptions of HR practices by adapting policies to the
context of their subunit. They also include, however, the possibility that managers
influence perceptions of HR through their leadership style, personality and behaviour;
and by communicating their own perceptions of HR practices to employees. Overall,
however, the implication is a negative connotation of line manager actions in the sense
that they distort the uniform application of policy by creating “slippage” (Boxall and
Purcell, 2003: 220) or undesired variability (Nishii and Wright, 2008)

Research setting and methods

Setting. The organisation I studied was a large UK-based fashion retailer, with several
hundred stores nationwide. The company were leaders in the mass mid-market segment
and faced stiff competition from rival retailers. The time of this research (spring of 2010)
was dominated by the aftermath of the 2008 financial crisis, and the ensuing recession in
the British economy. Although fashion retail as a whole continued to grow through the
recession, midmarket players suffered relatively more (Mintel, 2010), limiting sales
growth for the focal company. Ahead of the recession, management responded by taking
a number of measures to cut cost, including lower inventory levels and lower levels of
staffing (company annual reports).
I selected this type of organisation because I expected a retail environment to offer a
number of advantages for my research. First, retail networks consist of a number of
fairly similar and comparable business units, the stores. The role of store and other
 4

managers was standard across all units. Second, the same HR policies were applicable to
all the stores. Finally, in retail organisations, Head Office often exerts a tight control on
activities in the stores. This is important because many retail networks thrive from
multiple replication of a successful template designed at the centre. In this case, the
organisation’s Head of HR commented: “Our managers do not have any discretion
whatsoever; they basically do what we tell them to do”. This eventually proved not to be
necessarily the case, as shown and discussed below.
Data gathering. I collected data through semi-structured interviews with 21 respondents
in six stores in the Midlands region of England. An HR officer arranged for my visit to
each of the stores on separate days. Stores ranged in size from around 70 employees to
more than 150. The scale of some of these stores made them quite complex operations.
In-store activities were grouped under three main areas: commercial, operational and
administrative. Commercial staff focused on providing high quality customer service,
maintaining the commerciality 2 of the shop floor, and protecting against theft. On the
operational side, stockroom staff handled stock from delivery to shop floor, including
major routines around processing of deliveries and replenishment, often constrained by
lack of physical space and tight schedules. Finally, the administrative team carried out a
wide range of important control activities relating to the management of cash, tracking of
stock and prevention of employee fraud, among other responsibilities.
My point of contact at the stores was the store manager. After interviewing him or her, I
asked to interview three further members of staff 3 . Respondents were selected by the
store manager, depending on their availability. I ensured that each respondent gave
consent to participating in the research and explained that responses were confidential.
All 21 respondents held managerial positions in the stores. Their job titles included Store
Manager (5), Assistant Store Manager (4), Commercial Manager (1), Sales Manager (7),
Sales Supervisor (1), Office Manager (2), and Stockroom Supervisor (1). 16 respondents
(76%) were female, 5 respondents were male. Median length of service was 6 years,
ranging from 1 to 21. 12 respondents (57%) had been hired in their teens or early
twenties into part time sales consultant positions; 5 respondents had been recruited
externally directly into a managerial position.
My goal in designing the interviews was to inquire about managers’ role in HRM: what
policies they were involved in, what specific activities they carried out within those
policies, and how HR decisions were made. I designed two sets of interview questions;
one for store managers and another for the other respondents. Store manager questions
were grouped under three headings: ‘About your prior experience’, ‘About your current
job’ –including questions on store characteristics-, and ‘About your role managing
employees’. I designed these interviews to last about 45 minutes. Respondents in other

2
Dealing with the optimal display of merchandise on the shop floor, in terms of arrangement, presentation
and stock levels.
3
There are only 5 store managers in my sample because at one of the stores the position was vacant, and
the assistant store manager was acting as store manager. At one of the stores I could only interview the
store manager and her assistant. At another, after a second visit, I could only interview the store manager
and two of her sales managers.
 5

jobs received the following three sets of questions: ‘About your job’, ‘About your
manager’, and ‘About working for [the company]’. I designed these interviews to last
around 20 minutes. Actual interviews lasted between 25 minutes and 1 hour for store
managers (median 43 minutes), and between 15 and 45 minutes for other respondents
(median 26 minutes). All the respondents gave consent to record the interviews, so I
could produce a total of 11 hours of audio material.
Data analysis. All interviews were transcribed verbatim and then coded using a
qualitative data-analysis software (NVivo 8). My initial coding strategy focused on
tagging material according to the main HR activities identified by respondents, including
staffing structure, management of man-hours, salary, bonus, personnel selection, shifts,
absence, time keeping, performance appraisal, training, employee consultation, sales
competitions and disciplining. It was soon apparent that responses had concentrated on
some of these activities more than others.
On a second stage of analysis, I developed my coding in two ways. First, I allowed
further detailed coding within each of the initial categories to emerge. For instance,
material that I had initially coded under ‘time keeping’, was further refined into
‘evaluation of policy’, ‘implementation issues’, ‘new system’, and ‘old system’.
Secondly, I connected activities together and eventually grouped some of them under
umbrella HR activities. For example, respondent accounts of ‘Absence’ and ‘Time
keeping’ were so closely related to performance management processes that I grouped
them as special cases of the latter. Through this process two major HR activities
emerged, namely progression and performance management.
In a final third stage of my analysis, I sought to identify narratives of actual past episodes
of progression and performance management, as distinct from general accounts and
evaluations of policies and practices. This allowed me to better substantiate and illustrate
some of my conclusions.

Findings
The company had a set of well developed HR policies, which were widely known,
respected and adhered to by respondents. I found a high degree of consistency between
different respondents’ accounts of HR policies, procedures and activities. In general,
managers viewed the HR department as a value-adding partner to the business. HR was
praised for their support to decision making in the store, especially in terms of the
relevance and user-friendliness of HR policies, and the accessibility of HR officers. At
lower levels of management, some voices were raised against the HR organisation
regarding the excessive focus on cost containment and the sometimes callous application
of policy, in particular absence policy.
Centralised HR. Some components of the HR system were highly centralised, in the
sense that decisions were taken at Head Office and fed down to individual stores, often
without prior consultation. For instance, Head Office determined the staff structure for
each store depending on a generic template for in-store processes, and the specific
revenue targets and opening hours of the store. This structure specified each position to
be filled in every department in every store, with their corresponding level, salary grade,
shift and contract hours. Target structures were adjusted frequently because of redesign
 6

of standard store processes, or changes in target revenue for the store. Adjustments in the
target structure of a store often triggered other HR processes such as new job openings,
job ‘red-ringing’ (elimination), job downgrades and upgrades.
Another important HR component that was determined centrally was the staffing level of
the stores. Full time employees worked a standard number of hours every week, no extra
time was allowed for them. However, for part time employees, contract hours tended to
be less than the time they would actually be needed for. The additional hours were
budgeted by Head Office according to revenue targets; store managers would then
distribute them among part timers. For this purpose, most stores used a corporate
software application called “Staff Planner”, which worked out the precise individual
shifts needed, based on historical data. Usually, only last minute adjustments due to
unforeseen absence needed to be worked out by managers.
Finally, salaries and bonus payments were also determined centrally. Each position
defined in the target structure of a store was attached to a salary grade. Salary grades had
only three salary levels: entry –for new entrants-, competent –automatically attained on
completing a three-to-six month probation period to the satisfaction of the manager-, and
above –only exceptionally awarded to high contributing members of staff, following their
manager’s proposal. Exceptions to these set salary levels were very rare. Bonuses were
paid monthly and seasonally, based on attainment of store level goals. Individual payouts
were calculated by Head Office according to each person’s salary grade and hours
worked during the period.
Decentralised HR. Some elements of the company’s HR system were decentralised, in
the sense that decisions were made by line managers, within more or less defined
corporate policy, and with more or less consultation with Head Office. My investigation
showed that there were two major HR areas where decisions were largely made by line
managers: progression –practices and procedures aimed at moving employees from one
position to another within the company-, and performance management –practices and
procedures aimed at correcting undesired employee behaviour.

Progression
Throughout my interviews in the stores, respondents spoke profusely about progression.
Often this referred to the upward moves (i.e. moves to higher-grade positions) that many
of them had experienced through their time of employment at the company. I found,
however, that upward moves operated in a similar fashion to other kinds of voluntary
move. In my description here and later discussion, therefore, I am expanding the
definition of progression to include the practices and procedures that enabled all
employee voluntary moves whether to a higher-grade position or otherwise.
I found evidence for employees seeking progression on three different dimensions. First,
many employees sought progression in terms of pay. For part-timers, this meant seeking
more hours and eventually becoming full-time employees; whereas for full-timers it
meant moving to successively higher grade positions. Second, some employees sought
progression in terms of life balance, including working closer to home, working
weekdays only and reducing total hours worked. Finally, some employees sought
progression in terms of job-content preferences, such as managerial versus non-
 7

managerial roles, commercial versus non-commercial roles, or store versus Head-Office

roles.
It is important to remind the reader here that, as discussed earlier, management at the
store level had virtually no influence on the number and nature of positions in their store,
or on the salary and bonus paid to incumbents. This meant that, first, the single best way
–often the only possible way 4 - for an employee to progress on pay, life balance or job
content was by moving to a different position within the structure; and second, that
availability of progression opportunities depended on vacancies coming up, either
because a new position had been created due to a review of the store’s structure or –more
often- because an existing incumbent had moved to another job or quit the company.
Progression was competitive. Applicants expected to undergo a formal selection process
in which their suitability for the position was assessed against fellow applicants.
Managers generally sought to maximise the number of quality applications by publishing
the vacancy internally and by encouraging anyone they considered a good candidate to
apply. Indeed, most vacancies 5 were published in a weekly companywide vacancy
bulletin, which featured prominently on staff-room notice boards. Advertisements
included detailed standard information on title, department, grade, shift and hours. All
current full- or part-time employees interested in a position were entitled to apply.
Although there was a formal requirement for prospective applicants to speak to their
current managers before applying, these conversations were described by respondents as
“developmental”, focusing on advice on whether the prospective applicant was “ready” to
take up the position. Ultimately, no formal permission from the current manager was
required 6 .
The selection process typically involved an interview with the hiring manager. This
could consist of a standard set of questions recommended by HR, or the manager could
design their own set of questions or even activities for the candidates to perform (e.g.
plan a change of season). Responsibility for the selection process rested with the hiring
manager. Others –the manager’s own manager and peers, HR- were sometimes involved
in support roles (e.g. note taking, second opinion) at the request of the hiring manager.
There was full agreement across all respondents that hiring decisions were the hiring
manager’s 7 . Within a store, each manager took responsibility for selecting their team.
As one store manager put it:

4
Occasionally, positions were upgraded (or downgraded) as a result of reviews to a store’s structure, and
this meant the incumbent’s employment terms were adjusted accordingly. However, this happened rarely,
and the timing and direction of the move was out of the control of anyone within the store.
5
Occasionally, direct appointments were made which did not involve publication, but these tended to be
restricted to lower level positions and to cases in which the appointment provided a solution to an existing
placement problem (e.g. placement of an employee whose position had been ‘red-ringed’).
6
There was evidence of at least one respondent who successfully applied for a position against the advice
of his then manager.
7
There was evidence of hiring decisions against the opinion of HR and of the recruiting manager’s
manager. Only in the case of some showcase stores, would the Area Manager supersede the store manager.
 8

I think it is important for [my managers] to be able to make their own [hiring]
decisions, whether they be right or wrong; live with them and manage them.
There was no suggestion by any of the respondents that such freedom for the hiring
manager resulted in unfair selection processes or unequal opportunities for progression.
On the contrary, a manager’s right to choose their team members was recognised as
legitimate by all my interviewees 8 . Respondents attributed success in selection processes
to an applicant’s “ability to do the job”, but also to their “fit” with the current team. In
practice, a candidate’s chances improved significantly through demonstration of prior
experience in the role and through enhanced visibility outside their own store. In
particular, there were three ways in which employees could acquire experience and
visibility.
First, they could cover for absences in roles larger than their own and in stores different
than their own. Covering involved temporarily performing someone else’s job. Usually,
it implied leaving one’s job 9 –which may in turn also need to be covered. A cover could
last anything from a few hours to several months; sometimes its duration was determined
in advance and sometimes it was open-ended. Some covers were routine, planned events
(e.g. vacation). When a planned absence was long (e.g. maternity leaves), covers could
involve a competitive selection process similar to a full appointment, including
publication in the vacancy bulletin and an interview. Often, however, covers were
unexpected, most notably in the case of illness. In such cases, the absentee’s manager
would network with other managers in nearby stores, who would selectively offer the
opportunity to their staff.
Second, employees could acquire experience by taking delegated functions from their
managers. This job enrichment or enlargement was seen as a way to develop employees
as well as to try them for higher roles. Arrangements were made at the discretion of each
manager, and had no implications for employee rewards. There were numerous
references to this practice at all levels. For instance, an experienced store manager took
up responsibilities from her area manager, carrying out store visits for her and leading
fellow store managers in specific activities. Likewise, a sales supervisor whose manager
had given notice of leaving in a few month’s time was effectively acting as manager,
even as a selection process was underway in which she was competing against other
applicants for the position.
Finally, employees could improve their visibility by participating in area- or corporate-
level project work, such as new store openings or the development of next season’s
collections. Appointments for these projects were made informally through the
recommendation of managers.
All three activities above were widespread, with employees on progression track
participating in one, two or all three. None of the activities earned higher rewards or

8
The fact that all employees had access to information on vacancies, were entitled to apply, and usually got
the chance to defend their application at an interview seemed to infuse a measure of fairness to the system.
9
Sometimes covering someone meant doing two jobs at the same time, especially when the two jobs
involved similar functions in the same organisational unit (as when covering for one’s manager).
 9

formal recognition; the only advantage for participants was precisely their increased
chances of progression.

Performance management
Respondents used the expression “performance management” to refer to a set of practices
and procedures aimed at correcting undesired employee behaviour. This included a
continuum of activities ranging from on-the-job supervision and guidance, to informal
verbal counselling, formal performance reviews, and disciplinary procedures leading to
formal sanctions. Performance appraisals were not considered part of performance
management because they were used as a form of praise and as a guide for employee
development, rather than for discipline.
On the most common, routine, day-to-day end of the performance-management spectrum,
managers engaged in supervision on the shop floor. One of the main functions of all
layers of management was to regularly review the work done by lower organisational
levels. This was seen as crucial for ensuring conformity with company standards. For
instance, store managers regularly started their work day with a morning floor walk in
which feedback was given on all aspects of commerciality. Likewise, sales managers and
supervisors spent time on the sales floor expressly watching their team work and
“corrective-coaching” them:
We are told corrective coaching is the best way, so just quickly as soon as you see
something, nip it in the bud, go along and say ‘Oh, try it like this, it may work
better’.
Whenever a supervisor felt that an issue with an employee’s behaviour on the job was not
likely to be resolved through corrective coaching on the shop floor, he or she could
decide to ask the employee to have an informal private conversation. Sometimes this
also happened on the initiative of an employee who might feel the need to address an
outstanding issue.
The next level of performance management was the formal review. Typically, employees
were put on review when informal performance management had not been effective. For
the formal review, the manager produced a document describing the specific behaviour
that needed to change, along with a set of actions and a time when performance would be
reviewed again. In a ‘counselling session’, this document was discussed with the
employee. After the time had elapsed, if the behaviour was corrected, the employee
came off formal review; otherwise a time for a new review was set. After a number of
formal sessions, if the behaviour persisted, the manager could decide to proceed to
disciplinary. The following excerpt offers a good description of how formal reviews
operated:
I spoke to him verbally about things to start with, and then I didn't think he'd
improved so then we go to a formal counselling session, we get those documented,
so I had three of those documented on certain issues that I wasn't happy with,
personally written down on a duplicate format so he would sign it, I would sign it,
we would agree some action points, go back in a week, see if those action points
were being improved on, if they weren't, then we went to a disciplinary.
 10

Disciplinary was the final stage in performance management. It included a first written
warning, a final written warning, and the imposition of formal sanctions on the employee.
Sanctions ranged from inserting a note in the employee’s personnel record to dismissal.
However, dismissal was only used in grave cases of employee misconduct, such as theft
or fraud. The worst sanction imposed in most other situations was a forced change of
jobs, maybe to a lower grade position.
Faced with unwanted behaviour, managers chose an approach to each situation based on
corporate policy, their prior experience as managers and the advice of HR officers.
Sometimes a single issue would go through the whole spectrum of performance-
management actions, from corrective-coaching to formal sanctions. Often however, less
important issues did not make it past a certain action, even if they were not corrected.
Likewise, grave offences were taken directly to disciplinary, without any of the other
actions.
The types of behaviours that needed to be corrected were varied. In practice, three types
of issues were most frequently mentioned by respondents. First, employees were often
performance-managed because of repeated lateness. At the time of this research, the
company was in the process of implementing a new clock-in system. In the past,
managers had to manually fill in a time sheet with the times worked by each employee.
Often, late arrivals went unaccounted for. With the new system, employees themselves
clocked in and out at the till at the start and end of their shift. The clock-in system was
directly linked to payroll, so that a delay of more than 5 minutes meant 15 minutes were
taken off that day’s pay. Conversely, if an employee clocked out late, he was only paid
until the end of his shift 10 . Because the new system removed confusion about acceptable
lateness and had direct pay implications, it was expected to help reduce overall tardiness.
In terms of performance management, the new system was also expected to help
managers detect and manage lateness issues at an earlier stage.
A second major reason for performance management was absence. The company
operated a system of trigger points, by which if an employee was absent a certain amount
of days in any 12-month period, he or she would automatically be given first written
warning, followed by final warning and dismissal if further trigger points were reached.
This policy had been strictly enforced by HR over the past few years and was generally
perceived as harsh. More experienced respondents recalled a time when managers judged
whether an employee’s absence was justified or not before taking performance
management actions. Employee attitudes towards the current policy ranged from
accepting it as a sign of ‘professionalism’ to openly resenting it, perhaps depending on
how vulnerable the employee felt. Some of the resentment was expressed by means of
‘horror stories’ about employees who, in the most strenuous circumstances, had been
given warning.
Finally, performance management actions were often used to correct poor customer
service behaviours. One of the main duties of any manager in the company was to ensure
compliance with its high standards for customer service. Many of the corrective-

10
If a manager wanted an employee to stay longer on a particular day, she could manually alter the shift
information so that the extra time was paid.
 11

coaching and informal counselling activities were geared towards correcting customer
service behaviours. Additionally, the company had in place a ‘traffic-light system’ to
help manage more serious issues. The system involved rating all employees as being
green, amber or red in terms of quality of their customer service. If an employee was
classed as amber, he would be put on performance review. If he was classed as red, he
would be taken to disciplinary. Implementation of this system seemed patchy, however.
Some managers seemed more inclined to use it, even though they may have mixed
feelings about the efficacy of the ensuing performance management actions. Other
managers did not pay much attention to the traffic-light system, basically rating every
member of their staff green.

Discussion
In this section, I discuss my findings through the concept of HR discretion. First, I
describe how companies design HR policies to enable varying levels of discretion and
propose a three variable model to explain these decisions. Second, I explore two ways in
which companies keep their managers’ HR decisions aligned with organisational
objectives. Finally, I discuss different arrangements under which managers may exercise
HR discretion.

Designing HR discretion
HR discretion is partially determined by the design of HR activities. Companies design
their HR policies with varied levels of manager discretion. Within any single HR system,
some activities may enable a high level of HR discretion, while others may not. In the
case under study, the company made centralised decisions for activities such as staff
structure, staffing levels, pay, absence and lateness. Conversely, decisions on
progression and most areas of performance management were left to the managers on the
field.
Three variables can be used to explain why companies may choose to restrict manager
discretion on some HR activities and not others. First, companies will tend to restrict
discretion for HR decisions with a direct impact on cost. In our case study this would be
the case for staff structure, staffing levels and pay. Second, companies will tend to give
managers discretion on HR activities that depend on information that is distributed and
difficult to systematise. For instance, performance management for quality of customer
service relies on rich observations by managers, which might be too costly to build into a
centralised system. Finally, companies will tend to give managers discretion for HR
activities that enable productive exchanges between managers and employees, and
restrict it for those that do not. For instance, the decision to centralise the management of
lateness was expected to reduce ambiguity around the control of time worked, which was
a source of friction between manager and employee.
Further insight can be gained by looking at the changes in the management of absence.
In the past, decisions to performance-manage employees because of repeated absence
were made by managers who took into account a number of context variables, such as the
apparent causes of the absence, the history of the employee, and his or her personal
circumstances. A few years back, HR started applying a strict standard policy of
 12

disciplinary after a certain level of absence was reached, irrespective of any of the above
circumstances. This policy was cited by many of my respondents as one of the most
controversial in the management of HR at the company, both from the point of view of
employees who felt vulnerable, and from the point of view of managers who thought this
caused more problems than it solved at the store level. This on-going resistance to the
policy may be explained through the three variable model proposed above. First, the
company sought to restrict managerial discretion on absence because of the important
cost effects of paying absentee salaries and covering their position in the store. In terms
of information, it is easy to control and centralise information on days not worked. It is
however much more difficult to centralise information which is also relevant to the
decision to apply discipline, such as apparent causes, employee history and personal
circumstances. Finally, absence management may be a source of productive exchanges
between managers and employees. Several of the respondents interviewed recalled
instances earlier in their career when they had a period of increased absence. The way in
which their managers treated their case was recalled as a source of gratitude which was
reciprocated with enhanced commitment to both the company and the manager
concerned. They subsequently had long and productive careers in the company. With
the new system, not only were these exchanges thwarted, but there was a strong sense of
unfairness around the policy, manifested in ‘horror stories’ about employees being
disciplined in very difficult personal circumstances.

Aligning HR discretion
By designing HR activities to afford discretion to line managers, companies run the risk
that managers make HR decisions to the detriment of the organisation. This concern is at
the heart of the economics conception of managerial discretion. Companies manage this
risk in a number ways. First, companies limit managers’ latitude of objectives through
performance pressure (Shen and Cho, 2005). In our case study, performance pressure at
the store level was considerable. The basic measures of commercial activity were total
sales, average transaction value and units per transaction. Seasonal, monthly and weekly
targets were set by Head Office for each store, broken down by department. Measures
were produced instantly through the information provided by the tills. Every day started
with a review of the previous day’s figures and their evaluation against weekly targets. A
significant amount of everyone’s compensation was paid monthly and seasonally only if
targets were met. Stores or departments that did not meet their targets could be
downgraded and their structure downsized, which meant reductions in contract hours,
losses in salary grades and red-ringing of positions. All of this limited the incentives for
managers to make HR decisions that damaged the store’s performance. One respondent
explained:
I have a friend who works in another store, and people will go to that manager
and in some ways she'd change maybe hours to suit people, and in the end that
doesn't end up helping in store, and she does it because she wants to keep her
staff happy. But you know, yes, I want to keep my staff happy, but I'm not going to
do that in detriment of the store, because it only ends up… you end up having to
change things anyway.
 13

Another way in which companies limited the risk that managers would make ‘bad’ HR
decisions is through the support of the HR function. In our case study, for instance, store
managers could have the incentive to ‘poach’ employees from other stores, while holding
on to their own. This is however limited through the efficacy of HR to provide
candidates for a replacement via the vacancy bulletin and the external jobs website.
Likewise, HR plays an important role advising managers in performance management
processes and maintaining an appeals process.

Exercising HR discretion
An important insight in the management literature is that a manager’s discretion is
defined by the context as much as by his/her own actions: managers can create discretion
by seeing possibilities that others had not seen, or by persuading others of a course of
action previously ruled out (Hambrick and Finkelstein, 1987). From my field work, I
was able to identify four different arrangements under which managers exercise HR
discretion. First, policy or practice may dictate specific areas of manager prerogative.
For instance, respondents clearly agreed that line managers made hiring decisions at our
focal organisation. Head Office did not intervene; their role in selection was merely to
give administrative support for publishing the vacancy and processing applications. Even
within a store, each manager was largely free to choose who to bring in as their direct
reports.
Second, influence on some HR outcomes may be a consequence of influence on a related
HR outcome. For example, managers at the company were able to influence an
employee’s progression outcomes through their ability to decide whether they would
cover absences outside their store. Such links between different HR activities may be
established by policy design, as the common case in which a performance appraisal
outcome is used for merit salary increases. Often, however, the links may be unintended
and undetected by policy makers.
Thirdly, managers were sometimes able to influence HR outcomes by engaging in
negotiations with Head Office. This is illustrated by an episode where a store manager
challenged HR on the automatic application of a standard level of discipline following a
serious breach of security procedures. In this particular instance, a junior manager had
lost her store keys, which was considered a serious offense that carried a final written
warning. The store manager was able to argue with HR, and effectively persuade them,
that the circumstances surrounding the incident justified a much lesser sanction. Two
things are important about this case. First, the store manager made a decision to pursue a
different kind of action to policy; and second, his success in pushing his decision
depended on his ability to influence his HR counterpart.
Finally, some managers were able to influence HR outcomes by dodging policy controls.
In one instance, Head Office directed all part time contracts to be reduced in hours. In
one of the stores, one particular employee had his contract hours reduced, but was given
assurance that he would continue to do the same weekly hours as before through extra
time. This finding echoes some previously reported in the literature. For instance,
Purcell and his colleagues reported that, in organisations where managers seemed to have
little discretion over formal rewards, there were “‘unofficial’ practices creeping in as a
means of motivating people such as allowing time off, access to training, and providing
 14

more challenging work.”(Purcell et al., 2009: 63). Similarly, Nadisic (2008a, 2008b)
proposed that managers allocate benefits for “uses other than those for which they were
formally intended”, particularly to compensate for unfairness in HR policy areas outside
their influence.

Conclusion
In this paper I have given an account of how managers carry out their HR activities. I
have shown that companies design HR systems to enable varying levels of discretion, and
that they take steps towards aligning line manager HR decisions with organisational aims.
I have also shown how line managers can exercise HR discretion under a number of
different arrangements. The resulting picture involves the recognition that managers can
have a determining influence on the outcomes of many HR processes. To a great extent,
HR decisions are locally configured by line managers who exert influence in the political
processes leading up to specific decisions about the employment terms of their
employees, often in ways not contemplated in policy and unexpected by top management.
Thus, managers do not only act as agents of the organisation, but can exercise discretion
in several different ways, as they pursue their own objectives as managers of a business
unit. How they exercise this discretion is an important component of an HR system and
will have implications on such outcomes as employee behaviours and attitudes, and
organisational performance. Further research on the role of line managers and the HR
discretion construct may therefore prove fruitful in our understanding of the performance
effects of human resource management.

References
BARTEL, A. P. 2004. Human resource management and organizational performance:
evidence from retail banking. Industrial & Labor Relations Review, 57, 181-203.
BOSELIE, P., DIETZ, G. & BOON, C. 2005. Commonalities and contradictions in HRM
and performance research. Human Resource Management Journal, 15, 67-94.
BOXALL, P. & PURCELL, J. 2003. Strategy and human resource management,
Basingstoke, Palgrave Macmillan.
BREWSTER, C. & SODERSTROM, M. 1994. Human resources and line management.
In: BREWSTER, C. & HEGEWISCH, A. (eds.) Policy and practice in European
human resource management : the Price Waterhouse Cranfield survey. London;
New York: Routledge.
GUEST, D. E. 1987. Human resource management and industrial relations. Journal of
Management Studies, 24, 503-21.
GUEST, D. E. 1989. Human resource management: its implications for industrial
relations and trade unions. In: STOREY, J. (ed.) New perspectives on human
resource management. London; New York: Routledge.
 15

HAMBRICK, D. C. & FINKELSTEIN, S. 1987. Managerial discretion: a bridge between

polar views of organizational outcomes. Research in Organizational Behavior, 9,
369.
HOWARD, J. L. & FERRIS, G. R. 1996. The Employment Interview Context: Social and
Situational Influences on Interviewer Decisions. Journal of Applied Social
Psychology, 26, 112-36.
LEGGE, K. 1989. Human resource management: a critical analysis. In: STOREY, J. (ed.)
New perspectives on human resource management. London; New York:
Routledge.
LEGGE, K. 1995. HRM: rhetoric, reality and hidden agendas. In: STOREY, J. (ed.)
Human resource management: a critical text. London; New York: Thomson.
LEGGE, K. 2005. Times are a-changing: Human Resource Management, Rhetorics and
Realities ten years on. Preview/postcript for anniversary edition. Human resource
management: rhetorics and realities. Houndmills, Basingstoke, Hampshire; New
York, N.Y.: Palgrave Macmillan.
LEVY, P. E. & WILLIAMS, J. R. 2004. The Social Context of Performance Appraisal: A
Review and Framework for the Future. Journal of Management, 30, 881-905.
LIDEN, R. C., MARTIN, C. L. & PARSONS, C. K. 1993. Interviewer and applicant
behaviors in employment interviews. Academy of Management Journal, 36, 372-
86.
MARTÍNEZ LUCIO, M. & WESTON, S. 1992. Human resource management and trade
union responses; bringing the politics of the workplace back into debate. In:
BLYTON, P. & TURNBULL, P. (eds.) Reassessing human resource
management. London; Newbury Park, Calif.: SAGE Publications.
MCGOVERN, P., GRATTON, L., HOPE-HAILEY, V., STILES, P. & TRUSS, C. 1997.
Human resource management on the line? Human Resource Management
Journal, 7, 12-29.
MINTEL 2010. Fashion: Impact of The Recession - UK - June 2010. London: Mintel
International Group Ltd.
MINTZBERG, H. 1978. Patterns in strategy formation. Management Science, 24, 934-
48.
NADISIC, T. 2008a. The fair hand of managers: Managers’ visible and invisible
corrective justice strategies and their antecedents. Doctorat en Sciences de
Gestion.
NADISIC, T. 2008b. The Robin Hood effect. Antecedents and consequences of
managers using invisible remedies to correct workplace injustice. In:
GILLILAND, S., SKARLICKI, D. & STEINER, D. (eds.) Justice, morality, and
social responsibility. Charlotte, NC: Information Age Pub.
NISHII, L. & WRIGHT, P. M. 2008. Variability within organizations: implications for
strategic human resources management. In: SMITH, D. B. (ed.) The people make
 16

the place : dynamic linkages between individuals and organizations. New York:
Lawrence Erlbaum Associates.
PURCELL, J. & HUTCHINSON, S. 2007. Front-line managers as agents in the HRM-
performance causal chain: theory, analysis and evidence. Human Resource
Management Journal, 17, 3-20.
PURCELL, J. & KINNIE, N. 2007. HRM and business performance. In: BOXALL, P.,
PURCELL, J. & WRIGHT, P. (eds.) The Oxford handbook of human resource
management. Oxford: Oxford University Press.
PURCELL, J., KINNIE, N., SWART, J., RAYTON, B. & HUTCHINSON, S. 2009.
People management and performance, London [u.a.], Routledge.
SCOTT, B. A., PADDOCK, E. L. & COLQUITT, J. A. 2009. An Actor-Focused Model
of Justice Rule Adherence and Violation: The Role of Managerial Motives and
Discretion. Journal of Applied Psychology, 94, 756-69.
SHEN, W. & CHO, T. S. 2005. Exploring involuntary executive turnover through a
managerial discretion framework. Academy of Management Review, 30, 843-54.
TRUSS, C. 2001. Complexities and controversies in linking HRM with organizational
outcomes. Journal of Management Studies, 38, 1121-49.
TRUSS, C. & GRATTON, L. 1994. Strategic human resource management: a conceptual
approach. International Journal of Human Resource Management, 5, 663-86.
WILLIAMSON, O. E. 1963. Managerial Discretion and Business Behavior. The
American Economic Review, 53, 1032-57.
WRIGHT, P. M. & BOSWELL, W. R. 2002. Desegregating HRM: a review and
synthesis of micro and macro human resource management research. Journal of
Management, 28, 247-76.