IJOPM

22,1
30
International Journal of Operations &
Production Management,
Vol. 22 No. 1, 2002, pp. 30-49.
# MCB UP Limited, 0144-3577
DOI 10.1108/01443570210412060
Dispelling the modern myth
Employee satisfaction and loyalty drive
service profitability
Rhian Silvestro
Operations Management Group, Warwick Business School,
University of Warwick, Coventry, UK
Keywords Job satisfaction, Loyalty, Productivity, Efficiency, Profitability, Supermarkets
Abstract This paper reports some empirical findings which appear to challenge the received
wisdom prevailing in the operations management, service management, TQM and HRM
literatures, namely, that employee satisfaction and loyalty are key drivers of productivity,
efficiency and profit. An empirical study of one of the UK's four large supermarket chains reveals
an inverse correlation between employee satisfaction and the measures of productivity, efficiency
and profitability, the most profitable stores being those in which employees are least satisfied.
Employee loyalty, measured in terms of length of service, also appears to be inversely correlated
with productivity and profitability. It also emerges that the pressure to maximise store efficiency
may be leading to dysfunctional managerial behaviour at store level. These preliminary findings
suggest two imperatives for managers and academics. For managers, it is advocated that they
analyse the relationship between employee satisfaction, loyalty and financial performance in their
own organisations rather than assuming that the rhetoric of the management literature applies in
all operational contexts. For academics, four contingent variables are proposed which distinguish
those service contexts in which the assumed relationship may pertain: services where customer
contact with staff is high; services where there is little opportunity for technological substitution;
services where staff contact is critical to the customer value proposition; and services with high
labour costs.
Introduction
It has been widely argued in the operations management, total quality
management (TQM), human resource management (HRM), and service
literatures that improving employee satisfaction and loyalty leads to higher
service productivity and profits. Employee satisfaction and loyalty are seen as
critical to the capability of service organisations to respond effectively to
customer needs, whilst also driving down costs through reduced recruitment
and training expenditure and all the cost efficiencies which accrue from skilled
workers who are up to speed and familiar with both the tasks at hand and their
customers.
Yet there is a noticeable dearth of empirical evidence to support what has
become so commonly an accepted part of modern management philosophy.
This paper reports the results of a preliminary study of the links between
employee satisfaction, loyalty, productivity and profit in one of the UK's four
The current issue and full text archive of this journal is available at
http://www.emeraldinsight.com/0144-3577.htm
The author wishes to acknowledge the contribution of Stuart Cross, Head of Strategy
Development, Boots the Chemist, Nottingham, and Warwick Business School alumnus, 1997, for
his assistance in collecting and analysing the data during the early stages of this research
project.
Dispelling the
modern myth
31
large superstore retailers. Correlation analyses of the links between these four
key performance variables reveal a picture which is quite different to that
described in the literature and so often expounded by management consultants.
For the evidence suggests that the most productive and profitable stores are
those where staff are least satisfied and loyal.
This raises the question of whether productivity and profit are being
generated in this sector despite rather than because of the management of its
human resources; a timely issue, in view of the scrutiny which this sector has
recently attracted for its supposed exploitation of the supply network (see for
example, Nuki and Bevan, 1998). Having been repeatedly accused of squeezing
their suppliers on cost without passing on reduced prices to the end customers,
this empirical study signals the possibility that superstores may be
pressurising their employees as well as their supply networks.
The received wisdom
The American TQM ``gurus'', such as Deming (1986) and Juran (1989), are
unanimous and unequivocal in the view that increasing process ownership
and job satisfaction will yield returns in both quality and productivity. The
Japanese quality experts also emphasise the importance of the ``human factor''
in creating an environment for production excellence. Ishikawa (1985, p. 96),
for example, claims that Deming prize winning organisations are
characterised by ``showing respect for humanity, nurturing human resources,
considering employee happiness, providing cheerful workplaces''. He insists
that effective quality control relies on an understanding of the ``human
drives'' which include the satisfaction of doing a job well, the happiness
which comes from co-operation with others and from personal growth and
fulfilment (Ishikawa, pp. 27-8). Shimada (1993) similarly sees ``human
technology'' and motivation as being necessary preconditions for production
control and improvement, and for outputs including higher quality, low price,
growth and profit.
This TQM rhetoric, which directly links employee satisfaction to
organisations' abilities to sustain long term business growth and profitability,
also permeates the HRM literature, as expounded by, for example, Walton
(1985) and Guest (1987). As Legge (1995, p. 240) explains:
The rhetoric of high value added or zero-defect products and of ``customer care'' at the
organisation-market interface, generally includes a mantra about the importance of well-
trained, skilled and committed employees. This is almost tautological . . . Demands for
employees' commitment to the organisation and responsiveness to the customer logically
demand a reciprocal commitment to the employee and responsiveness to her demands as
internal customer.
However, the thrust of Legge's text, as is explicit in the subtitle, Rhetorics and
Realities, is that there is a gap between the management rhetoric of these
exponents and the realities of implementation. There have been numerous
empirical accounts of production environments where, despite the rhetoric of
empowerment, new wave manufacturing methods, including TQM and just-in-
IJOPM
22,1
32
time (JIT), have in fact led to reduced employee discretion, increased control,
peer surveillance and labour intensification (for example, Turnbull, 1988,
Delbridge and Turnbull, 1992; Garrahan and Stewart, 1992; Kerfoot and
Knights, 1995). Such examples have led critical theorists to argue that TQM
and JIT environments, far from increasing employee satisfaction, in fact,
generate even more stressful working environments:
. . . it would be tempting to dismiss moves towards employee involvement under TQM as just
another ruse constructed by management to encourage workers to engage more freely in their
own exploitation (McArdle et al., 1995).
The organisational behaviour literature of the past decade can therefore be said
to have taken a more ambivalent and questioning line with regard to the
teachings of TQM and its assumptions about the relationship between
employee satisfaction and loyalty, productivity and profit. Indeed Ledford
(1999) argues that the continued survival of the ``happy productive worker
ideology'', despite the absence of empirical evidence to support it, is due to an
historical convergence of the interests of managers, workers and researchers.
From the management perspective, the thesis that happy workers are most
productive obscures the legitimate grounds for conflict between labour and
management and encourages management manipulation of employees
(Perrow, 1986); whilst employees can use exactly the same thesis to support
their demands for improved pay and working conditions.
Academic researchers, meanwhile, use the argument to dispel ``any guilt we
might feel for theory and research that otherwise would make us feel like an
unwitting tool of either management or labour'' (Ledford, 1999, p. 27). This is
accompanied by a natural inclination towards the academic aesthetic of
elegant, neat and inclusive theories. According to Ledford (1999, p. 28) there is a
tendency to over-generalise and ``the emphasis is on finding ways to
simultaneously improve such outcomes as morale, productivity, quality,
teamwork, customer service, employee commitment, and turnover''.
Whilst practitioners and academics alike appear to be easily seduced by the
case for directly linking employee satisfaction with improved productivity, the
mantra is particularly strong in the service management literature. Indeed it
would seem highly unfashionable, as well as politically incorrect and
managerially retrograde, to argue to the contrary! As far back as the 1970s
Marriott was famed for his quotation ``you can't make happy guests with
unhappy employees'' (Hostage, 1975); and Carlzon (1987) is just one example of
a charismatic leader who explained his renowned turnaround story in terms of
nurturing happy and productive front liners. More recently, Rucci et al. (1998)
describe what they call ``the employee-customer-profit chain'' in Sears, in the
US, and demonstrate how focusing managerial efforts on improving employee
satisfaction and loyalty resulted in higher levels of customers satisfaction and
financial performance. There is considerable anecdotal evidence, then, to
support the view that employee satisfaction is a key driver of business growth
and profitability.
Dispelling the
modern myth
33
In the service literature, awareness of the ways in which employees can
directly impact upon customer perceptions of service has led to general
acceptance of what Heskett et al. (1997) refer to as the ``satisfaction mirror'' in
which employee satisfaction is ``reflected'' in terms of customer satisfaction,
which in turn generates business growth and profitability. Conversely, they
maintain that failure to satisfy employees will eventually be reflected in
customer dissatisfaction. Schlesinger and Heskett (1991) similarly claim that
employee frustration leads to high labour turnover, and reinforces a ``cycle of
failure'' characterised by minimal investment in training, poor rewards and
declining levels of customer service. Reichheld (1996) further argues that
improving employee loyalty reduces operational costs, as well as improving
customer service, thus driving improved profits.
Heskett et al. (1997) have developed and integrated all this work, with their
conceptualisation of the ``service profit chain'' in which employee satisfaction
and loyalty initiate a chain of performance links between quality, productivity,
service value, customer satisfaction and loyalty which in turn drive profit and
growth.
The past decade has seen the emergence of several studies which endeavour
to lend empirical support to these relationships. However most of these studies
provide evidence of links between service quality, customer satisfaction,
loyalty and financial performance (see for example, Reichheld and Sasser, 1990;
Rust et al., 1995; Zeithaml et al., 1996) rather than explore the effect of employee
satisfaction on the rest of the chain. Although Schneider and Bowen (1993)
provide documented evidence of the relationship between employee and
customer satisfaction, the mirror effect is not linked to business performance.
More recently, Loveman (1998) reports the results of a major study of a
slightly modified version of the service profit chain in a network of branches in
a regional US bank. His study considers the link between employee
satisfaction, loyalty and profitability, but does not consider branch
productivity. The study incorporates the results of an extensive employee
satisfaction survey; however, Loveman recognises the limitations of the
financial measures which were used. Branch profit could not be measured due
to the high level of cost allocations at branch level. Revenue was therefore used
instead of profit; but Loveman concedes that this is a proxy measure which
``fails to capture any positive profit consequences associated with productivity
or cost control from, say, loyal employees serving loyal customers'' (Loveman,
1998, p. 30).
Whilst Loveman's results lent some general support for the service profit
chain model, an element of complexity emerged in the relationship between
employee satisfaction, employee loyalty (measured in terms of employment
tenure and employees' stated loyalty) and financial performance. The findings
demonstrated a positive correlation between employee satisfaction and
employees' stated loyalty but there was no link between employee satisfaction
and employee tenure. Furthermore, employment tenure was linked to financial
performance; but stated loyalty was not. Thus the link between employee
IJOPM
22,1
34
loyalty, satisfaction and financial performance depends crucially on the way in
which employee loyalty is measured and the study is inconclusive about the
assumed status of employee satisfaction as a driver of financial performance.
In studying the links between employee satisfaction, loyalty, productivity
and profitability, as part of their work on the service profit chain, Heskett et al.
(1997) provide surprisingly little empirical evidence. In fact, only six companies
are cited and in none of these companies are links established between all four
variables; which, in fact, represents a failure to provide empirical evidence for a
linked chain.
To conclude, a cross-functional review of the management literatures reveals
general acceptance of the view that employee satisfaction is a driver of service
profitability, yet a dearth of convincing empirical support. In this study a
sample of stores from a large retail network in the UK was studied with a view
to exploring the relationship between employee satisfaction, loyalty,
productivity and profitability, and establishing whether or not these four
variables were correlated in a single, multi-site organisation. If the received
wisdom is correct, one would expect to find that employee satisfaction and
loyalty was highest in the most productive and profitable stores.
Methodology
Access was provided to a leading player in the UKgrocery retail sector, a market
dominated by a small number of major players. With a national network of large
family supermarkets, this company was particularly appropriate for the study
because its senior management had, apparently, fully bought into the
management philosophy that employee satisfaction is a driver of business
success. Indeed, in the 1997 Annual Report the Chief Executive Officer (CEO)
stated that customer satisfaction and loyalty were ``the real drivers'' of this
company's profit and growth, and that they were strongly influenced by how
employees felt about their work, their rewards and their manager. This was also
a central theme in the company's management training programme.
Fifteen branches from a single region of the network of stores were selected
for the study. Given the strong corporate concept, the stores were comparable
in terms of design format, service range, appearance, layout etc.; but the stores
were evenly spread along a continuumof high through to lowperformance.
The analysis was facilitated through a study of a combination of both
quantitative, hard measures and qualitative, soft measures. The performance
indicators selected for the analysis are explained below.
Hard measures
One hard measure of employee loyalty was used, which was available within
the company, namely, staff turnover at store level. The company could also
provide data on the profit margins and productivity of each store. Store profit
margin was measured in terms of a ratio of store contribution to sales. Three
measures of store productivity were provided: sales per square foot,
contribution per square foot, and contribution per staff hour. Sales and
Dispelling the
modern myth
35
contribution per square foot represented key measures of capacity utilisation;
whilst contribution per staff hour was the main labour productivity indicator.
Another measure which was given considerable management attention at
both store level and head office was the operating ratio. This was a ratio of
actual to planned working hours and targets would be set for each store for
each period based on long term sales forecasts. As sales forecasts were
adjusted in the short term, store management would have the discretion to
change the number of planned hours in order to meet demand. Exceeding the
planned hours would result in a ``high operating ratio''; and management were
encouraged to adhere to their targets. Whilst this was not strictly a financial
performance measure nor a productivity indicator (since it is not a ratio of
inputs to outputs (Fitzgerald et al., 1991)), it is considered by the company's
management to be an important efficiency measure. Given the importance
placed by management on this performance measure and its clear conceptual
proximity to productivity, store operating ratios were collected, with the
expectation that the most productive stores would also be the most efficient.
Soft measures
It was found that no measures of employee satisfaction were available at store
level. A small survey was therefore conducted based on just six of the 15 stores;
the small sample size being due to the resource and time constraints of the
research project. Again, just like the 15 stores, these stores were evenly spread
in terms of performance. A sample of between 20 and 30 employees in each
store were asked to rate the following on a 1-5 scale:
.
overall satisfaction with work at the stores;
.
employees' perception of service quality in the store;
.
employees' perception of the physical environment in the store;
.
employees' satisfaction with the style of supervision in the store;
.
employees' perceived ability to affect store performance; and
.
employees' perceived ability to affect the customer experience.
The employee satisfaction questionnaire was also used to collect two further
measures of employee loyalty, namely, length of service and employees' stated
willingness to refer the store to friends and relatives as a place of work.
Correlation analysis
Measures of the performance of each store were collected, and each was then
tested against the rest in order to identify correlations. A null hypothesis was
developed that no correlation existed between the two variables involved in
each test. A significance test of 95 per cent was used, and only where Pearson's
correlation coefficient was greater than the minimum accepted at the 95 per
cent significance level was the null hypothesis rejected. The minimum value of
the correlation coefficient necessary to reject the null hypothesis was 0.81 given
IJOPM
22,1
36
the sample size of six stores, and 0.51 where the sample size was 15 stores
(Morris, 1993).
The main methodological limitation of this study is the small sample size.
Although the small number of stores could be said to be representative of the
region, the sample is too small to be statistically representative of the entire
retail chain. Nevertheless, the findings were not considered by the company's
management to be untypical of the company, and, as will be seen later, the
results were taken seriously by the managers. Indeed they were at pains to
interpret the results and did not attempt to ``explain away'' the findings on the
basis of the small sample size.
The study is also limited in terms of the choice of measures used to analyse
the relationships between employee satisfaction, loyalty and business
performance. Employee loyalty is particularly difficult to measure: Loveman
(1998) found that the relationship between employee loyalty and other aspects
of performance varied depending which measure was used. The difficulty of
obtaining single measures, which fully represent each aspect of service
performance, is one which has no ``magic'' solution, but the performance
measurement literature generally promotes the adoption of a range of measures
or surrogate indicators of each phenomenon, with a view to understanding and
managing service performance (Fitzgerald et al., 1991; Kaplan and Norton,
1997). In the absence of a single ``perfect'' measure of this complex social
phenomenon, three different measures of employee loyalty were used which
included both hard and soft measures.
Research findings
Tables I and II give the correlation coefficients for each of the pairs of variables.
Analysis of productivity and profit
It is first worth noting that there is a strong positive correlation (significant at
the 99.9 per cent level) between the three measures of productivity which
suggests that they are indicators of the same phenomenon and can be classified
together as productivity indicators. There was also a strong positive
correlation, again significant at the 99.9 per cent level, between each of the
productivity measures and store profit margin. This certainly makes intuitive
sense in that one would expect stores which are highly productive both in
terms of capacity utilisation (sales and contribution per square foot) and in
terms of labour productivity (contribution per staff hour) to be highly
profitable.
Analysis of employee satisfaction
The employee satisfaction results were striking; although it should be
recognised that they must be treated with caution, given the small sample of
stores analysed. Employee satisfaction was negatively and significantly
correlated both with store profit margin (refer to Figure 1), and with the labour
productivity indicator (contribution per staff hour, refer to Figure 2). The
Dispelling the
modern myth
37
correlation between employee satisfaction and the other productivity measures
(sales and contribution per square foot) was also negative, though not
statistically significant.
This of course calls into question the received wisdom emanating from
consultants and academics alike, that employee satisfaction is a profit driver.
Furthermore, the negative correlation with labour productivity could be
interpreted as an indication that employees are being over-stretched in the most
profitable stores in the retail network. Far from confirming the view advocated
Table II.
Correlation analysis
between employee
satisfaction and loyalty
Employee loyalty
Staff
turnover
Length of
service
Willingness to refer
store as place of work
Correlation
significance (R)
Employee satisfaction
a
0.1 0.57 0.61 0.81*
Notes:
a
data based on only six stores. *Significance level of 95%
Figure 1.
The correlation between
store profit margin and
employee satisfaction
Table I.
Correlation analysis
results
Employee
loyalty
Operating
ratio
Planned/ Productivity Correlation
Length of
service
actual
hours
Sales/sq
foot
Contribution/
sq foot
Contribution/
staff hour
Profit
margin
significance
(R)
Employee
satisfaction
a
0.57
a
0.76
a
(0.61)
a
(0.68)
a
(0.82)
a
(0.87)
a
0.81
*
Employee loyalty
a
0.65
a
(0.90)
a
(0.91)
a
(0.90)
a
(0.84)
a
Operating ratio (0.65) (0.71) (0.69) (0.75) 0.64
**
Sales/sq foot 0.97 0.79 0.77 0.76
***
Contribution
/sq foot 0.90 0.86
Contribution
/staff hr 0.97
Notes:
a
data based on only six stores; parentheses indicate negative correlation. Significance
level of:
*
95%;
**
99%; and
***
99.9%
IJOPM
22,1
38
by the company's CEO that customer satisfaction and profit are driven by
employee satisfaction, it would appear that profits are being achieved despite,
rather than because of, the nurturing of employee satisfaction.
Correlation tests were conducted between overall employee satisfaction and
each of the other employee perception measures collected as part of the
employee survey, with a view to identifying which of the particular aspects of
the employee's experience was linked to overall satisfaction. The results are
reported in Table III. Whilst there was no correlation between overall
satisfaction and employees' satisfaction with the physical environment in the
store, there was a positive correlation with each of the other aspects. However
only two of the correlations were statistically significant, namely, satisfaction
with the style of supervision within the store (significant at the 95 per cent
level), and employees' perceived ability to affect the customer experience
(significant at the 99.9 per cent level).
Interestingly, this does accord with the assumption in the management
literatures that empowered staff are better equipped to meet customer needs
and more likely to find job satisfaction. Moreover, both the TQM and HRM
literatures promote the view that empowerment is appropriate to improve the
customer experience and that good relationships between employees and
supervisors are conducive to employees' taking pride in their work and
achieving job satisfaction. The fact that the employees who were happiest with
working in the stores were also the most satisfied with their supervisors and
Figure 2.
The correlation between
store productivity and
employee satisfaction
Table III.
The relationship
between employee
satisfaction and other
employee perceptions
Correlation between overall satisfaction and the following: R value
Correlation
significance (R)
Employees' perception of service quality in the store
a
0.71 0.81*
Employees' perception of the physical environment in the store
a
0.27
Employees' satisfaction with the style of supervision in the store
a
0.82
Employees' perceived ability to affect store performance
a
0.72
Employees' perceived ability to affect the customer experience
a
0.93 0.91**
Notes:
a
data based on only six stores. Significance level of: *95%; and **99.9%
Dispelling the
modern myth
39
had a perceived ability to affect the customer experience, meshes well with this
management philosophy.
Analysis of employee loyalty
As was found by Loveman (1998) in his study, analysis of the relationship
between employee loyalty and the other performance measures was rather
more inconclusive. Table II indicates that there is a stronger positive link
between employee satisfaction and two of the measures of loyalty (length of
service and staff's stated willingness to refer the store as a place of work to
friends and colleagues), than there is with the other measure of employee
loyalty, namely, staff turnover; however, the former correlations are not
significant at the 95 per cent confidence level. Nevertheless the notion that the
more satisfied employees are more likely to refer the store and stay the longest,
accords with intuition and, partially, with the findings of Loveman (1998), who
found a positive correlation between employee satisfaction and employees'
stated loyalty (though not with tenure).
The only measure of employee loyalty which correlated significantly with
the hard measures of performance was length of service, as indicated in Table I.
There was a significant and negative correlation with profit and all the
productivity measures. Again we have a surprising result here which goes
against the service management rhetoric which deems that increasing
employee loyalty, in terms of length of service, should reduce costs (by, for
example, reducing recruitment and training costs) and hence improve
profitability (Reichheld, 1996). This study seems to suggest that, in this
organisation, those stores where employees are happiest and stay the longest,
achieve the lowest profits. This is contrary to Loveman's findings, for he found
a positive link between employee tenure and financial performance.
What these conflicting results suggest is that the link between employee
loyalty (measured in terms of length of service) and profit is at best more
complex than is suggested in the formulaic service management literature, and
that the relationship may well vary by operational context, and may, indeed,
vary on the basis of cost structure differences between services. Higher
recruitment and training costs, resulting from high employee turnover, may
only translate directly into the bottom line if employee related costs represent a
significant proportion of total costs. In this company store labour costs
typically represent only 7 per cent of sales, thus having a less significant
impact on profit than, for example, gross margins.
Analysis of the operating ratio
There were strong correlations between the operating ratio and all the profit
and productivity measures, significant at the 99 per cent confidence level. The
operating ratio is the ratio between planned and actual hours. If the ratio is
high, actual hours exceeds planned hours. This means that the short term sales
forecast is higher than originally planned, so that more hours are actually
worked than were planned for and in the short term store management recruit
IJOPM
22,1
40
extra staff to meet the actual required hours. Where the actual hours are
equivalent to planned hours the sales forecast and short term planned hours are
unchanged.
It appears that the operating ratio is positively, though not significantly,
correlated with employee satisfaction and employee loyalty (measured in terms
of length of service), but negatively and significantly correlated with all the
other hard measures of productivity and profit. That is to say, the higher the
operating ratio (i.e. the more actual hours exceed target), the lower the
productivity and profitability of the store. Again, this makes intuitive sense in
that one would expect the most efficient stores (i.e. those with low operating
ratios) to be the most productive and profitable.
Interviews with managers further suggested that manipulation of the
operating ratios by managers to maximise efficiency may be occurring at the
expense of employees. Employees are most pressurised in the stores where no
short term adjustment is made to planned hours. Managers who meet their
planned hours may not be adjusting their actual hours either because they are
less sensitive to short term changes in demand (i.e. they fail to adjust sales
forecasts appropriately), or because they are reluctant to respond to those
adjustments by changing planned hours accordingly. Interviews with
management certainly confirmed that the operating ratio was regarded as a
critical performance measure within the company and that there was pressure
on store managers not to exceed planned hours. This analysis would appear to
suggest that the company's focus on the operating ratio was leading to
dysfunctional behaviour which was resulting in stressful working
environments leading to employee dissatisfaction and shorter length of service.
We are reminded of the contributions of critical theorists who have argued that
TQM and JIT production environments can result in labour intensification and
the creation of more, rather than less, stressful places of work.
Explaining the findings
Caution is required in considering these findings because, despite their
statistical significance, the employee satisfaction measures were based on a
very small sample size. However, when the findings were discussed with the
company's management at head office and at store level, interestingly, they
were not at all surprised, and did not even attempt to ``explain away'' the
apparent anomalies in terms of sample size. Indeed they had heard anecdotal
evidence of this phenomenon from the other major retailers too.
Their explanation was that there was another key determinant of business
success which had not been considered in the analysis thus far, namely, store
size. In the large grocery retail chains, store performance tends to correlate with
store size: the larger stores being more profitable than the smaller ones.
However, employee satisfaction tends to be higher in the smaller stores. Indeed
the company's management were able to profile in some detail the key
differences between two store archetypes, referred to as ``achieving'' stores and
``coasting'' stores; these may be profiled as follows.
Dispelling the
modern myth
41
``Achieving stores'' are the large, highly profitable stores, serving high
volumes of customers. These stores have to cope with a high degree of
variability in terms of customer spend, known as basket size: they have large
numbers of shoppers loading large trolleys on the one hand and many ``top-up''
shoppers on the other; this makes these stores difficult to manage. As a result,
staff tend to have to work harder; they are more tightly supervised; mistakes
are more likely to be picked up on; the work place is more stressful.
Larger stores employ hundreds of staff so that it is more difficult to cultivate
team spirit. Store sites tend to be green field or in the city, so employees have to
travel to work. In doing so, they may well drive past ten other potential
employers; and there is likely to be direct competition from one or more of the
other large grocery retailers. Staff are therefore less loyal and tend to serve
shorter terms of service; labour turnover is higher.
``Coasting stores'' are not usually loss making but have lower turnover and
tend to be less profitable than the achieving stores. Customers' shopping
behaviour is less variable here, there being greater uniformity both in terms of
basket size and in terms of the amount of time customers spend in the store.
Customers therefore tend to be less impatient, easier to manage and generally
expect more interaction with staff, spending more time at the specialist
counters than would be typical in the large, fast-moving stores. The market
tends to be better defined, with less competition; and for employees there are
fewer other employment opportunities. However, these stores are less likely to
operate long, unsociable opening hours which are unpopular with staff. So in
these shops staff tend to work at a slower pace, there are more full-timers, and
staff stay much longer with the company (note that the negative correlation
between length of service and profitability demonstrated in Table I also
appears to support this view). It is therefore easier to cultivate team spirit, and
build on the sense of community.
Store size, then, seems to be a critical variable which needs to be taken into
account when considering the relationships between employee satisfaction and
loyalty, productivity and profit in this industry; relationships which appear to
be markedly different from the way in which they are characterised in the
management literatures. This implies that these relationships, which are so
often treated as universal, in fact vary according to operational context, and
may well vary between industries and businesses.
The distinction between the two store archetypes also fitted with the above
analysis of the relationship between the operating ratio, employee
satisfaction, productivity and profit. The high operating ratios in the less
profitable stores suggest that managers of those stores were more willing to
adjust their sales forecast and hence increase staff numbers in the short term,
at the expense of the operating ratio. This could be because short term
increases in demand would be easier to predict in the smaller stores than in
the larger, more volatile store environments. However, store managers'
behaviour could be supply, as well as demand, driven. It may be that because
employees are more satisfied and loyal in the smaller stores, they are more
IJOPM
22,1
42
flexible and more willing to increase their hours, or work extra shifts, than
employees in the larger stores. It may, therefore, be easier for managers to
adjust working hours at short notice in the smaller stores. Thus an
explanation for the reluctance of managers' of large stores to adjust planned
hours may be, not merely the difficulty of forecasting demand in the larger
stores, but also their inability to attract staff into the store at short notice.
This, combined with pressure from head office to keep down the operating
ratio, could well discourage managers in large stores from increasing their
actual working hours, despite the fact that this will make the stores more
stressful to work for the employees. Hence the correlation between the
operating ratio and employee satisfaction, not to mention the implication that
superstore managers may be pressurising the workforce to boost productivity
and short term profits.
The managers' explanation of the research findings in terms of the two store
archetypes provides an interesting and unusually coherent explanation of the
relationship between many different hard and soft measures of performance in
this particular operational context. It should, however, be emphasised that
what we have here is a set of hypotheses which require further testing. The
hypotheses, presented in Table IV, postulate the contingent relationships
between various aspects of service performance as they pertain in the
superstore industry.
This study has generated some limited evidence to the effect that employee
loyalty and employee satisfaction are, in some operational contexts, inversely
related to store profitability. A further research study is now under way to
further test the above hypotheses on a larger sample of stores. In the
meantime the store sizes of the 15 stores were obtained from the company to
conduct a preliminary test of hypotheses. The correlations between store size
and store profit, operating ratio and employee satisfaction are presented in
Table V and are statistically insignificant. However, the positive correlation
between store size and profitability, and the negative correlation between
store size and both operating ratio and employee satisfaction, are in line with
the hypotheses. Further testing on a larger sample of stores is now required
and the next phase of this research project is to further test the hypotheses in
Table IV.
Table IV.
Hypothesised
relationships between
store size, employee
satisfaction, loyalty
and profitability
Hypotheses
Hypothesis 1 Store size is positively correlated with store profitability
Hypothesis 2 Larger stores are more stressful working environments than smaller
stores
Hypothesis 3 It is easier to cultivate team spirit within smaller stores than larger stores
Hypothesis 4 Employee loyalty is higher in the smaller stores than in larger stores
Hypothesis 5 Large, profitable stores generate lower levels of employee satisfaction
than the smaller, less profitable stores
Dispelling the
modern myth
43
Implications for academics
It has been argued that the universal prescriptions of management consultants,
academics, and many of the TQM ``gurus'', are simplistic and insensitive to the
contingencies of varying operational contexts: ``TQM recommendations tend to
be context independent and, therefore, implicitly universal'' (Dean and Bowen,
1994). There is an urgent need for studies, such as Bowen and Lawler's (1992)
work on empowerment, which attempt to identify the contingencies which
render different management practices appropriate.
What the present, preliminary study suggests is that the relationship
between employee satisfaction and business performance is contingent upon
service context. Although this is a high contact service, in the sense that
customers spend relatively long periods of time in the service process (Chase,
1981), customer contact with employees is relatively low. The role of front
office staff focuses less on customer contact than on stock availability and
ensuring a smooth customer flow; even at the check-out customer priorities
tend to be speed and accuracy of service, rather than friendliness and
personal care. Indeed, in this industry there seems to be a general move away
from strategic differentiation based on staff/customer contact. This
company's introduction of self-scanning technology is very much in line with
this trend.
Perceptual data from this company's customers also seems to suggest that
the drivers of service value for customers are primarily product range and
value for money, rather than the interface with employees. So it may be
surmised that, when the interaction between customers and employees is not a
key driver of customer satisfaction, Heskett et al.'s (1997) ``mirror'' effect of
employee and customer satisfaction no longer applies. The disconcerting
implication here is that the dissatisfaction of highly productive workers may
not be translated in terms of customer dissatisfaction or indeed lowprofits.
The relationship between employee loyalty and profitability also appears to
be contingent upon operational context. Whilst the literature tends towards a
universal prescription which deems that reducing labour turnover will yield
increased profits, this study has shown that the increased recruitment and
training costs incurred through high levels of staff turnover, will not
necessarily translate directly into the bottom line, unless labour costs are a
significant proportion of total costs. Whilst labour costs in a superstore are an
important management concern because they are controllable at store level,
they do not impact upon the bottom line as significantly as do, for example,
gross margins and the degree of customer spend on high margin products. It is
Table V.
Correlation analysis
based on store size
Store profitability Operating ratio Employee satisfaction
Store size 0.35 (0.33) (0.44)
a
Correlation significance (R) 0.51 0.81
Note:
a
data based on only six stores; parentheses indicate negative correlation
IJOPM
22,1
44
therefore proposed that the nature of the relationship between employee loyalty
and profitability is contingent upon service cost structure.
This suggests four contingent variables which determine whether or not
there is a link between employee satisfaction, loyalty, service productivity and
profitability:
.
the level of contact between customers and staff;
.
the opportunity for technological substitution;
.
the criticality of staff/customer contact to the customer value
proposition; and
.
the cost structure of the service operation.
Thus, in services where there are high levels of contact between customers and
staff, few opportunities for technological substitution, staff/customer contact is
a critical aspect of service value to the customer, and labour costs represent a
significant proportion of total costs, then there may well be a direct link
between employee satisfaction, loyalty, unit productivity and profit. In other
service contexts, another set of performance levers have to be applied. For
example, in this network of superstores the drivers of profit appear to be store
size, product range, level of investment in terms of store refurbishment,
perceived value for money, and, as for most retailers, location. The critical
performance levers in this industry, then, may be effective management of
tangible assets and of the supply network to protect volumes and margins,
rather than maintaining a highly motivated workforce with high levels of
satisfaction.
This leads inevitably to a debate as to whether in low contact services
managers should, as Ishikawa (1985) maintains, concern themselves with
``employee happiness''; a debate which has clear ethical implications which go
beyond the scope of this paper. However, the point being made here is,
emphatically, not that employee satisfaction does not matter, but rather, that it
is no use pretending that profits cannot be achieved without happy employees.
This points again to the importance of taking a long term, rather than a short
term view of business survival. Just as cutting service levels may increase short
term profits but seriously jeopardise the long term future of a business, so
employee satisfaction may be critical to long term business success, even if it
does not ensure short term profits. Ironically we are reminded here of Deming's
(1986) rhetoric.
Implications for practitioners
This company's management training promotes the CEO's view that
employee satisfaction is a key driver of business success. Yet, interestingly,
the company's strategy of differentiation through the application of IT does
not in fact rely on the supposed link between employee satisfaction and profit;
and so the findings of this study do not call into question the company's
strategic orientation. This would appear to be a classic example of the
Dispelling the
modern myth
45
difference between explicit and implicit or emergent strategies (Mintzberg
and Waters, 1985): the explicit strategy being to compete on the basis of staff/
employee contact and the implicit strategy being to differentiate
technologically.
But there is a practical danger here that is relevant to all operations
managers who discern the absence of a direct link between employee
satisfaction and profit in their business. The danger in promoting the ``happy
productive worker'' concept to management and staff (in this case, through the
management training scheme) is that employees receive messages which do not
mesh with their own personal experiences. If the company teaches that
employee satisfaction will result in high profits, and employees perceive that
the business units where employees are happiest are in fact the lowest
performers, disenchantment with the company philosophy is bound to arise. It
is precisely the dysfunctional behaviour which results from employees' cynical
view of the company ``mantra'', that can lead to the ``fizzling'' of corporate
initiatives.
And what of the implication that store profits may be achieved at the
employees' expense? The large UK retailers are already being accused of
achieving high margins at the expense of their suppliers; is this study
indicative of their exploitation of employees? Certainly the company's
explanation of ``achieving'' stores stacks up well with the growing evidence that
companies associated with new wave manufacturing methods and excellence
are creating stressful and difficult working environments (see for example
Garrahan and Stewart, 1992, on Nissan). Can service excellence similarly be
achieved at the expense of creating a pleasant working environment for
employees?
Further study would be necessary in order to explore what makes for
employee satisfaction in this type of retail environment and how this is related
to store profitability. However, what this study has highlighted is the mismatch
between this company's stated management priorities and its measurement
systems. If we take on board Kaplan and Johnson's (1987) view that ``what is
measured is managed'', then the ready availability of profit and productivity
data on each store, the focus on the operating ratio and the dearth of employee
satisfaction measures found in this organisation, may well raise some concerns.
Despite the CEO's advocacy of concern for the well-being of employees, the
company's performance measurement systems focus heavily on financial
performance and productivity. And the dysfunctional effects on managerial
behaviour of such measurement systems are well documented (Kaplan and
Johnson, 1987).
In this company a strong focus on the operating ratio also appeared to be
leading to dysfunctional behaviour, where strict adherence to planned hours
and an inflexible approach to the manning of stores may have led to a stressful
working environment and staff dissatisfaction. Again, routine monitoring of
employee satisfaction might encourage more positive managerial behaviour,
and consideration of the long term welfare of employees as well short term
IJOPM
22,1
46
productivity and profit. This points towards Kaplan and Norton's (1997)
concept of a balanced business scorecard, where measurement of employee
satisfaction as a long term aim would counter the short termism associated
with a narrowfocus on financial performance.
This study, like other studies of the complex social phenomena of employee
satisfaction and loyalty (e.g. Loveman, 1998), reminds us of the importance of
selecting appropriate measures of performance. Employee loyalty may be
measured in a variety of ways and the relationship between loyalty and other
aspects of performance appears to vary depending which measure is being
used. In the absence of a single ``perfect'' measure of employee loyalty, the
answer here seems to be to select a range of measures, so as to reveal and
understand, the complexities of performance relationships.
Conclusions
This empirical study presents preliminary evidence which calls into question
key assumptions in the operations management, TQM, HRM and service
management literatures on the link between employee satisfaction and loyalty,
service productivity and profitability. The negative correlations demonstrated
here should not, however, be interpreted as disproof of those assumptions.
Rather what the study suggests is that the relationship between employee
satisfaction, loyalty and business performance is more complex than the
literature implies.
Two imperatives are suggested here. First, there is the academic imperative
to explore further, and in different service contexts, the links between employee
satisfaction, loyalty, productivity and profitability, in order to identify the
contingent variables which determine a positive link. In this paper four possible
contingent variables are proposed which distinguish those service contexts in
which employee satisfaction may be a driver of profitability: first, services
where staff contact is high; second, services where there is little opportunity for
technological substitution; third, services where staff contact is critical to the
customer value proposition; and fourth, services with high labour costs. It is
suggested that in services of this type, there is likely to be a direct link between
employee satisfaction, loyalty and unit profitability; whilst in other service
contexts, another set of performance levers may have to be applied.
These four service characteristics are very much aligned with the typology
of services which distinguishes between professional services, service shops
and mass services (Silvestro et al., 1992, Silvestro, 1999). On the volume-variety
continuum, professional services, defined as low volume, high contact and high
variety services which are people, rather than equipment based, would appear
to be those where employee satisfaction and loyalty can be expected to drive
service profitability. It might be that in mass services, and in service shops at
the lower end of the volume-variety continuum, the link between these
phenomena needs to be re-investigated and re-evaluated; for the management
literature appears to fall short of explaining the drivers of profitability in these
service environments.
Dispelling the
modern myth
47
The second prescription calls upon practitioners to measure both
profitability and employee satisfaction at unit level, in order to ascertain
whether the link does pertain in their particular operations. Fifteen years ago it
was relatively unusual to find service companies regularly measuring customer
satisfaction; this is no longer the case. Yet few companies today routinely
monitor employee satisfaction. It is proposed that in future this will also be
regarded as a key indicator which is measured with a view to deterring a short
term managerial focus on financial performance and productivity.
The key message here is for practising managers not to take it from the
management consultants and academics as gospel that employee satisfaction
drives profit, but to take the initiative to do the analysis in their own
organisations. Just as Bowen and Lawler (1992) argue that ``before service
organisations rush into empowerment programs, they need to determine
whether and how empowerment fits their situation'', so too senior managers
need to understand the dynamics of the relationship between employee
satisfaction, loyalty and financial performance in their own organisations. This
paper has outlined a methodology, using simple correlation analysis, which can
prompt and facilitate preliminary investigation of the issues.
If, having conducted the analysis in one's own organisation, employee
satisfaction and profitability are found to be positively correlated, then the sort
of management philosophy promoted by the CEO of this superstore company
can be bolstered with empirical evidence which will be convincing to managers
and employees. That is to say, senior management will be in a position to
demonstrate to its people that the most profitable units are those in which
employees are happiest. If, on the other hand, no direct link is apparent (or, as in
this case, the correlation is negative) then there is a need to delve further in
order to understand the real drivers of financial performance for the business.
And whilst it may still be desirable to maintain high levels of employee
satisfaction, the na ve notion that improving their satisfaction will directly
translate into profits will be dispelled, and senior management will not be
caught in a situation where the company philosophy is met with scepticism
from a disbelieving workforce.
References
Bowen, D.E. and Lawler, E.E.E. (1992), ``The empowerment of service workers: what, why, how
and when'', Sloan Management Review, Vol. 33 No. 3, Spring, pp. 31-9.
Carlzon, J. (1987), Moments of Truth, Ballinger Publishing Co., Cambridge.
Chase, R.B. (1981), ``The customer contact approach to services: theoretical bases and practical
extensions'', Operations Research, Vol. 29 No. 4, pp. 698-706.
Dean, J.W. Jr and Bowen, D.E. (1994), ``Management theory and total quality: improving research
and practice through theory development'', Academy of Management Review, Vol. 19 No. 3,
pp. 392-418.
Delbridge, R. and Turnbull, P. (1992), ``Human resource maximisation: the management of labour
under just-in-time manufacturing systems'', in Blyton, P. and Turnbull, P. (Eds),
Reassessing Human Resource Management, Sage, London, pp. 56-73.
IJOPM
22,1
48
Deming, W.E. (1986), Out of the Crisis, Center for Advanced Engineering Study, Massachusetts
Institute of Technology, Cambridge.
Fitzgerald, L., Johnston, R., Brignall, T.J., Silvestro, R. and Voss, C. (1991), Performance
Measurement in Service Businesses, Chartered Institute of Management Accountants.
Garrahan, P. and Stewart, P. (1992), The Nissan Enigma: Flexibility at Work in a Local Economy,
Mansell, London.
Guest, D.E. (1987), ``Human resource management and industrial relations'', Journal of
Management Studies, Vol. 24 No. 5, pp. 503-21.
Heskett, J.L., Sasser, W.E. Jr and Schlesinger, L.A. (1997), The Service Profit Chain: How Leading
Companies Link Profit and Growth to Loyalty, Satisfaction and Value, Free Press, New
York, NY.
Hostage, G.M. (1975), ``Quality control in a service business'', Harvard Business Review,
July-August, pp. 98-106.
Ishikawa, K. (1985), What is Total Quality Control? The Japanese Way, Prentice Hall, Englewood
Cliffs, NJ.
Juran, J.M. (1989), Juran on Leadership for Quality: An Executive Handbook, The Free Press, New
York, NY.
Kaplan, R.S. and Johnson, H.T. (1987), Relevance Lost, Harvard Business School, Boston, MA.
Kaplan, R.S. and Norton, D.P. (1997), The Balanced Business Scorecard: Translating Strategy into
Action, Harvard Business School Press, Boston, MA.
Kerfoot, D. and Knights, D. (1995), ``Empowering the quality worker? The seduction and
contradiction of the total quality phenomenon'', in Wilkinson, A. and Willmot, H. (Eds),
Making Quality Critical, Routledge, London.
Ledford, G.E. Jr (1999), ``Happiness and productivity revisited'', Journal of Organizational
Behaviour, Vol. 20, pp. 25-30.
Legge, K. (1995), Human Resource Management: Rhetorics and Realities, Macmillan Business,
London.
Loveman, G.W. (1998), ``Employee satisfaction, customer loyalty, and financial performance: an
empirical examination of the service profit chain in retail banking'', Journal of Service
Research, Vol. 1 No. 1, pp. 18-31.
McArdle, L., Rowlinson, M., Procter, S., Hassard, J. and Forrester, P. (1995), ``Total quality
management and participation: employee empowerment or the enhancement of
exploitation'', Wilkinson, A. and Willmott, H. (Eds), Making Quality Critical: New
Perspectives on Organisational Change, Routledge, London.
Mintzberg, H. and Waters, D. (1985), ``Of strategies, deliberate and emergent'', Strategic
Management Journal, Vol. 6, pp. 257-72.
Morris, C. (1993), Quantitative Approaches in Business Studies, Pitman, London.
Nuki, P. and Bevan, S. (1998), ``Stores mark up `bargains' by 60 per cent'', The Sunday Times,
8 November, p. 4.
Perrow, C. (1986), Complex Organizations, 3rd ed., Random House, NewYork, NY.
Reichheld, F.F. (1996), Loyalty Effect: The Hidden Force behind Growth, Profits and Lasting
Value, Bain and Co., Boston, MA.
Reichheld, F.F. and Sasser, W.E. Jr (1990), ``Zero defections: quality comes to services'', Harvard
Business Review, September-October, pp. 105-11.
Rucci, A.J., Kirn, S.P. and Quinn, R.T. (1998), ``The employee-customer-profit chain at Sears'',
Harvard Business Review, January-February, pp. 83-97.
Dispelling the
modern myth
49
Rust, R., Zahorik, A.J. and Keiningham, T.L. (1995), ``Return on quality (ROQ): making service
quality financially accountable'', Journal of Marketing, Vol. 59, pp. 58-70.
Schlesinger, L.A. and Heskett, J.L. (1991), ``Breaking the cycle of failure in services'', Sloan
Management Review, Spring, pp. 17-28.
Schneider, B. and Bowen, D. (1993), ``Human resource management is critical'', Organisational
Dynamics, pp. 39-52.
Shimada, H. (1993), ``Japanese management of auto production in the US: an overview of ``human
technology'', Lean Production and Beyond, International Labour Office, Geneva.
Silvestro, R. (1999), ``Positioning services along the volume-variety diagonal: the contingencies of
service design, control and improvement'', International Journal of Operations &
Production Management, Vol. 19 No. 4, pp. 399-421.
Silvestro, R., Fitzgerald, L., Johnston, R. and Voss, C. (1992), ``Towards a classification of service
processes'', International Journal of Service Industries Management, Vol. 3 No. 3, pp. 2-75.
Turnbull, P. (1988), ``The limits to Japanization just-in-time, labour relations and the UK
automotive industry'', New Technology, Work and Employment, Vol. 3 No. 1, pp. 7-20.
Walton, R.E. (1985), ``Toward a strategy of eliciting employee commitment based on policies of
mutuality'', in Walton, R.E. and Lawrence, P.R. (Eds), Human Resource Management,
Trends and Challenges, Harvard Business School Press, Boston, MA, pp. 35-65.
Zeithaml, V.A., Berry, L.L. and Parasuraman, A. (1996), ``The behavioral consequences of service
quality'', Journal of Marketing, Vol. 60, pp. 31-46.