The customer is not always right: customer aggression and emotion regulation of

service employees

Meaning
The trading policy that states a company's keenness to be seen to put the customer first.

Origin

Several retail concerns used this as a slogan from the early 20th
century onward. In the USA it is particularly associated with Marshall Field's department
store, Chicago (established in the late 19th century). The store is an icon of the city,
although it is set to lose its name in 2006 when, following a takeover, it becomes
renamed as Macy's. In the UK, Harry Gordon Self ridge (1857-1947) the founder of
London's Selfridges store (opened in 1909), is credited with championing its use. The
Wisconsin born Self ridge worked for Field from 1879 to 1901. Both men were dynamic
and creative businessmen and it's highly likely that one of them coined the phrase,
although we don't know which. Of course, these entrepreneurs didn't intend to be taken
literally. What they were attempting to do was to make the customer feel special by
inculcating into their staff the disposition to behave as if the customer was right, even
when they weren’t. The trading policy and the phrase were well-known by the early 20th
century. From the Kansas City Star, January 1911 we have a piece about a local country
store that was modeled on Field's/Selfridges:[George E.] "Scott has done in the country
what Marshall Field did in Chicago; Wannamaker did in New York and Self ridge in
London. In his store he follows the Field rule and assumes that the customer is always
right. “Whether the phrase was coined by Field or Self ridge it is fair to call it American.
What we can't do is credit them with the idea behind it. In 1908 Cesar Ritz (1850-1918),
the celebrated French hotelier is credited with saying 'Le client n'a jam is tort' - 'The
customer is never wrong'. That's not the phrase that people now remember, but it can
hardly be said to be any different in meaning to 'the customer is always right'.

Introduction
According to recent researchers and the popular press, people are less able to treat each
other in a courteous and respectful way at work (Andersson & Pearson, 1999; Daw, 2001;
Grimsley, 1998; Marks, 1996). In the study of this phenomenon, work behaviors such as
yelling, rudeness, and threats have been studied under multiple labels, including
interactional injustice (Bies & Moag, 1986), work aggression (LeBlanc & Kelloway,
2002), workplace bullying (Leyman, 1996), incivility (Andersson & Pearson, 1999), and
interpersonal conflict (Spector & Jex, 1998). In general, these behaviors are related to
deleterious work reactions such as stress and health problems, retaliatory behaviors, and
turnover, making it a critical topic for study. The theory, research, and scale development
on these forms of work aggression has generally focused on intra-organizational
members as the source and targets of these behaviors (e.g., Andersson & Pearson, 1999;
Donovan, Drasgow, & Munson, 1998; Frone, 2000; Spector & Jex, 1998). A missing
piece from the existing literature is the recognition that such antisocial behaviors may
come from the very people the organization is trying to help, namely, the customers.
Furthermore, researchers have been more likely to study rare violent episodes from ex
employees or the public (e.g., Barling, Rogers, & Kelloway, 2001) than the effects of the
more common verbally abusive customer. In fact, aggression from customers is critical to
understand due to the potential frequency and challenge of responding to aggression from
this source (Grandey & Brauburger, 2002; Maslach, 1978). In this paper we extend the
literature on work aggression by: (1) outlining reasons why customer aggression should
be studied in its own right; (2) examining psychological and behavioral consequences
of employees’ experiences with customer aggression; and (3) exploring how employees
regregulate emotions when dealing with aggressive customers.
Research on work aggression or anger has typically focused on supervisors and co-
workers as the instigators of aggression; however, aggressive customers are also likely
and may have unique consequences for the employee. We explore this phenomenon with
a sample of 198 call center employees at two work sites. The employees reported that
customer verbal aggression occurred 10 times a day, on average, though this varied by
race and negative affectivity. Using LISREL, our data indicated that both the frequency
and stress appraisal of customer aggression positively related to emotional exhaustion,
and this burnout dimension mediated the relationship of stress appraisal with absences.
Stress appraisal also influenced employees ‘emotion regulation strategies with their most
recent hostile caller. Employees who felt more teatened by customer aggression used
surface acting or vented emotions, while those who events more stressful, and
implications of these results are discussed.

Defining Boundaries: Customer Verbal Aggression

The increased attention to the ‘dark side’ of workplace behavior in the last decade has led
to the need to disentangle any similar concepts (Glomb, 2002; O’Leary-KellyQ2, Duffy,
& Griffin, 2000; Neuman&Baron, 1998).

Verbal aggression
We wanted to focus on behaviors that occur frequently, rather than extreme or rare
situations. Though intense events like physical violence are certainly highly stressful, the
more mundane daily hassles have also been implicated as causes of stress and health
symptoms by themselves (Lazarus, 1984).Since verbal forms of aggression (e.g., yelling,
insults, and cursing) are the most frequently experienced forms of aggressive behaviors
(Barling et al., 2001; Diaz & McMillin, 1991; Glomb, 2002),we were particularly
interested in these types of behavior. Several terms are related to our behaviors of
interest, though none are a perfect fit. Interactionaljustice (Bies & Moag, 1986) refers to
perceptions about how one is treated during decision-making or allocation processes,
typically in terms of rudeness from intra-organizational members (Donovanetal., 1998).
The term interpersonal conflict refers to negative treatment that may be verbal or
behavioral (e.g., being rude or ‘nasty;’ Spector & Jex, 1998). Workplace bullying is
defined as persistent and long-term exposure to psychologically aggressive behaviors in
which the targets have difficulty defending themselves (Leyman, 1996; Rayner, Hoel, &
Cooper, 2002), typically referring to repeated interactions with the same people. Incivility
refers to behaviors that violate interpersonal norms with ambiguous intentions, such as
rudeness or ignoring someone (Andersson & Pearson, 1999), and its uniqueness revolves
around its ambiguous intentions. In this study, we focus on a behavioral term that refers
to verbal communications of anger that violate social norms: verbal aggression, or
hostility (Glomb, 2002; Neuman & Baron, 1998). This term is clear about the behaviors
of interest, avoids the fuzzy boundaries associated with the different terms noted
previously, and is not necessarily associated with intra-organizational members as the
previous terms have been. Consistent with this variable of interest, in the current study
we focus on call center employees who only engage in voice-to-voice service
interactions, allowing verbal aggression to be the primary method of communicating
anger or hostility.

Customer verbal aggression

Why would customers as a source of aggression be important to study? First, customer
aggression is likely to be a common hassle of work life for boundary-spanners who are
the intermediaries between the company and the public. Such a job requires high
frequency of interpersonal contact, a characteristic associated with the frequency of work
aggression (LeBlanc & Kelloway, 2002).Furthermore, the mantra ‘the customer is always
right’ communicates the unequal power in the customer–employee transaction, which is
also a key aspect of being a target of aggression (Allan & Gilbert, 2002; Hochschild,
1983). Research has supported, though not focused on, the occurrence of verbal
aggression from customers. For example, Spratlen (1995) found that 21 per cent of
university staff reported an interpersonal mistreatment event from someone categorized
as ‘other’ (i.e., not a co-worker, supervisor, or subordinate). In a diary study of part-time
service workers, the majority of anger-inducing interpersonal events reported over a 2-
week period were due to customers (43 per cent), and most of these were verbal
aggression behaviors (Grandey, Tam, &Brauburger, 2002).Second, the experience of
customer aggression may have unique consequences. With the growth of the service
economy, there are high expectations on boundary-spanners to maintain customer
satisfaction (Bitner, Booms, & Tretrault, 1990). Supervisors are often monitoring service
employees to ensure organizational control over emotional expressions with ‘service with
a smile’ as the goal (Rafaeli &Sutton, 1987)Q3. If customers do verbally attack
employees, a ‘spiral of incivility’ (Andersson &Pearson, 1999) may occur where the
hostile customer arouses anger in the employee, who must regulate his or her response or
risk venting at the customer. Thus customer verbal aggression maybe particularly
stressful for the employee and problematic for the organization.

A Model of Customer Verbal Aggression

Following the lead of other work aggression researchers, (Rogers & Kelloway, 1997;
Schat &Kelloway, 2000), we drew on the traditional stressor–stress–strain framework for
our hypotheses (Kahn & Byosiere, 1992; Lazarus, Delongis, Folkman, & Gruen, 1985).
In this model, objective stressors in the environment evoke cognitive appraisals about the
situation and coping responses. The stress appraisal of the situation induces a strong state
of emotional and physiological arousal. Chronic levels of this state can result in
psychological and behavioral signs of strain. In our model, the frequency of customer
aggression is the work stressor, and we were interested in the extent that employees feel
Threatened by customer aggression (stress appraisal), while emotional exhaustion and
absences represented psychological and behavioral forms of strain, respectively.
Customer verbal aggression and stress appraisal Stress appraisal refers to the individual’s
perception of a stressor as being stressful or threatening(Lazarus & Folkman, 1984),
typically accompanied by a negative emotion (e.g., fear, anger) and physiological arousal.
Given our human discomfort with being the target of anger, and the fact that this behavior
communicates that the goal of satisfied customers is not being met, it is likely that verbal
aggression is highly stressful (Averill, 1983). The more often such an event occurs, the
more likely that the event may be found to be stressful due to heightened states of arousal
and apprehension. For instance, prior research has found that targets of frequent
aggression are likely to experience more fear at work (Barling et al., 2001; LeBlanc &
Kelloway, 2002). Those who regularly interact with hostile customers may be ‘primed’
for the next call to be another yelling person. It should be acknowledged that an alternate
possibility is that higher frequencies may induce habituation to the negative stimuli;
however, people seem to be hard-wired to react strongly to such potentially threatening
events (Frijda, 1988). Therefore, we expected that a higher frequency of verbally
aggressive customer is likely to be associated with a stronger appraisal of threat from
such an event.

Negative affectivity and customer aggression

An individual difference that is likely to contribute to increased frequency and stress
appraisal of customer aggression is the negative affectivity (NA) of the service provider.
Surprisingly, NA’s role in self-reported work aggression has been omitted from recent
research (Barling et al., 2001; LeBlanc& Kelloway, 2002; Schat & Kelloway, 2000). A
person high in NA is more likely to have a negative world-view and to interpret
ambiguous comments as negative (Spector, Chen, & O’Connell, 2000; Watson & Clark,
1984; Watson, Clark, & Tellegen, 1988). These tendencies may mean high-NA persons
evoke more aggression from a customer (Buss, 1987). For example, a customer may
make an ambiguous comment and someone with high NA may respond more
defensively, evoking hostility in the customer. Similarly, someone high in NA has fewer
coping resources and thus may perceive the event as more threatening (Spector et al.,
2000). Therefore, we expected NA to correspond with both the reported frequency and
stress appraisal of hostile customers.

Autonomy and customer aggression

Call center employees may objectively have low levels of job autonomy (e.g., calls are
monitored, encounters are scripted), but differences in their perceptions of control at
work should influence how stressful they find customer aggression. Stress research has
long pointed out the importance of perceived sense of control on stress reactions to
stimuli (Folkman, Lazarus, Gruen, & DeLongis, 1986; Friedland, Keinan, & Regev,
1992); similarly, a sense of autonomy has been associated with less Who feel that they
have control at work should feel more empowered in customer encounters, including
aggressive ones, and research has supported job autonomy’s relationship with the stress
response to work aggression

Customer verbal aggression and emotional exhaustion

Burnout has long been established as a likely outcome for service workers and caring
professionals, due to the boundary-spanning nature of the job tasks (Singh, Goolsby, &
Rhoades, 1994). Being the target of frequent hostility from the same people for whom
you are supposed to be providing ‘service with a smile’ requires constant emotion
regulation that may eventually deplete one’s resources (Brotheridge & Grandey, 2002;
Hochschild, 1983). The experience of verbally hostile customers may thus ‘take an
emotional toll’ (Neuman & Baron, 1998, p. 397), resulting in a state of burnout (Deery,
Iverson, &Walsh, 2002; Maslach, 1978). LeBlanc and Kelloway (2002) did not find that
‘public aggression’ predicted health outcomes beyond other work factors, though their
measure included violent (less common) behaviors as well. We expected that more
frequent interactions with verbally aggressive customers would relate to emotional
exhaustion. Stress emotions have a unique effect on health and organizational outcomes
beyond the frequency of work aggression (e.g., Schat & Kelloway, 2000), and thus may
explain the relationship between the frequency of events and outcomes (Lazarus et al.,
1985). As the frequency of customer aggression increases, so should the resulting state of
burnout due to the enhanced states of stress arousal. We assessed these relationships with
emotional exhaustion while allowing negative affectivity to also predict emotional
exhaustion. Since NA has also been found to be a strong predictor of
emotional exhaustion (e.g., Brotheridge & Grandey, 2002), we conduct a conservative
test of our variables of interest on emotional exhaustion by controlling for NA (Spector,
1994).Regular hostility from customers creates an unpleasant working environment that
employees may seek to avoid whenever possible. Withdrawal is a form of mood control
—regulating emotions by avoiding situations that create negative mood states (George,
1989; Hackett, Bycio, & Guion, 1989). In support of this, Hackett et al. (1989) found that
a ‘mental health day’ was almost as common a reason for absence as minor illness among
nurses. Surprisingly, frequency of conflict has a weak relationship withself-reported
withdrawal behavior (Donovan et al., 1998; Spector & Jex, 1998), and no studies were
found that linked the frequency of verbal aggression to actual absences. We expect that
the frequency of customer aggression predicts hours absent from the workplace as a form
of avoidance. Stress appraisal of customer aggression may result in absences as a form of
mood regulation—high appraisal of work events as stressful may mean more arousal and
negative emotions at work and thus the desire to remove oneself to stabilize internal
arousal (George, 1989).Another explanatory mechanism for why customer aggression
may relate to absences is through its influence on burnout. There are costs to avoiding
work, and the desire for mood regulation or avoidance may not be strong enough
motivation to be absent (Harrison & Martocchio, 1998). However, if
customer aggression predicts an employee’s state of burnout, then they may need a day
off from work to restore lost resources (Hobfoll & Freedy, 1993; Muraven & Baumeister,
2000). A few studies have shown a link between burnout and work absences (Firth &
Britton, 1989; Saxton, Phillips, & Blakeney, 1991). Another study with call center
employees found only a marginal relationship of burnout to absences administration,
making interpretation of the relationships difficult. Our study tested a mediation
relationship of emotional exhaustion between customer verbal aggression and subsequent
absences.
.
Emotion regulation with an aggressive customer
Lazarus et al.’s model of stress also proposes that emotion regulation, or emotion-focused
coping, follows a stress appraisal (Lazarus et al., 1985). Though employees may be
trained in how to do problemfocusedcoping (i.e., how to technically help upset
customers), there may be uncertainty in how to handle their internal emotions with
customers. We were interested in how call center employees regulate emotions in
response to customer aggression, given the demands for friendly service. In our
study, a specific event is used to cue reports of regulation rather than asking about
general responses, as suggested by Lazarus (2000). Since this information is at the acute,
rather than chronic, level of analysis, it was analyzed separately from the hypothesized
model above. Two general ways that service workers can emotionally regulate is through
surface acting (i.e., engaging in behavioral change) and deep acting (i.e., engaging in
cognitive change) (Grandey, 2000; Hochschild, 1983). Surface acting entails modifying
behaviors by suppressing or faking expressions (Brotheridge & Lee, 2002; Grandey,
2000). Deep acting refers to internal change: changing cognitions through perspective
taking (reappraising the situation by taking another’s point of view) or positive refocus
(focusing attention on positive things to regulate feelings) (Gross, 1998). We also
examined venting emotions—a dysfunctional response to customers (Bitner et al., 1990;
Grandey,2003) that happens when the individual does not regulate emotions. Surface
acting has been linked to burnout and lower service performance, while deep acting has
been positively related to serviceperformance (Brotheridge & Grandey, 2002; Grandey,
2003; Totter dell & Holman, 2003). Thus, the type of regulation matters. A person’s
stress appraisal (e.g., ‘How threatening is this to me?’) is considered an antecedent of
coping strategies (Lazarus et al., 1985; Lazarus, 2000). Surface acting is likely to be
frequently used during customer aggression by employees who appraise hostile
customers as very stressful, because the negative emotions will need to be masked when
interacting with these customers (Grandey, 2000; Hochschild, 1983). In experience-
sampling studies, service employees responded to highly negative events with surface
acting (Grandey et al., 2002; Totter dell and Holman, 2003). Those employees who
appraise aggressive callers as highly stressful may also be more likely to vent their
frustration than those who find them only mildly stressful; intense emotions may only be
‘bottled up’ for a limited time. When negative events are viewed as less stressful, deep
acting is more likely to be reported;attentional resources are more available for modifying
cognitions about the situation (Gross, 1998).

Organizational Context

Service Industry
With the continued growth of the service sector in the United States, it has become
increasingly important for for-profit organizations to differentiate themselves from the
competition by going to great lengths to accommodate customer demands and maximize
service efficiency. The rise of Internet technology has enabled self-service in many
sectors of the economy, while also exacerbating the demands for efficient and customer-
centric exchanges of information during face-to-face and telephone service encounters.
Providers of service in the United States are expected to be swift, thorough, and
accommodating. Customers, however, are frequently impatient, angry, and
argumentative.

The Rise of Call Centers

Call centers are a multi-billion dollar industry in the United States. Customer
representatives are the’ face’ of the organization displayed one-on-one to consumers over
the telephone, and are essential to many organizations, even in an Internet economy, for
exchanging information, generating revenue and ensuring customer satisfaction and
retention. Call center employees usually have the opportunity to interact with a diverse,
geographically large population, and to provide services and support to large numbers of
individuals each day. Although satisfying customers can provide substantial intrinsic
rewards, the work often takes the form of an assembly line of telephone calls, frequently
with customers who are calling to resolve a problem. Difficult or angry customers are
Encountered often, and absenteeism and burnout are common.

Employees
Study participants were customer service associates who worked full-time and had long
tenures with the organization. They spent all of their working time on the phone with
customers, while finding and recording account information at a computer in a small
space surrounded by other employees. The employees were union-represented and could
bring issues or concerns to a local union representative. Employees answer inbound
telephone inquiries about account services and billing, working atone of two large call
centers in a large, successful for-profit services organization. High accountability for
customer service quality is the norm. Supervising managers can monitor any call, lengths
and numbers of calls are generally tracked and monitored, and continued customer
satisfaction with service representatives is essential to the company’s brand.

Time
The study took place in mid 2001, during a fairly strong economy with low
unemployment and many job opportunities in the U.S. labor market. The study was
completed prior to the decline of corporate revenues in the technology sector, the Internet
sector and elsewhere, and prior to concerns about corporate health associated with
accounting practices and terrorism. There were no substantial changes in the company
workforce (e.g., lay-offs, promotions) prior to the study.

Call center context

As is typical of call centers, the service encounters involved frequent monitoring, strict
emphasis on efficiency and timeliness (e.g., a risk of poorer performance evaluations for
being 1 minute late from a break), and the expectation to field calls in rapid succession.
The service representatives answered calls from small workstations situated closely
together and within sight of the supervisor, who would routinely listen to calls and
evaluate the subordinates’ performance (e.g., how well they worked with difficult
Customers and how effectively they solved customers’ problems). Other sources of
evaluation included ratings of live and recorded calls from: (i) quality monitors, who
specialized in quality control and performance assessment; and (ii) customer surveys. All
of these evaluations affected the official appraisal rating that a customer service
representative received. Management wished to learn about and improve the call center
employees’ work experience, while retaining the overall culture of monitoring
and efficiency that is typical of the industry and a key element in customer satisfaction.
Interview participants and procedures To assess the appropriateness of a study on
customer aggression with this sample, the first author conducted semi-structured
interviews (20–30 minutes) by phone with 12 customer service representatives
from the target organization. They had been selected from the work floor by a site
supervisor with the goal of having a wide range of demographics and experiences
represented. Eight individuals (67 per cent) were female. The interviewees had been with
this organization for an average of 4 years (range¼2–9 years).

All 12 interviewees said that yelling, threatening, and insulting behaviors from callers
occurred daily, and their description of events was used in the content of the survey. The
frequency varied by interviewee from one a day, to 20 a day, to ‘a majority of our calls’
(50–60 calls per day is typical).When asked about the most stressful part of their job, half
(six) of the interviewees said hostile, difficult, or angry customers. In fact, one female
employee told us that she was quitting that day because she ‘just can’t take it anymore.’
Low autonomy was another commonly mentioned stressor. When asked how they
responded to an aggressive caller, they had a variety of responses, including distracting
themselves with pictures or desk toys, putting the caller on mute and talking to co-
workers, and just recognizing that the caller is mad at the company and not the employee.
Two admitted to ‘losing it’ and yelling at a customer. Four mentioned that the only
training was being told ‘Don’t TIP,’ meaning’ Don’t Take It Personal.’ Three stated that
there was a need for training sessions for handling aggressive callers.

Implications
Future researchers who are examining aggression from supervisors and co-workers
should measure its occurrence from customers as well and examine the comparative
stress of aggression from intra- and extra-organizational sources. Helping boundary-
spanners feel like they have control in their jobs may aid in decreasing the stress of
abusive customers. This may be as simple as telling employees they have the freedom to
take a break if they need one after a rough customer or to tell customers when they have
crossed the line. When customer aggression was viewed as less stressful, emotional
exhaustion was lower, and there were subsequently fewer absences, thus related to call
centers’ bottom line. Furthermore, engaging in deep acting was also associated with
lower stress appraisal. The results of this study demonstrate that hostile customers are
part of work life for boundary-spanners. Management is encouraged to enhance the sense
of job autonomy for service representatives so that such events are less stressful, and
provide employees with training in emotion regulation for responding to customers who
are not always right.