Ethnic gatekeeping on the shopfloor: a study of bases,motives and approaches

Introduction

A common criticism of studies of discrimination and exclusion in organizational settings is that they fail adequately to document the experiences of employees, especially those from BME groups (see review by Proudford and Nkomo, 2006). Scholars are increasingly addressing this concern by providing empirical studies that analyse different aspects of the work experiences of BME communities (e.g. Kamenou and Fearfull, 2006; Ogbonna and Harris, 2006; Van Laer and Janssens, 2011). The general conclusion of these studies is that discrimination remains a pervasive and deleterious aspect of the work experience of BME groups. A number of these contributions also suggest that there has been a notable shift in the type of discrimination and they posit that direct, overt racism has been replaced by subtle forms of discrimination that are, to some extent, rooted in everyday work experiences (see also Cortina, 2008; Deitch et al., 2003; Essed, 1991). Consequently, these scholars call for additional research to generate further understanding of different forms of such covert discrimination (see Cortina, 2008; Deitch et al., 2003; Van Laer and Janssens, 2011). These studies are set against research contributions that continue to report the persistence of overt discrimination (e.g. McGinnity and Lunn, 2011), to the extent that existing research can be characterized as binary, with BME discrimination commonly viewed as either covert or overt.

Further, while researchers have made advances in examining discrimination at institutional/organizational levels (see Kamenou et al., 2013; McGinnity and Lunn, 2011) and at individual/group levels (e.g. Kamenou and Fearfull, 2006; Ogbonna and Harris, 2006), limited attention has been paid to understanding the interactions and combined impacts across the levels. To gain additional insights, it is important to explore discrimination from multiple and integrated perspectives involving both victims and potential perpetrators. This not only requires studying discrimination at the individual/group level but also necessitates an understanding of the organizational context in which discrimination occurs. Thus, although the main concern of this article is the role of shopfloor workers, the empirical data extends to the managerial and organizational context to facilitate a broader understanding of how discrimination is linked across different levels.

The primary focus on shopfloor employees reflects the important role they play in attracting, socializing and retaining new and existing workers (see Harris and Ogbonna, 2013). Thus, any discrimination at this level is likely to have a significant effect on the participation and progress of BME groups. In order to elucidate issues of shopfloor discrimination, it is important to understand the role of white majority shopfloor workers in their interaction with existing and potential BME counterparts. In this regard, this article builds on the suggestion of Riordan et al. (2005) and others who argue that scholars should develop a fuller appreciation of the explanatory power of relational demography (how relationships are influenced by individual and group identity construction on the basis of shared demographic attributes) in exploring the dynamics between different ethnic groups in organizations (see also Avery et al., 2008; Park and Westphal, 2013). The key questions that this article addresses are whether (and if so how) shopfloor workers (individuals and groups) engage in acts of overt or covert discrimination, and whether any such acts of discrimination are encouraged or facilitated by management or by organizational practices. To frame the discussion, insights are drawn from gatekeeping theory, which explores how individuals and groups influence decisions regarding who or what is desirable (and thus include) or undesirable (and thus potentially exclude).

The aim of this article is to provide an empirical exploration of the role of shopfloor workers and management/organizational context in relation to gatekeeping and ethnic discrimination. To achieve this aim, the study explores the nature of social relationships between demographically similar groups (in this case white employees) and demographically dissimilar groups (various BME groups) in order to uncover whether gatekeeping occurs, and, if so, to document the bases, motives, approaches and the organizational/managerial context in which this occurs. The findings of the study suggest that shopfloor workers routinely engage in ethnic gatekeeping which, data analysis suggests, are discriminatory behaviours that are neither fully overt nor entirely covert. Such discrimination occurs with the knowledge of managers and supervisors who appear tacitly to approve the disregard of the organizational policies and processes that are designed to eliminate disadvantage.

Discrimination and disadvantage of BME groups

A vast number of contributions report the disadvantages of BME groups in gaining access to employment and in working in desirable roles (e.g. Heath and Li, 2010; McGinnity and Lunn, 2011). Recent studies highlight three important themes. First, there is a divergence of scholarly interest, with some researchers focusing on the persistence of direct, overt forms of discrimination (e.g. McGinnity and Lunn, 2011), while others call for a shift to covert racism to reflect what they view as modern forms of discrimination (e.g. Cortina, 2008; Van Laer and Janssens, 2011). The second theme of research calls for greater acknowledgement of how multiple social particularities (for example, culture, gender, class, race, age) combine to intensify the gravity and consequences of the discrimination of BME groups (see Acker, 2006; Fearfull and Kamenou, 2006; Kamenou et al., 2013). The third theme recognizes the role of capital in explaining the disadvantaging of BME groups. Here, researchers explore the impact of human capital (see Rafferty, 2012) and social capital (see Mouw, 2006; Stewart and Garcia-Prieto, 2008) as explanatory variables to organizational outcomes. The latter stream of research is consistent with social discrimination wherein the treatment of individuals is based on their social group membership rather than their ability to perform a given role (see Scheepers et al., 2006). The conception of gatekeeping adopted in this article falls within this strand and it is useful to provide a brief overview of this literature.

Social discrimination theorists argue that understanding the social resources at the disposal of different groups provides a more appropriate lens for framing issues of discrimination and disadvantage (see Avery et al., 2008; Riordan et al., 2005). The underlying rationale is that individuals tend to form close association and network ties with others that are demographically similar to them; a phenomenon which researchers label ‘social homophily’ (see Mouw, 2006; Stewart and Garcia-Prieto, 2008), and which commonly results in the direct or indirect exclusion of demographically dissimilar work colleagues (see Avery et al., 2008).

Social discrimination theorists commonly illustrate the impact of demographic similarity by drawing on social identity theory to discuss the challenges that diverse groups (for example mixed ethnic groups) are likely to encounter in organizations (see Pelled et al., 1999). Interestingly, the desire to minimize such challenges may provide a potential ‘business case’ argument which may be linked to organizational level discrimination during recruitment and selection (see Brief et al., 2000; McGinnity and Lunn, 2011). Indeed, the conclusion of researchers is that BME groups are either denied the opportunity or are unable to obtain access to the social capital that is necessary for success in organizations (see Li, 2004). Studies also demonstrate the impact of taken-for-granted social ties and opportunities which are commonly unavailable to BME groups through discrimination and because they lack the critical mass in strategic positions to benefit from such ties (see James, 2000; Park and Westphal, 2013).

Although the absence of social capital can impact negatively on employees of all social categories, the impact is especially acute for BME groups whose visible dissimilarity generates an additional barrier. Indeed, while the surreptitious and clandestine nature of many forms of social capital (see Harris and Ogbonna, 2006) make it difficult to assess the full range of impacts, it is estimated that up to 50 percent of job vacancies in the USA are filled through social networks (see Mouw, 2003), suggesting that BME groups with fewer social networks are routinely disadvantaged. It is for this and other reasons that scholars call for additional research into the various forms and approaches that characterize this phenomenon (see Ogbonna and Harris, 2006; Park and Westphal, 2013; Riordan et al., 2005). This study contributes to understanding this aspect of discrimination by exploring the bases, motives and approaches of gatekeeping by shopfloor employees and by explicating the managerial and organizational context in which such behaviours occur.

Gatekeeping

Although the concept of gatekeeping has a distinguished history, it is surprising that it has not been more utilized in research and theorizing in work and organizational studies. Indeed, the notion of gatekeeping can be traced to one of the founding editions of Human Relations wherein Lewin (1947) adopted this construct to explain the role of different social actors in maintaining continuity and encouraging change. Lewin (1947: 145) defines gatekeepers as individuals or groups that are ‘in power for making decisions between in and out’. This power, however, need not be formally assigned, as Lewin’s (1947) illustration focuses on the role of a homemaker in influencing the diet of a family. In this sense, the gatekeeper is one who asserts direct or indirect influence and one who achieves this by acting as the intermediary between those that are ‘in’ and those that are ‘out’ (Barzilai-Nahon, 2008; Gray, 2002; Shoemaker et al., 2001).

Following Lewin’s (1947) work, a number of scholars from a diverse range of disciplines have adopted the construct of gatekeeping in explaining social and organizational processes. For example, Huczynski (1994) adopts the gatekeeping construct to describe the role of business academics in influencing the dissemination of knowledge and adoption of new business practices. Brekke et al. (2007) applies gatekeeping theory to discuss the role of general medical practitioners in restricting patient access to hospitals and specialist care. However, it is within the fields of journalism, communication and information science that gatekeeping is studied extensively in explaining the influencing behaviour of social actors (see Barzilai-Nahon, 2008; Shoemaker, 1991; Shoemaker et al., 2001).

Although the construct of gatekeeping offers a useful heuristic device for analysing work and employment issues (e.g. Gray, 2002), it remains relatively under-utilized. A few studies apply gatekeeping in addressing issues of discrimination and exclusion of disabled groups however. Specifically, a number of studies position human resource (HR) managers as employment gatekeepers through their role and influence in developing and managing organizational policies and practices, especially those that relate to recruitment, selection and promotion. The consensus of these studies is that HR managers play an active influencing (gatekeeping) role that impacts negatively on the employment of disabled staff. This includes the tendency of these managers to identify obstacles to employing disabled staff rather than opportunities (see Duff et al., 2007) and their unwillingness to persuade their organizations to make reasonable adjustments (Woodhams and Danieli, 2000). Other studies also reveal how managers externalize discrimination by shifting the blame for their discriminatory practices to external clients (see Hammond, 1997).

While the limited but insightful studies of gatekeeping demonstrate the usefulness of adopting this construct in organization, work and employment research, existing conceptions of gatekeeping commonly identify gatekeepers as those in positions of power such as newspaper editors (Barzilai-Nahon, 2008; Shoemaker et al., 2001) and HR managers (Duff et al., 2007; Woodhams and Danieli, 2000). The conception of gatekeeping in this article partly derives from Lewin’s (1947) original work in that gatekeepers are conceived as those whose influence capacity is not necessarily formally constituted but instead arises from their social position of being an intermediary. This conception is also consistent with social discrimination theories of demographic similarity in that it is not the individual ability or organizational position that defines their role as gatekeepers but rather gatekeepers’ access to social networks that arises from their demographic similarity (see Avery et al., 2008; Scheepers et al., 2006). Thus, we define shopfloor gatekeeping as influencing activities by shopfloor employees that intentionally or unintentionally exclude existing or potential co-workers from a particular area of work/work role or from potential or actual work-based benefit. Ethnic gatekeeping refers to instances where the primary focus of gatekeeping is race or other social characteristics that are linked to ethnicity. This definition shifts the focus of gatekeeping from isolated powerful individuals to mutually reinforcing influencing behaviours of individuals and groups that are based on the similarity and dissimilarity of salient demographic characteristics (in this case ethnicity and race). Moreover, this definition incorporates both intentional and unintentional gatekeeping. Thus, this article focuses on the impacts of the influencing behaviour of perpetrators of gatekeeping (whether or not such impacts were intended as discriminatory).

Research design and methods

Given the desire to explore and elucidate the key aspects of gatekeeping in relation to shopfloor workers and BME groups, an exploratory research design was considered most appropriate.

The authors are two researchers (one white and one BME) who have a long history of research collaboration. This pairing proved particularly useful as the study unfolded. Indeed, the initial phase of data gathering mirrored the findings of scholars on the impact of similarity and dissimilarity on salient demographic characteristics (e.g. Park and Westphal, 2013) in that there was an element of bias in the level of trust and openness displayed by the interviewees. Consequently, the researchers decided that the BME researcher would conduct the interviews with the BME participants while the white researcher conducted the interviews with the white participants. The white researcher had previously worked as a shopfloor worker in a similar organization and this researcher’s experience proved invaluable in helping the team to understand and navigate informal ‘shopfloor culture’ (including jargon, appearance and socializing) in a manner that engendered trust and cooperation.

A total of 48 in-depth, personal interviews were conducted with employees of six outlets of a large retail firm. The case organization granted research access to study the dynamics between different shopfloor groups on the proviso of anonymity but may be described as a very large general retail/supermarket firm. A typical branch employed 400 staff and had an annual turnover of around £60 million. Four of the branches were in areas where between 10 and 15 per cent of the population were from BME groups, while two branches were in areas where around six per cent of the population came from BME communities.

Although the focus of the study was on shopfloor workers, we interviewed four managers and seven supervisors (all of whom were white) to develop an understanding of the organizational context (strategies, policies, practices). White respondents constituted 78.4 per cent of the sample of shopfloor workers (29 of the 37) while the remaining respondents came from a variety of BME groups. Reflecting the gender division in the branches studied, most shopfloor employees were female (65.9 per cent) while eight of the 11 supervisors and managers were male. The average industry experience was just under 11 years while the mean tenure in the current firm was 7.3 years. The personal interviews typically lasted for 50 minutes. To promote the independence of the researchers and confidentiality of the study, interviews were conducted at a venue of the informant’s choice (a number of which included a park bench, a café, fast food outlets, a bar and the individual researcher’s car). All interviews were audio-recorded (with the permission of the informants) and transcribed.

Researching issues of discrimination and disadvantage is difficult because very few people view themselves as racist or are willing to talk openly about such behaviours. Thus, although some of the quotations presented in this article can be viewed as clear examples of prejudice, some are less clear-cut and reflect the researchers’ interpretation of the interview and contextual data generated. Such interpretations were helped by the notes maintained by the research team which detailed the background information including ‘body language during the interviews’ and discussions at the start and end of the formal interviews which were not recorded. Similarly, given the potentially sensitive nature of the interviews, the research team was careful to ensure that contentious individual interpretations (especially of actual events) were explored and alternative views/interpretations considered. As such, importance was placed on third party (preferably independent – that is, outside the immediate group) verification/contradiction (see Eisenhardt and Graebner, 2007).

Data collected were analysed both during and post data collection. Following the procedural recommendations of Strauss and Corbin (1998), three types of coding were utilized to analyse the data. First, ‘open coding’ was employed to uncover and identify the dimensions and properties of concepts in the data. Second, ‘axial coding’ was used to explore and to link the core categories together at the level of dimensions and properties. Third, ‘selective coding’ was adopted as a mechanism of both integrating and refining theory. Results were discussed with a sample of participants and factual inaccuracies were amended.

Findings

Our data analysis found widespread examples of gatekeeping in the research organization wherein shopfloor workers systematically excluded existing and potential employees who were demographically different. Over 70 per cent of the informants cited at least one example that fell into our definition of gatekeeping, denoting the widespread nature of the phenomenon.

Bases of similarity–difference and the motives for shopfloor employee gatekeeping

To guide the subsequent discussion, Figure 1 presents a framework depicting the bases, motives, approaches and tactics of discrimination via gatekeeping inclusion and exclusion by shopfloor employees.

OPEN IN VIEWER

Figure 1.

The bases, motives, approaches and tactics of discrimination via inclusion and exclusion.

The interview analysis consistently found that discrimination for varying motives and via different approaches and tactics (see later) was grounded in perceived demographic differences and similarities among shopfloor workers. While numerous bases were found, ethnicity and race were the predominant bases, although these were also intertwined with religion and, to a much lesser degree, gender and sexuality (the differing font sizes of the bases of similarity and difference in Figure 1 depicting frequency). Of the eight BME interviewees, six indicated that they had been victims of gatekeeping, with two Middle Eastern, two Pakistani, one Eastern European Romani and one black African employee falling into this category. Indeed, data analysis revealed that the most commonly targeted groups were Middle Eastern and Pakistani employees. Consequently, while acknowledging that various bases of discrimination were noted, the findings largely focus on issues of ethnicity and race as the most prevalent bases of perceived similarity and difference. Data analysis led to the emergence of two main motives for gatekeeping activities: blocking access and facilitating access. In both cases, the recipients of such actions were gauged according to their ‘difference’ from the gatekeeping perpetrator. Overall, gatekeepers believed that their actions were noble and helped to maintain group integrity. While a minority of white informants (nine) accepted that their actions disadvantaged others, the majority rejected any suggestion that their inclusive approaches were discriminatory.

The most common aim of influencing gatekeeping was to block access to ‘undesirable’ (that is, perceived to be ‘different’) employees or potential employees. As such, this motive was deliberate, considered and (largely) accepted as common practice among groups of employees. The blocking of access occurred for potential employees as well as existing workers. In the first instance, the blocking of potential ‘different’ employees from gaining employment within the firm was espoused to be significantly rarer and was often recognized as direct discrimination and therefore contrary to company regulations and legal requirements:

The company rules and indeed the law are very clear on such issues … We don’t discriminate against any type of employee. I know that every employer says that they are an equal opportunities employer but here it’s genuinely true. If you look at our employees at this store and indeed most of our stores, we employ a huge percentage of employees from ethnic minorities and a big proportion of our checkout staff are women. [HR Manager, white, female, aged 42, 17 years’ experience]

However, the interviews uncovered sporadic instances where potential employees who were perceived to be ‘different’ were excluded from positions under a thinly disguised, ‘politically acceptable’ veneer ranging from ‘poor social skills’ to ‘limited team-working skills’ to ‘incompatible experience’. For example:

The thing is, the night replen [replenishment shift] is a funny crowd. You’ve got one team of guys, all locals [all white, working class with a strong local accent] who have been here for years and work really well together. You put someone in there who disrupts all that and you’ve got a problem. It’s much easier to get someone who’ll fit right in straight away and just get on with it! We put a Romani fella on night shift last year – bloody disaster. Teamwork went out the door – had to move him on in the end. [Replenishment Supervisor, white, male, aged 58, 27 years’ experience]

While direct discrimination of blocking the employment of ‘different’ workers was largely clandestine and disguised, the blocking of access of ‘different’ employees within the firm was more overt and more common. The most common form of blocking access for co-workers centred on blocking access to particular teams or shifts. Many teams formed tight-knit work groups with strong social links and were deeply distrustful of employees that were either visually dissimilar or who were perceived to have dissimilar values. For example:

Everyone’s the same. Nobody likes working with people that aren’t the same as them. I’m not saying that they can’t work here! They can work for everyone and anyone they want! I just don’t want to work with people who aren’t speaking English! You never know what they’re saying about you! I bet you if you ask anybody here and they’ll want to work with the people they know! People like them. [Kiosk Assistant, white, male, aged 22, three years’ experience]

Such teams would jealously guard access to their groups, arguing that changes were disruptive to their work and ultimately ineffective for the firm. Particular shifts of work were also viewed by different workers as especially desirable, leading to blocking of access to the extent that, in some outlets, certain jobs and shifts of work were dominated by particular ethnic groups. For instance:

Different people are suited to different jobs. The management here are old hands at this game – they know we won’t let them down. We won’t let them get someone who’ll not fit in with our ways. We soon weed out the undesirables [nods head toward a group of Asian males outside window] – that lot never pull their weight. [Bakery Assistant, white, male, aged 45, eight years’ experience]

The second main aim of employee gatekeeping pivoted on facilitating access to potential employees viewed as ‘suitable’ (that is, similar to the group). A supervisor explained:

It happens quite frequently. One of the local boys will come along to me and say a friend, neighbour, cousin or someone that they know could do with a bit of work experience and would it be alright if they come in for a shift or two? I have to say, so far they’ve never let me down. You get an extra pair of hands and they can often turn out to be someone that fits right in! If I think about it, John came that way, so did James. As did Fred and George [all pseudonyms of four white employees out of an all-white team of 10]. [Warehouse Supervisor, white, male, aged 51, 22 years’ experience]

Such experiences were viewed very favourably by recruiting managers who often contacted informal team leaders for feedback or references. One manager’s comments provide what was tantamount to a ‘business case’ for promoting informal methods of recruitment even though this contradicted the formal company position on avoiding discrimination:

As with all retail, finding and retaining quality staff is our main concern. If we get someone who’s had a bit of experience here, that’s an advantage. An internal reference does count more than an external reference from someone you don’t know. [Non-Foods Manager, white, male, aged 28, seven years’ experience]

As such, while this may be viewed as merely the patronage of friends, family or acquaintances, the outcome of such patronage entailed the exclusion of other potential dissimilar employees, especially BME workers who lack the social network to secure such informal work experience placements.

Approaches and tactics of gatekeeping

While not all efforts at gatekeeping were always immediately successful, respondents argued that, over time, what this article describes as gatekeeping was effective, with estimates of success at both exclusion and inclusion averaging at around six out of seven times. Data analysis revealed three approaches through which shopfloor employees could exert a level of influence over the hiring, promotion and job allocation of other workers or potential employees. Two of the approaches focused on exclusion while the third focused on inclusion (see Figure 1). Each of the three approaches was accomplished and facilitated by a series of tactics that employees utilized.

The first excluding approach of shopfloor worker gatekeeping involved attempts to influence key decision-makers with regard to hiring, promotion or job allocations. This approach entailed subordinate shopfloor employees deliberately targeting superordinate decision-makers with the aim of influencing their decisions. As such, shopfloor employees had no direct authority to control decisions but employed surreptitious means to influence outcomes and thereby exerted some control over their working environment. Two equally common tactics were utilized by shopfloor workers: biased reporting and tacit agreements. The first tactic of decision-maker influence centred on the generation, fabrication and supply of deliberately misleading feedback regarding the performance, skills and abilities of co-workers perceived to be unwanted members of the team. This approach was used both to criticize demographically ‘different’ co-workers and also (less commonly) to endorse ‘similar’ co-workers with a view to improving their promotion or job allocation possibilities. For example:

Ali [a pseudonym] wasn’t necessarily a bad guy. It’s just there’s nothing we can talk to him about. We have absolutely nothing in common with him. It wasn’t anything personal. After the probationary week, we just said that he can’t talk to customers in the right way – blunt, rude, like. They [the management] always listen to us. He wasn’t really rude but he just didn’t fit in! [Pharmacy Assistant, white, female, aged 24, three years’ experience]

Variations of this tactic ranged from partial, misleading feedback to the simple telling of untruths. For instance:

Well, with some of the idiots they try foisting on us, are you surprised that we’ll do anything to keep ‘em out? [Whispering] There was this **** [Asian minority ethnic] guy. Urrgghh the smell was terrible! We couldn’t have that! We told them ‘he spent all his time skiving off, avoiding work’. Said he had no backbone! The [white] manager’s keen on backbone. [Customer Service Assistant, white, female, aged 22, two years’ experience]

Overall, the aim of such tactics was to present existing employees or potential employees as unsuitable for the required role or (arguably) more suited to other roles so as to maintain the preferred composition of the group.

The second tactic of decision-maker influence entailed the development and exploitation of tacit understandings between groups/teams of current employees and certain influential decision-makers. Particularly where the influential decision-makers had been promoted to their role internally, cohesive groups of similar employees felt able to draw upon and exploit past working or social links to develop an understanding of mutual benefit. For example:

John’s [a manager with responsibility for job allocations within a department] been here for years! He knows the score! We joined the same week and we still take our breaks at the same time, six years later! He and I understand each other about these things. He knows who we can work with and who’ll get in the way. He’s not going to upset the applecart by [whispering] putting one of ‘that lot’ [pointing chin towards a group of workers of Asian descent] in with us! [Night Replenishment Assistant, white, male, aged 43, six years’ experience]

Many of the managers interviewed claimed that bowing to such pressures merely reflected their appreciation of group dynamics and allowed them to find suitable, alternative jobs and roles for excluded workers. Interestingly, of the interviewees who could remember, 62.5 per cent of BME workers and 27.6 per cent of white employees were employed in roles which were less appealing than those for which they had originally applied, suggesting that BME groups may be more than twice as likely to be victims of gatekeeping.

The second approach of shopfloor worker gatekeeping focused on excluding newly hired, promoted or assigned ‘dissimilar’ employees from the team/group. Three tactics of exclusion emerged as most common. First were tactics of ‘hazing’, where existing workers played a variety of ‘practical jokes’, ‘tricks’ or ‘pranks’ designed to embarrass, humiliate or demean new employees. For example:

Oh the new ones always get a joke or two played on them! They tried to foist a young black lad on us last year! So we told him that if you wanted to smoke, you had to be a mile away from the store! He soon got the point! He is now working on the customer service desk, I think. [Bakery Assistant, white, female, aged 44, eight years’ experience]

Where new employees informally complained of such practices, co-workers would publically and vociferously deride and ridicule the ‘humourless’ employee while official complaints were met with social exclusion and almost inevitably a change in role for the complainant to ‘solve inter-personal differences’ (a term used euphemistically by a number of managers).

A second exclusion tactic employed by shopfloor workers involved concerted and orchestrated plans and actions calculated to make ‘different’ workers feel uncomfortable in both their role and socially. For example:

I did an induction week in the warehouse [shudders] – yerchh! I don’t think anybody talked to me for a week, let alone smiled. I asked to move after four days and was out of there by the weekend! I may be dumb but I ain’t stupid – I know when I’m not wanted! [Checkout Operator, BME, female, aged 24, two years’ experience]

Perpetrators claimed that such approaches were highly efficient and often rapidly led to targeted employees requesting different roles, assignments or in some cases leaving the company. Indeed, one apparent victim of such a tactic (now working elsewhere in the store) stated:

You know when you’re not wanted. The silences when you walk in, the looks, the jokes that ‘you wouldn’t get’. Being black doesn’t make you blind – just thick skinned! You ain’t gonna win the argument – you ain’t gonna change their minds – just best off away from that kinda c**p!’ [Household Team Assistant, BME, male, aged 34, four years’ experience]

The final tactic of exclusion via unfair work allocation was commonly presented as a longer-term approach and was the least common. This tactic involved teams/groups of similar workers allocating activities so that new ‘dissimilar’ employees were allocated roles or tasks that were difficult to complete within allocated times. One employee explained:

We work as a really good team here. If somebody doesn’t fit in with us, we just let them sink on their own. We’ll just get the new boy to take on something that they can’t do! And watch them fail! You do that night after night after night and it’ll soon get picked up! [Warehouse Assistant, white, male, aged 46, 14 years’ experience]

The aim of such tactics was, over time, for either the victim to feel inadequate and request a change of role/resign, or for the manager to recognize such ‘inadequacies’ and suggest a change in role.

The third approach focused on inclusion via promoting, advocating, supporting or endorsing new or existing employees typically based on demographic similarity. The patronage of new or existing employees was viewed by perpetrators as a constructive means of ensuring group/team stability, harmony and durability and also prolonging existing group efficiencies, store performance and ultimately organizational success. A shopfloor worker commented:

They always say at those management briefings, about how much the store depends on us [the shopfloor customer-contact staff]. Getting the right staff here is crucial. It’s not just a matter of how hard they work but is also whether they’ll stay here. If you get somebody who doesn’t fit, there’s no point, they will only leave after a few weeks! So, us finding them good staff is doing them a huge favour – they ought to pay us a bounty! [Internet Shopping Assistant, white, female, aged 39, 11 years’ experience]

In this regard, such patronage was viewed as not simply ethical but a helpful, constructive and pro-social activity. Suggestions that such processes might be viewed as discriminatory were vehemently and vociferously rejected by perpetrators. A typical response was:

Oh, that kind of political correctness is absolute c**p! It’s spouted by people who really don’t know what I’m talking about! It’s not about exclusion or being unfair, it’s about getting the best people! The managers here trust us to get the right people. [Phone Shop Assistant, white, male, aged 24, four years’ experience]

Three main tactics of patronage were observed: entry, new role and club membership patronage.

Entry patronage tactics involved existing workers actively seeking new demographically similar employees either within the outlet or outside of the firm. Potential sources of ‘non-different’ co-workers ranged in order of attractiveness from (the most common) family members, family friends, neighbours, co-participants of social activities to friends of friends. For example:

I’ve known Jack [a white locally born employee] for years. We used to play in the same rugby team when we were a lot younger … and slimmer! When he told me he was looking around, I got him straight in here. He’s a good lad! [Bakery Assistant, white, male, aged 47, 10 years’ experience]

New ‘non-different’ employees were also patronized by in-firm benefactors during probationary periods and early stages of employment or new role activities. For instance, one employee in the all-male, all-white warehouse department commented:

If they’re one of the boys, you don’t mind helping. Everyone gets confused over the first few days! So what we do is take them under our wing. Help them out. Show them around. You know? [Warehouse Assistant, white, male, aged 22, three years’ experience]

As such, patrons of new employees perceived to be ‘suitable’ were helped to achieve organizational expectations and guided through the informal ‘rules’ of team behaviour during work.

Finally, employees viewed as ‘similar’ were also patronized by automatic inclusion in (otherwise closed) group activities both during working hours and outside of work. Examples in work varied from invitation to join team-only break activities to sharing in work-based horseplay. For instance, one newly hired white worker observed:

I was at school with two of the crew [both locally born white employees]. They’ve been brilliant they have! Shown me the ropes. What to do, what not to do, who to watch out for, who’s a bit more of a lark, where we all go for a smoke, which table the crew sits at on break … all that stuff! [Checkout Operator, white, female, aged 31, under one year’s experience]

Outside of working hours ‘appropriate’ employees were ‘invited’ (expected) to participate in shared activities ranging from charity work to drinking sessions to sporting activities (the ability of new workers to participate in such activities being guaranteed via earlier assessments of ‘acceptability’). A long-serving group member of the night shift commented:

I knew he’d be fine! Fitted into the team straight off! He’s a damn good goalie too – plays for us now [us being the staff five-a-side football team]. [Night Replenishment Assistant, white, male, aged 43, five years’ experience]

Overall, this study found a range of motives, bases and approaches through which colleagues were included or excluded from potential or actual work-based benefits on the basis of salient demographic characteristics of ethnicity and race.

Conclusions and implications

The first contribution of this study is to the literature on discrimination of BME groups in organizations. Several studies suggest that the nature of workplace prejudice has changed and that overt discrimination has been replaced by covert, subtle forms of discrimination variously labelled as ‘everyday racism’ (Deitch et al., 2003; Essed, 1991), ‘subtle discrimination’ (Van Laer and Janssens, 2011) and even ‘modern racism’ (see McConahay, 1986), and which stems from the social construction of disadvantage through perceptions of similarity and difference. Indeed, scholars call for additional empirical insights into how these covert forms of discrimination impact on BME groups in the workplace (e.g. Cortina, 2008; Van Laer and Janssens, 2011), with the implication that a focus on uncovering covert discrimination is the appropriate way of advancing knowledge on BME disadvantage (e.g. Deitch et al., 2003). While concurring with the need for additional studies of covert discrimination, this should not imply a binary opposition (overt–covert discrimination, old-fashioned–modern racism) or a zero-sum research calculation wherein a focus on one is achieved at the expense of the other.

The adoption of gatekeeping theory provides a theoretical lens through which mechanisms and dynamics of various forms of discrimination can be illuminated. Further, the lens of gatekeeping theory supplies a novel perspective from which individuals’ positions within organizations can be incorporated into analyses of potential means and avenues of discrimination. Adopting this perspective leads to the finding of a common form of discrimination that is neither fully overt nor entirely covert, but that shares elements of both. Ethnic gatekeeping mirrors aspects of overt discrimination in that it can be a conscious act that is designed to privilege some people and disadvantage others on the basis of ethnicity and race. It also echoes elements of subtle discrimination in that it is frequently unintentional activation of behaviours that are subconsciously influenced by stereotyping and preconceived ideas of individuals and groups that are driven by demographic dissimilarity. Thus, this study suggests that understanding both covert and overt discrimination remains important in developing insights into the dynamics of discrimination that are not easily captured by a focus on one approach. The investigation of ethnic gatekeeping reveals an insidious form of discrimination which has been overlooked, and this study presents a framework which future researchers can build on to develop a more comprehensive understanding of this phenomenon.

Moreover, this study extends existing polarized conceptualizations wherein discrimination is either studied at individual/group level (e.g. Kamenou and Fearfull, 2006; Ogbonna and Harris, 2006) or at institutional/organizational level (see Kamenou et al., 2013; McGinnity and Lunn, 2011), with limited attention paid to the interactions and combined impacts across the levels. Through incorporating the managerial and organizational context, this study finds that what is seemingly discrimination at a single level of individuals and groups reveals more sinister managerial and organizational level discrimination. That is, not only did managers and supervisors abrogate their responsibility for ensuring fairness, they also tacitly approved and encouraged informal discriminatory practices (such as segregated work groups, word-of-mouth recruitment and informal placement provision) that helped them to maintain their control of costs and profit margins. Thus, although the findings of this study strongly suggest that shopfloor workers exert an influence on gatekeeping and discrimination, it is arguable that such discrimination is only possible in contexts where there are dysfunctional and discriminatory managerial and organizational processes. This suggests that to reveal additional insights into the interaction and combined impacts of discrimination, it is necessary to study discrimination simultaneously, not only involving the victims and perpetrators, but also incorporating processes at different levels.

Notwithstanding the position taken on whether organizational level discrimination facilitated individual and group level discrimination, a striking insight arises from the finding that shopfloor workers are at the frontline of gatekeeping in the organization studied. Indeed, not only does this study suggest that lower level employees may have a greater degree of influence over some aspects of their work than has been previously theorized, but this influence can also be deployed negatively in a manner that has been understudied so far. One way to interpret this finding is to focus on the implications for control and resistance in that shopfloor workers are active agents who seek to derive a level of influence over their work and work processes in the face of overly demanding and controlling managers and customers (see Lawrence and Robinson, 2007; Mulholland, 2004). This interpretation may, however, be misguided in that the ‘employee on employee’ discrimination moves this away from an ideological position that is based on hierarchical power to one that is centred on the domination of vulnerable minority groups in a manner that is morally undesirable and potentially illegal. This suggests that conceptions of the role of shopfloor workers should be broadened beyond the focus on resistance and opposition to management control to incorporate the range of influences they may have and to understand how some of their activities may be clandestine and sinister as well as morally undesirable.

The findings of our study confirm existing insights on the role of social capital and demographic similarity in understanding work-based discrimination (see Ibarra, 1995; Mouw, 2006; Stewart and Garcia-Prieto, 2008). Moreover, we extend these insights through uncovering a strong tendency of shopfloor workers not simply preferring to work with those who are demographically similar (social homophily), but, more intricately, proactively seeking to influence job allocation and to control the demographic dynamics of their immediate work teams. This leads to the conclusion that homophily and similarity-attraction are not (necessarily) benign social science constructs but are often driven by prejudices and bigotry with pernicious discriminatory consequences.

Linked to the above is the finding of a disproportional impact of gatekeeping across BME groups. Typically, Pakistani groups and new immigrants, particularly those from the Middle East and Eastern Europe, appear especially targeted by gatekeepers. In contrast, non-Arab black groups appear to be less targeted. This suggests that other factors may combine with demographic dissimilarity to trigger gatekeeping or to increase the intensity of such activities. Indeed, incidences of discrimination may be linked to ensuing public debates, with current discourses on immigration in the UK mass media being especially negative with particular hostility towards asylum seekers, Muslims and Eastern Europeans. In this regard, it is difficult to ascertain the impact of such general societal identification and hostility towards specific groups on both the behaviour of our research participants and their willingness to discuss these openly with the research team. This finding, however, suggests that understanding BME group discrimination may be tempered by the differential impacts on specific minority groups with narratives of ‘acceptable’ and ‘unacceptable’ BME groups constructed to suit prevailing socio-political discourses in the country concerned.