Beyond Policies and Social Washing: How Social Procurement Unfolds in Practice

Social procurement is increasingly used by organizations to create social value. An important feature of social procurement used to mitigate issues with social exclusion is employment requirements, which aim to create internships for unemployed marginalized people. However, little is known of their effects on people working at an operative level. Through 23 semi-structured interviews with practitioners in the Swedish construction and real estate sector, this paper adopts a practice lens to analyse the effects of employment requirements (ER). Findings show that practitioners must handle the tension between old and new practices, and strike a balance between fulfilling formal responsibilities and performing new practices on an ad hoc basis, and finding the time and resources to do so. Practitioners act as practice carriers for both traditional work tasks and new employment requirement practices, which can lead to role ambiguity. The paper provides novel details for how employment requirements unfold in practice. It also adds to practice theory by suggesting an important relational aspect between first-order, premeditated practices, and second-order, emergent practices, and how both types of practices are vital for working with employment requirements.


Introduction
When societies face challenges, such as mass migration, fiscal constraints, inequality gaps, and increasing poverty, private and public actors have tried to find new tools to help alleviate these issues. One such tool is for companies to use their purchasing power and procurement process to create social value. This is called social procurement and has, over the last decade, been increasingly used as a way to mitigate societal problems [1]. Social procurement encompasses a wide range of social criteria related to, for example, collective agreements and fair working conditions, health and safety, procuring from local, women's, or minority-owned enterprises, and employment creation for disadvantaged groups [2][3][4][5][6][7].
In Sweden, social procurement has mainly focused on formulating social criteria called employment requirements (ER). These aim to create job opportunities through internships for long-term unemployed, marginalized people, like immigrants, youths, or disabled people [8]. The focus on employment creation stems from issues with social exclusion, segregation, and unemployment [9]. Today many cities are segregated, and many of the buildings in these segregated neighbourhoods were built during the 1960s and 1970s, and are now run down and in need of refurbishment. The people who live in these neighbourhoods tend to be stigmatized in the labour market and often long-term unemployed [8][9][10][11]. In addition, in 2015, there was a large inflow of refugees who were subsequently unemployed, and often housed in the segregated neighbourhoods, which created an urgency for social procurement aimed at increasing employment and thereby decreasing social exclusion among these target groups [12,13]. This situation, also shared by countries other than Sweden, has led many organizations in the construction and real estate sector, such as housing companies, to implement requirements to hire unemployed people, sometimes even their own tenants, in refurbishment projects and the maintenance of buildings and green areas in the neighbourhood (cf. [8]). This also means that the construction and real estate sector is especially suited for employment-focused social procurement, due to the sector's close ties with social exclusion issues [1,14,15]. In Sweden, employment requirements have mostly been operationalized by creating internships. Therefore, this paper will henceforth refer to internships, and the individuals who get an internship through employment requirements will be referred to as employment requirement interns, or ER interns for short.
Despite employment requirements being seen as strategically important, both in industry and among policymakers, and that employment requirements present an opportunity for organizations to create social value in the form of employment for marginalized people, research is scarce in regard to how employment requirements actually unfold in practice [16][17][18][19]. Furthermore, although the empirical setting for this paper is the Swedish construction sector, employment requirements are novel and practices are still diffuse in many different sectors and geographical contexts, so the issue with diffuse practices and lack of knowledge is an international problem (cf. [1,5,7,16,18,20]). Therefore, to fill this empirical and theoretical knowledge gap, this paper aims to analyse the effects on operative level practices when employment requirements are implemented. The operative level in this paper refers to the daily work that individual and collective actors engage in when working with employment requirements in construction projects and building maintenance. By conducting a qualitative interview study, the findings highlight three specific areas related to the effects of employment requirements at an operative level: (1) for the daily project management practices and the practitioners, (2) for the internship, and (3) for the companies and projects, in the context of the Swedish construction and real estate sector.

Social Procurement
Social procurement has been used throughout the twentieth century to enact social policies, and has focused historically on issues like fair working hours and wages, the employment of disabled veterans in the UK, fair treatment of aboriginal populations in Canada, and fair treatment for African Americans in the United States [21]. Today, a new wave of social procurement initiatives, legislation, and policies are taking form, which aim to create employment opportunities for disadvantaged people. This includes the 2012 UK Social Value Act, which outlines how public contracts must acknowledge economic, environmental, and social well-being [22], and the EU directive (2014/24/EU) [23], which opens up the use of social procurement practices to a wider extent. In Australia and Canada, policies are mainly aimed at benefiting the indigenous populations [5,21].
These newer legislative acts and policies have spurred both public and private organizations to engage in social procurement [3,7,24,25], and in fact, social criteria relating to employment of vulnerable groups are today the second most used type of social criteria in public procurement in the construction sector [26]. In North Ireland and Scotland, new work roles which exclusively deal with social procurement are increasingly common, and in many ways, the work with social procurement has become business as usual [17,25]. This development, with the creation of new roles, is also seen in Sweden [19].
Previous studies have investigated common perceptions and experiences of social procurement in the construction sector. Many actors within the sector are in favour of social procurement, and studies have found that it is a useful tool to deepen the collaboration throughout the supply chain, to enable knowledge sharing, to fulfil client demands, to build competences, and to create employment [1,16,17,25]. Furthermore, social procurement is found to have benefits such as improved work environment, less employee turnover, lower absenteeism and improved productivity [27].
In Northern Ireland, Erridge [16] studied a project which used multiple contracts that embedded employment requirements. This study provides an important insight into how employment requirements can influence construction practitioners and projects, and it was found that few actors working in the project perceived that employment requirements increased their administrative workload. Although training was lacking for the newly employed, who mostly had no construction background, the jobs created in the project turned out to be sustainable over time, as 46 out of 51 people maintained their job after the project ended. Similar results were found in a large Swedish social housing refurbishment project where employment requirements were used [28]. There, 18 out of 50 people who were taken in on internships were given permanent employment after the internship ended, suggesting that jobs can also be sustainable over time in Sweden. However, to ensure sustainable positive outcomes, Erridge [16] highlights how commercial goals should not be overemphasized, as this can subsequently undermine the achievement of socioeconomic goals.
Another perspective on employment requirements is to see it as a service innovation and a way for organizations to provide new business opportunities [1,3,[16][17][18]. Kurdve and de Goey [29] conducted a case study of a project where unemployed people were given jobs to build standardized modular housing. This created more housing, and more importantly, it also created simple jobs in the construction sector for people who were lacking construction experience. The employment of marginalized people functioned as a service for the municipality, who is often the buyer of modular housing [28]. However, in contrast to these findings, Murphy and Eadie [25] found that contractors in the construction sector in North Ireland rather see social procurement as a contractual obligation driven by legislation than as a social innovation.
Previous research has found that, in the construction sector, there is a general lack of knowledge about social procurement [2,3,5,30,31], and the perceptions about the effects of social procurement vary. For example, some doubts persist about how social procurement might require more resources than traditional procurement [2,3,5,16,25,27,32], how it might displace "ordinary" workers [14,15,27], or that social value and employment requirements are difficult to evaluate (cf. [1,26,31,[33][34][35]). Other concerns relate to whether, despite its potential business opportunities, engaging in social procurement benefits brand goodwill or not [27], and whether collaboration around social procurement is difficult [32].
To address these issues and to ensure that actors in the construction and real estate sector are willing and able to implement and engage in social procurement, it is suggested that social procurement practices need to be tied to artefacts, resources and best practice examples [1]. These artefacts must also be complemented by clear arguments as to why these practices should be adopted [1]. Furthermore, Murphy and Eadie [25] suggest that, to ensure that the "right" social value is created, a more person-centric approach should be adopted when implementing social procurement, where practices should be individually tailored and bespoke for each intern or newly employed person and match their needs and skills.

A Practice Perspective on Social Procurement
Despite finding that social procurement has the potential to create social value, previous research has shown how social procurement can be difficult to work with in practice, due to many actors still being sceptical about it. Part of this attitude could be due to social procurement being underpinned by other institutional logics [36] than traditional construction procurement [37]. Firstly, rather than focusing on criteria that are easy to measure like price, social procurement instead embeds a social value logic rather than a market logic [37]. However, delivering social value typically lies outside the contractor's area of expertise, and therefore social procurement deviates from traditional work practices when mainly aiming to deliver social value [25]. Secondly, social criteria do not pertain directly to the object of procurement (i.e., the building), but instead put focus on other goals. Third, the construction sector is characterized by decentralized actors who retain their independence and individual decision-making, while simultaneously collaborating with other actors [38]. However, when employment requirements are implemented, clients are dictating what type of workers their contractors should hire, such as long-term unemployed people. This diminishes the autonomy of organizations that are used to collaborate, while maintaining a high degree of autonomy [37]. Thus, social procurement differs radically from traditional construction procurement, creating conflicting situations for those working with employment requirements.
In a study also looking at the introduction of a sustainability initiative in the construction sector, Hargreaves [39] used practice theory to study interventions to inspire pro-environmental behaviour change in a UK construction company. Here, it was found that it is important to study not only individual practices, but also their connections and conflicts with other practices. Building on this notion, in order to study the effects from employment requirements in practice, while simultaneously having the complex institutional context and the practical issues described in previous research in mind, this paper adopts the theoretical perspective of practice theory. A practice perspective is a theoretical construct that is particularly useful to study organizational phenomena that are complex, novel or dynamic, much like previous research has found social procurement to be. In addition, the practice perspective has a bottom-up approach and tries to understand practices and untangle the relationships between different practices and people [40][41][42]. This is in line with our aim to analyse the effects on an operative level when employment requirements are implemented.
Practices and routines are central to organizational life, development and outcome, and can be seen as the building blocks that make up the social reality of organizational life [40][41][42]. Practices are made up of multiple interdependent and interrelated activities [43]. Organizations can be better understood and studied through a practice perspective, through the examination of the ongoing, everyday life and experiences of people in organizations [40,42]. In order to understand grander organizational matters, an analysis of what people do in organizations is important [41]. Taking a practice approach to organizational research enables a closer description and understanding of what is involved in different phenomena-in our case, employment requirements-as they unfold and comes in closer to the "real" work that happens in organizations [42,44].
A practice perspective also emphasizes the relational aspects of practices and their performance, where practices are constituted socially, rather than individually, and where a shared understanding of practices constitutes how they are organized. Such relations are, however, not always equal, and practices can be arranged in bundles in different ways, which benefit some and not others. This means that power and politics are a large part of a practice perspective and that practices subsequently can be a source of conflict and are constantly in a state of tension caused by imbalances in power, resources, and interests [40][41][42]. Power and social relations thus strongly influence when and how practices are created, as well as maintained and can result in more or less room for different actors to manoeuvre changes [39]. Such features of a practice perspective may be especially pertinent when studying social procurement, which is a multiparty phenomenon with interorganizational collaborations and activities.
Practices can be spoken of as routines and seen as an interconnected assemblage of elements that together make up the practice, like bodily and mental activities, know-how, and emotional states [41,45]. A practice perspective enables an investigation of how and why practices are continually practised-unconsciously or consciously, how they may lead to institutionalization and norm-creation, and how they are changed [44,46]. This also means that, in practice theory, people are not seen as rational or as norm-following, but rather as practice carriers. Carrying practices means both creativity in terms of new inventions of practices and preservation in terms of iteration of old practices. This means that the experiences and identities of professionals influence how practices are reproduced and changed [39]. In other words, although people are constrained by the social system, they have the possibility to influence it through action [41].
Four interrelated concepts from a practice perspective shape the theoretical framework that informs the present study ( Figure 1). Firstly, from a bottom-up practice perspective, it is the everyday, lived experiences of individual actors that are of interest. Secondly, practices have relational aspects that impact their development and diffusion. Thirdly, tensions caused by imbalances in power, resources, and interests have an effect on vis-à-vis practices and how much room individual actors have to manoeuvre a change of practice in their everyday work. Lastly, individual actors are practice carriers, and the role they take impacts how they do this. These four concepts are used to guide the analysis of the empirical data, so as to be able to answer this question: What are the effects on an operative level when implementing employment requirements?
Sustainability 2020, 12, x FOR PEER REVIEW 5 of 19 analysis of the empirical data, so as to be able to answer this question: What are the effects on an operative level when implementing employment requirements? Figure 1. Conceptualization of the theoretical framework.

Methods
To study the effects from employment requirements on an operative level, a qualitative research approach was used, as it helps capture the actions, thoughts and beliefs of individual actors [47]. The research approach is empirically driven and abductive (cf. [48]), where we iteratively have moved between our empirical data and practice theory, in order to understand and conceptualize how employment requirements affect practices on an operative level. Following this approach, the design and focus of the study builds on previous findings from an explorative study, consisting of interviews with 23 people working strategically with employment requirements in the construction sector.
The study is based on three different cases where unemployed people were given internships through employment requirements (ER). The first case was a construction project ordered by a private housing company to build more apartment housing (AH). The second case was a construction project ordered by a municipality to build a pre-school (PS). Both of these construction projects were conducted by the same large Swedish contractor. The third case centred on a social procurement model developed by a corporate group of public housing companies (PHG), which demands that their subsidiaries take in unemployed people on internships to work with facilities maintenance, in other words, the public housing group posed employment requirements as an internal client.
We have chosen to refer to the interns working in the three cases as "ER interns". The reason for calling them "ER interns" and not just "interns" is that they are unlike "regular" interns, as they come from disadvantaged backgrounds and are stigmatized in the labour market. Those who are recent immigrants may have poor Swedish skills, may come from traumatic backgrounds like the war in Syria, or have undocumented and inconsistent schooling. Those who have disabilities may have physical or mental obstacles to overcome in the workplace. Therefore, ER interns often have special prerequisites that must be considered.
The interviewees chosen for the study were identified with the help of managers employed by the Swedish contractor in cases 1 and 2, and by the public housing group in case 3. All in all, 23 people working operatively with employment requirements were interviewed in a semi-structured manner [49]. Information about the interviewees included in the analysis of this paper are listed in Table 1 and have been given a personal anonymous code.

Methods
To study the effects from employment requirements on an operative level, a qualitative research approach was used, as it helps capture the actions, thoughts and beliefs of individual actors [47]. The research approach is empirically driven and abductive (cf. [48]), where we iteratively have moved between our empirical data and practice theory, in order to understand and conceptualize how employment requirements affect practices on an operative level. Following this approach, the design and focus of the study builds on previous findings from an explorative study, consisting of interviews with 23 people working strategically with employment requirements in the construction sector.
The study is based on three different cases where unemployed people were given internships through employment requirements (ER). The first case was a construction project ordered by a private housing company to build more apartment housing (AH). The second case was a construction project ordered by a municipality to build a pre-school (PS). Both of these construction projects were conducted by the same large Swedish contractor. The third case centred on a social procurement model developed by a corporate group of public housing companies (PHG), which demands that their subsidiaries take in unemployed people on internships to work with facilities maintenance, in other words, the public housing group posed employment requirements as an internal client.
We have chosen to refer to the interns working in the three cases as "ER interns". The reason for calling them "ER interns" and not just "interns" is that they are unlike "regular" interns, as they come from disadvantaged backgrounds and are stigmatized in the labour market. Those who are recent immigrants may have poor Swedish skills, may come from traumatic backgrounds like the war in Syria, or have undocumented and inconsistent schooling. Those who have disabilities may have physical or mental obstacles to overcome in the workplace. Therefore, ER interns often have special prerequisites that must be considered.
The interviewees chosen for the study were identified with the help of managers employed by the Swedish contractor in cases 1 and 2, and by the public housing group in case 3. All in all, 23 people working operatively with employment requirements were interviewed in a semi-structured manner [49]. Information about the interviewees included in the analysis of this paper are listed in Table 1 and have been given a personal anonymous code.
The interviewees from the AH and PS cases worked operatively, either with construction work on site or with project management from the construction client's side. Interviewees from the PHG case worked with maintenance of housing and green areas in the subsidiary housing companies in the public housing group. This means that all of the interviewees worked with implementing the employment requirements and/or with the ER interns on a daily basis, or worked as ER interns, and therefore, they had experienced effects from working with employment requirements.
The data were collected between December 2018 and May 2019. The interviews, which lasted for about an hour, focused on the interviewees' lived experience, positive or negative, of employment requirements, how employment requirements have influenced their daily work and role, and what changes of practices they had made to integrate the ER interns in the workplace. The interviews were transcribed verbatim, and to enable a systematic review, the data were coded in NVivo software. To identify common themes, the empirical data were first inductively and thematically coded [50]. The inductive coding was important considering how social procurement is underexamined both academically and empirically, and this enabled unexpected patterns to emerge [51]. After this, all items were recoded to refine the coding structure and to ensure that the codes reflected the empirical material as accurately as possible. A respondent validation session (see [52]) with individuals working in the PHG case was also conducted after approx. 2/3 of the interviews had been held, to ensure that the preliminary results and codes were valid.
After these two coding rounds, 11 categories of codes emerged (see Figure 2), from which three overarching themes were identified. These three themes related to effects (i) for the daily project management practices and the practitioners in the individual projects, (ii) for the ER internships, and (iii) for the companies and projects. We had a practice theory perspective in mind when we collected and coded the data, but it was not until after the two inductive coding rounds that we analysed the empirical data in an abductive manner (cf. [48]), using the conceptualization of the theoretical framework.

Effects for Construction Project Management Practice and Practitioners in Individual Projects
In their position as "receivers" of the ER interns and driven by personal beliefs, many of the

Effects for Construction Project Management Practice and Practitioners in Individual Projects
In their position as "receivers" of the ER interns and driven by personal beliefs, many of the interviewees explained that they felt pressure to provide the ER interns with meaningful work, an achievement that, according to them, needed the right conditions in terms of targets, supporting organization, and the opportunity to create social relations between interns and supervisors: Having targets [with employment requirements] is important, but other things are also important; however, the utmost goal is [to create] real jobs . . . You have to be able to set up the right conditions for things to work. It comes down to the people, the intern and the supervisor, but also the employer . . . It's about creating opportunities for relationships and situations where people can grow. (PHG1) Ensuring good conditions so the supervisors could support their ER interns was expressed as crucial to provide a 'high-quality internship' with fair working conditions. Issues that caused uncertainty and unwarranted stress on the interviewees were, for example, whether the ER interns received fair compensation for their work. In the pre-school project, this type of issue was repeatedly a concern for the supervisors and the project team. For example, they faced a situation where one ER intern went without pay for several weeks, and another only got paid approx. €3 per hour, as described by an interviewee: When I found out that the intern only got paid €3 per hour, I just said to [the employer] that either we hire him or we let him go, because I cannot ask someone to work for €3 per hour, that is below my dignity. I cannot ask him to work hard when he has that compensation, no way! So we hired him instead [of having him on an internship]. (PS4) The example above shows the importance of a perceived fairness in the job situation and how this also affects staff attitude to employment requirements. Nevertheless, with a short-term horizon and from a job-creation perspective, there were also voices among the interviewees that raised a need to accept this type of imperfect conditions, explained by one interviewee as follows: I think that for those who come here, they should be able to count on us and feel that when they've gone through with this [internship], they have a chance to get a job. That has to be the most important thing. (PHG2) Still, several interviewees struggled with doubts that their engagement might not actually have any long-term good effects for the ER interns. There was scant follow-up regarding what happened to ER interns after the internship ended, and how many of them actually got permanent employment. In a few cases, the interviewees knew only due to keeping personal contact with the ER interns after the internship, or accidentally by, for example, running into them outside of work: With some [ER] interns, I don't know what happened later. I think it's a shame that we don't get information on what happened with those who we've worked with for six months. But one lives here in this neighbourhood, so I see him sometimes. It's great when he tells me how things are going. When you work with someone three days a week, you talk about life, problems, and you get engaged in their lives, perhaps more than you should. (PHG5) The pressure to be a "good" supervisor made them personally engaged in the ER interns as people, and not only in their work. This was in contrast to the advice given from HR and employment officers, who promoted a strictly professional relationship with the ER interns, something that was found difficult to uphold in practice, since the ER interns asked for help with many private matters, for example, reading emails in Swedish, paying bills, writing CVs, and even helping them and their families find a new and better place to live: They come with their bills and ask for help how to pay them. We were told [at the supervisor course] not to do that, but it's difficult not to help when they don't understand how to do it. To help write CVs and fill in applications, which I had no idea how to do. But I just had to learn. . . . You're not supposed to do that, but it depends on the person, how much you engage. It becomes emotional. (PHG5) One of the supervisors has a young ER intern [now], and she helps him a lot, writing CVs, applying for jobs. Although she chooses to help, it takes her a lot of time; he needs so much support from her, in a way he needs a mentor. . . . But she feels a bit frustrated because she doesn't really have that time. (PHG6) This additional support requires much from the interviewees, especially those working as supervisors. They become involved in many activities they have not engaged in before and spend quite a lot of time trying to find appropriate tasks for the ER interns that match their prerequisites and interests: "I usually think about 'what is the most valuable thing for this particular intern, what will be most important for this individual person?'" (PHG1). One of the interviewees found out that one intern had previously worked as a painter in his home country, and therefore tried to find painting-related tasks, although some of them were not even needed, just so that the intern could do something he enjoyed.
However, despite the large amount of personal engagement needed, all the extra work and the creation of tasks suiting the interns' interests and previous experiences, the interviewees described how they felt that the work with the interns gave them a personal reward and a sense of contributing to an individual, as well as to wider society. As one interviewee (PHG2) said, "[The intern] told me that after he had got employment, he got his life back. I think that's big. It's very cool." Another interviewee expressed his satisfaction by saying: We don't live in a perfect world, but I think it's cool that [the employer] flexes their muscles and gives people internships and that they have the ambition to make these internships meaningful and lead to a permanent job. (PHG1) According to the interviewees, supervising ER interns opened up for meeting the people "behind the news reports", relating to the 2015 refugee crisis: "It's a big deal, it's rewarding to get a face-to-face perspective on events you have only seen on the TV news reports . . . to meet people who have been there" (PHG1). Thus, the stories of individual situations told by the interviewees demonstrate how employment requirements provided a space to meet people they would not normally meet. It caused them to reflect on differences and to care for people with another background. Being new in a cold country such as Sweden provided one such story: He [an intern originally from a country in Africa] had so many clothes on but was still cold... And it's not like he was saying 'I won't go out', because he does what he's supposed to do. The other day, it was really cold, and we were down by the harbour. I needed to change a bulb in a light post. It's kind of tricky, and it takes some time with the light fixtures, so I put the heat on and let him stay in the car. (PHG3)

Effects for the Internship
To be able to offer high-quality internships, the interviewees claimed that demands had to be made on the ER interns: It hasn't been easy; we've had interns we had to fire because it wasn't working out. We've had interns who stopped showing up to work, so we just had to terminate the internship and not waste any more time on them... We make our working place and resources available in order to help people. And if they don't want help, then I don't think it's our role to try and coax and nag them to come here. In those cases, we have simply ended [the internship]. (AH1) Furthermore, several of the ER interns were said to have become disillusioned with the idea of internships, as for many, the internship offered through employment requirements may just be one internship in a long string of internships, which have not yet led to a permanent job: "People go from internship to internship, but never land a permanent job" (PHG1). Both the ER interns and their supervisors described how some ER interns felt hopelessness and frustration with the system and had a feeling of never getting a permanent job. One interviewee explained that this frustration often led to repeated discussions: The interns are not always super motivated to learn Swedish or participate in the internship. They think things move too slowly since they just want a permanent job. They ask, 'Why should I be here [on an internship]? I just want to work'. That is the most common discussion I have with the interns, to try to convince them that they will get a proper job, but it will take some time. (PHG10) A few of the ER intern supervisors felt that some ER interns accepted the internship only to keep their welfare support, which they may lose if not partaking in various internships or other labour market initiatives. This reason for accepting an internship influenced the supervisors' approach towards these individuals, especially when compared to, for example, traditional apprentices: The [regular] interns I have had previously have done the internship as part of their education, so they have a much greater interest in the work and more prior knowledge, so that is a difference. The ER interns are not always so interested in facilities maintenance work. (PHG3) The interviewees, especially those working as supervisors of the interns, found the lack of motivation and interest in the tasks problematic, since it often caused discussions with the interns about why they should participate in an internship and required much effort to keep spirits high among them. At the same time, the supervisors felt that the effort had to be mutual, so the ER interns were expected to be as committed to their work as other employees: As a supervisor, I have some level of responsibility, but that is, of course, shared with the intern. You have a shared responsibility that the [internship] is a meaningful time because you don't get rich coming here. Instead, you hopefully gain experience and know more things when you leave. So that is a responsibility. Part of it is giving them work experience, but a large part of it is also to teach the language, and that is usually far outside my work description. . . . [However] I try to provide opportunities for those who are ready to take them, to practice their abilities to hold a conversation in Swedish. That is a strength with this internship. (PHG1) Below is an example given by an interviewee on how to learn Swedish by seizing all work as a learning opportunity, even when the ER intern, in this case attending meetings in a language foreign to them, cannot contribute much. The quote also illustrates the importance of making work meaningful by looking at it from an alternative viewpoint and with a broader perspective: Instead of thinking that this is a meeting where the intern cannot understand all the conversations and saying that they should rake leaves instead, I give them a notepad and tell them to jot down 20 Swedish words to learn. So it's about finding a meaningful perspective for the intern in various situations . . . Instead of zoning out, don't mind that and grab some words off the PowerPoint! (PHG1) In addition, the ER interns emphasized the problems caused by a language barrier. One interviewed ER intern (PS5), who had newly immigrated to Sweden, explained that he did not think he could ever get a (permanent) job because of his poor Swedish: "I can work well, but I cannot speak very much. The language is a big problem for me, so [my supervisor] has had to help me a lot".
An issue that appeared in cases where the internships were going well and where the ER intern had been successfully integrated into the organizations was that this occasionally led to an overexposure of the ER interns by, for example, using them for marketing purposes: When we take someone in, I think they are just like anybody else. I can notice a tendency that some want to raise this all the time, and I don't like that. It bothers me because they are people, and I have taken them in because of who they are, but there are many who want to market [employment requirements], and that doesn't feel right to me. (PS4) This means that some well-performing ER interns may be used as speakers at industry seminars, be featured in marketing materials, be posted on the company website, be interviewed in magazines or radio shows, etc. By doing so, they are labelled as something other than ordinary employees, which counteracts the intentions expressed by the supervisors of taking them in and 'treating them as any other employee'.

Effects for the Project and Organization
Many tasks of the supervisors, and by extension, the ER interns, include a fair amount of communication with sub-contractors, clients and tenants: "It's a lot of language in the role of working with facilities maintenance. It's about communication, both with tenants and contractors . . . many face-to-face meetings" (PHG1). Therefore, language issues were considered to be a major difficulty and barrier for employment requirements to be fully implemented: It's been more demanding than I thought it would be. The most difficult thing with the interns [refugees] is the language, to make yourself understood, because they need to understand me, and I need to understand them. That's the difficult part. (PHG5) Additionally, the ER interns often did not fully understand Swedish work culture, which led to unnecessary misunderstandings and some frustration within the team: "The interns did not know our social codes or how we act within the Swedish work culture and in our workgroups" (PHG9). Language and cultural barriers did not only hinder the socialization of ER interns into the workgroup, but also made supervision difficult. One of the major concerns was safety and a fear of accidents, since handling heavy equipment in both construction and facilities maintenance needs clear instructions. One interviewee explained: Safety is very, very important. And that includes everything from how you lift things to how you handle equipment. For example, a handheld grass mower with rotor blades: to try to explain to someone who doesn't know that many Swedish words that you can absolutely never ever put your fingers under the rotor blades. Things like that are very important. (PHG3) Another obstacle to broader implementation of employment requirements was the type of the projects themselves, where not all projects are suitable for employment requirements. An interviewee (PS2) raised issues regarding the size of the project, the nature of the work, and a lack of suitable candidates: In a large-scale project, they can offer much more diverse tasks, so there I can imagine that you can employ people without a background in construction. . . . We explained to the municipality [the client] that we cannot take just anyone. If they are supposed to be a carpenter apprentice, they must know some basics, to use the tools. So we can't just take in a layman carpenter . . . In this contract we formulated that we would take in ten interns. But after a while, we realized that we will never reach ten interns, so the original idea wasn't well-thought through. (PS2) The size and scope of the contract thus have an impact on the employment requirements, and vice versa, but there may well be a general shortage of suitable people to take in on an internship. Nevertheless, although there are many practical barriers related to employment requirements, as illustrated by the quote above, the interviewees emphasized that the ER interns are ordinary employees, and therefore are expected to do a job just like any other employee, as described by one interviewee (AH1): "I have chosen all of them because I think they add value to our group, not because of where they come from". Thus, they were expected to do real tasks on real terms: There are no simple jobs. Some think [the interns] should only pick up trash. But they do the same job we do. . . . They shouldn't only do the boring tasks. . . . They must feel like they're here on the same terms as we are, because I wouldn't want to go to Iraq and only pick up trash. They need to be involved and be able to see that they can advance [in their career]. The more you learn, the more you can climb the ladder. . . . They should have all the possibilities. (PHG2) Besides an ambition to perform work on conditions similar to any other employee, the ER interns and employment requirements were perceived to also create added value for the work team. According to the interviewees, when jointly engaging an ER intern at work, this shared responsibility tied the team closer together. Taking an ER intern aboard and socializing the person into the team requires joint efforts and an adjustment of work practices. If the team succeeds in doing this and puts up with the extra efforts needed, it was seen as confirmation that the team is functioning well: There has to be an interest from everybody to engage, and here everybody did get very engaged. . . . The team felt like it has been great fun . . . and of course that creates team spirit. And everybody was adamant that [the intern] would do well. So, in such a situation, it brings the team closer together... We support each other. (PS2) For the supervisors, added value was also found on a more personal level: "I think [working with the interns] gives me some sort of added value in my own employment, that I work for an employer who is a genuine builder of society" (PHG1). Another interviewee said: I feel all the time that I am happy to be able to help, to help a person who hopefully shall live and feel good here [in Sweden], to have a good life that works and that everybody benefits from. If people around us are feeling good, then we all feel good . . . It feels good to contribute in that way. (PHG3) Many interviewees, especially those working for public organizations, expressed that working with employment requirements and taking in interns should be a natural element in the work of their organizations: We have to give them a chance, absolutely. It has to be terrible not to have anything to do [when being unemployed]. It becomes a vicious circle where they don't get anywhere. It must lead to such a terrible frustration. So I think [employment requirements] are really important, it's our responsibility now. (PHG2) However, in spite of a general favourable attitude towards employment requirements and their effects on the supervisors, the team, the organizations, and society, many of the interviewees also stressed that it is not a be-all and end-all solution: I think it's great that we're doing this, we give these people a chance. But we have to ensure that we get results in the end. We can't succeed with everybody, but we should have the goal that everybody gets employment. (PHG2)

Discussion
The aim of this paper is to analyse effects at an operative level when employment requirements are implemented. Applying a practice lens, the analysis will now be discussed in terms of the theoretical framework focusing on (1) everyday lived experiences of individual actors, (2) relational aspects of practices, (3) tension caused by imbalances in power, resources, and interests, and (4) an individual actor's role as a practice carrier. The section ends with a discussion of first and second order practices in environments characterized by competing institutional logics.

Everyday, Lived Experiences of Individual Actors
The interviews in this study have focused on what individual practitioners do in their workday in relation to employment requirements and the ER interns. What became clear was that individual actors had to create many new types of practices as a result of employment requirements. The interviewees expressed a frustration that they did not have the time to carve out a space for dealing with the internships in the way they would like. Follow-up was scant, compensation bureaucracy complex, projects sometimes lacked necessary scope and scale for internships, the ER interns often lacked language skills and previous experience, and the employment requirements were imposed on them in a top-down manner and landed in the laps of individuals at the operative level without fully formalized support and necessary resources. This would suggest that established practices have not yet changed enough to fully accommodate new practices related to employment requirements (cf. [44,46]).
One thing that could help to firmly establish practices related to employment requirements is to make them more routinized and standardized. Creating routines could be a first step to more norm-creation when it comes to social procurement practice [44,46]. Firstly, routines relating to the administration of how to handle, for example, compensation for ER interns could be improved, starting with increasing the knowledge of different compensation schemes. This corroborates previous research on social procurement, which has found that there is a general lack of knowledge about how to practically implement and work with social procurement [1,3,5].
Secondly, routines for following-up individual internships would help resolve general uncertainties regarding the results from social procurement, which has also been highlighted by previous research as necessary for social procurement to be widely accepted (cf. [26,31,34,35]), and thus legitimize employment requirements in practice. Implementing follow-up routines would benefit supervisors of ER interns, as getting feedback on what happened to them after the internship could help assuring them that their engagement in the ER interns has had long-term effects. Perhaps Erridge's [16] finding that many people who get employed through social procurement actually maintain their employment after their internships end can indicate that ER interns in Sweden can also have the same outcome.
Despite the lack of standardized and routinized practices and that working with ER interns can lead to increased stress for the supervisors, who must engage in extracurricular tasks outside their normal work responsibilities, it is clear from the interviews that value is created for ER interns, for individual supervisors, for work teams, and for projects as a whole. Thus, social procurement has the potential to serve as a value-adding service in the construction and real estate sector in many ways (cf. [29]). This value creation, despite not being institutionalized [44,46], can be enabled through the relational and emotional aspects of working with employment requirements and ER interns.

Relational Aspects of Practices
The findings show how practitioners are strongly driven by a wish to do good: do good by the internships, good by their organizations, and good by themselves. This "doing good" is reflected in their personal expectations to provide meaningful internships and fair compensation, by tailoring work tasks to fit the ER interns' skills and interests, by personal engagement, and by doing extracurricular work tasks regarding the ER interns' private lives. This is a good illustration of the relational and emotional aspects of practices, how engagement and caring for the ER interns can lead to the creation of new practices not previously practised in their everyday work life [41,45].
Practices related to employment requirements can also be seen as relational in terms of supervisors' engagement becoming self-supporting and having a beneficial effect also for the companies. It is difficult to draw any firm conclusions based on three cases, but this study indicates that when a work team is well-functioning, team members are open-minded and everyone in the work team is equally engaged in the ER intern, it does seem that the internship goes better. Reciprocally, this may have a positive effect on the perceived quality of the ordinary workers' employment. The drive for doing good adds an extra level to the supervisors' working lives, in the sense that they become proud of their employer, satisfied with their work role, and part of developing a better functioning work group. This adds to the findings of Eadie and Rafferty [27] that social procurement can lead to an improved working environment and potentially increased productivity as well.
Previous research on practice theory has claimed that practices are socially constituted in relation to other people, rather than individually constituted, and thereby adding a relational aspect to practices [40][41][42]. In this study, the findings suggest that implementing employment requirements can impact favourably the relationship employees have with their employer and each other, for example, by creating a better team spirit. These relational aspects of employment requirement practices thus seem to have led to unexpected good outcomes in other respects besides employment for marginalized groups.
The importance of social relations is thus clear in the case of supervisors, the work teams in the projects, and the ER interns. When a work group is fully engaged and the ER intern receives support from more people than just the official supervisor, the interns are (1) socialized into the project more fully, (2) the possibility of learning more skills increases and (3) both practical skills and language skills are improved. These developments should lead to a better chance of finding permanent employment after the internships' term expires, either in the same organization or elsewhere (cf. [16]). Therefore, there seems to be positive reinforcement, a cumulative effect and an added value to individual workers, to work teams, and the ER interns. In such an environment that becomes self-reinforcing, actors likely have more space, resources and power to manage conflicting practices and to take the initiative to create new sustainable practices. In other words, they have been able to influence their working environment and have managed to strike a balance between new and old practices, making them work more in harmony [41].

Tensions Caused by Imbalances in Power, Resources, and Interests
Social relations between supervisors and interns also come at a cost, in terms of widening the supervisory role and responsibilities. The findings show that supervisors engage in extracurricular tasks like helping the ER interns read private emails, make phone calls on their behalf, write CVs, find new living arrangements, etc. These extracurricular tasks would suggest an increase in administrative burden and a need for more resources, especially in terms of time. This points to an imbalance between two competing practices, where the implementation of social procurement might lead to other established practices having to be cut down due to limited resources (cf. [41]). This finding, that working with employment requirements can require more resources and lead to increased administrative burdens, contradicts Erridge's [16] findings that social procurement does not lead to more administrative duties. Nonetheless, increasing resources are likely to lead to increased costs, something which previous research has found is a concern for the sector [1][2][3]5,16,27]. Therefore, what is best for individual ER interns is not necessarily what is best for construction and real estate organizations' bottom lines.
Another imbalance relating to ER interns concerns expectations regarding their work. The lack of construction experience among the ER interns is a problem at an operational level and for the supervisors, as the ER interns' inexperience leads to accommodations having to be made in daily work practices. For example, as the ER interns have language barriers and often no background in construction or facilities maintenance, some tasks they are expected to do become difficult (e.g., communicating with tenants), and some tasks even become dangerous (e.g., operating heavy equipment). That ER interns are (at least initially) expected to perform tasks they are ill-equipped to do creates a discrepancy between expectations and reality. This discovery also mirrors many of Erridge's [16] findings regarding a lack of training for ER interns.
Despite some ER interns being ill-equipped to perform certain tasks, the ER interns are said to be treated like the ordinary employees and perform the same work tasks as their colleagues and supervisors, as "there are no simple jobs". This means that there is a contradiction in the way the ER interns are viewed, where adjustment in daily work practices must be made (which is not easy to do), while at the same time the ER interns and their work are not to be acknowledged as any different. As such, it seems that the actors working with social procurement have not been able to fully influence old practices and make space for new practices, and a tension remains between what ER interns are expected to do and what they are able to do [39,41].
The question then becomes if this tension hinders the ER interns in their quest to find permanent employment and to learn Swedish. If the ER interns are not given enough support, they can miss important learning opportunities, because they have too much of a hill to climb. However, if the ER interns receive too much special treatment, they may feel cosseted and become incapacitated and less independent. This finding is in line with Murphy and Eadie's [25] conclusion that practices must be made bespoke in order to achieve maximum social value. However, how to actually create and establish these bespoke practices and how to achieve a balance between support and self-sufficiency for the ER interns is unclear. What is clear is that developing bespoke practices is time-consuming and adds more pressure on supervisors who express how they feel stressed to complete their non-intern-related work tasks in less time. As such, bespoke practices may increase social value for ER interns, but decrease value as well as increase stress for intern supervisors. Moving forward, organizations in the construction and real estate sector who want to engage in social procurement will have to balance different institutional logics (cf. [40][41][42]). One way for individual actors to handle this balancing act could be to change their ideas of what their role and identity should be in relation to employment requirements and their daily work.

Individual Actors' Role as Practice Carriers
As was previously mentioned, the scope of supervisors' responsibilities unexpectedly changed when the supervisors started working with ER interns. Besides their normal construction work and facilities maintenance tasks, they also became involved in helping the ER interns with private matters. Supervisors often became a very important 'Swedish friend' for (newly immigrated) ER interns. Taking this into consideration, formal work tasks, and resources to perform those tasks, need to be widened, so that supervisors have the mandate to also include work and responsibilities in their role which differ from traditional construction or facilities maintenance work. This is already done in an ad hoc manner by the supervisors, who act as practice carriers when taking initiatives to create a space for the establishment of these new extracurricular work practices [41].
Many of the extra-curricular tasks undertaken by the supervisors have traditionally been performed by social workers or similar, which suggests a hybridization of their role [53]. Thus, the role of supervisors and practitioners in construction and real estate companies change when using employment requirements. When supervisors engage in social-worker-like tasks, while at the same time being expected to perform their usual work tasks, they will need to prioritize which role they enact and navigate between different practices.
This means that carrying practices related to construction or facilities maintenance work, while at the same time carrying practices related to employment requirements, could create identity ambiguity and uncertainty about the scope of their responsibilities. A successful merger of the two identities may mitigate this ambiguity and facilitate creating sustainable practices (cf. [39]), while failing to do so may lead to the fragmentation of employment requirement-related tasks into daily work practices (cf. [53]). A hybridization of the role might mean that supervisors can carve out more space for their tasks than they previously have been able to [39]. This might enable the creation of new bundles of practices (cf. [40][41][42]), combining both traditional construction work and facilities maintenance tasks with more social worker-esque tasks. Such creative work (cf. [39]) could ultimately lead to the establishment and institutionalization of a new employment requirement practice (cf. [1]).

First and Second-Order Practices
Looking at the discussion thus far, it is clear that many resources and new practices are created when employment requirements are implemented. Some of these practices are more or less expected and were intentionally designed when deciding to implement employment requirements. Other practices were unexpected and created on an ad hoc basis. Based on the findings presented in this paper, both types of practices are important and serve different purposes (cf. [44,46]).
Some practices are expected and necessary when implementing employment requirements, such as recruiting the ER interns, assigning supervisors, introducing the ER interns to their new workplace and work tasks, setting the ER interns to work, and monitoring their progress. We call these first-order practices, as these are fundamental when implementing employment requirements in the first place. However, these first-order practices are not enough to make employment requirements work in practice. They must be complemented with unexpected and unplanned extracurricular practices (cf. [44,46]), such as giving the ER interns impromptu Swedish work culture and language training ("grab some words from the PowerPoint"), helping them read emails and pay bills, ensuring they get fair compensation, finding new living arrangements for them, and giving them appropriate work tasks that they can perform in line with their previous experience, skills, and interests to keep them motivated. We call these practices second-order practices, and these second-order practices can be said to be supportive of the first-order, fundamental practices, and something which individual actors can manoeuvre (cf. [39]) into their workdays to make it all come together.
The second-order practices are not the main foci of the implementation of employment requirements and were unforeseen by the interviewees. However, they are nonetheless vital to making the first-order practices work at all and were created on an ad hoc basis so that the internships would not fail. In other words, just conducting practices to implement employment requirements is not enough to make them work; therefore, the second-order practices are vital. This notion of firstand second-order practices adds a valuable insight into how to make social procurement become a sustainable practice, as well as adding to the idea of the relational aspects of practices [40][41][42] where the organic emergent nature of practices becomes visible.

Conclusions
The findings in this paper provide novel, detailed insight into the effects from social procurement and employment requirements at an operative level, for actors working in the construction and real estate sector. Employment requirements entail new demands on the practitioners as receivers of ER interns, which in turn calls for a personal engagement with the ER interns and their private lives. The practitioners' drive for "doing good" by the internship, good by their organization, and good by themselves leads to self-reinforcing effects like them becoming proud of their employer and satisfied at work. However, they often lack the time and resources to handle ER internships in a way that they would like. ER interns have to deal with demands from their supervisors as to how they should engage in their internship, while at the same time, for various reasons, they often feel demotivated to fulfil what is required of their internships. In the daily operative work of the project or maintenance duties, the findings point to several obstacles to effective use of employment requirements, such as the language barrier, safety issues and non-alignment with how work is structured. However, despite the fact that employment requirements are difficult to implement and work with in practice, the ER interns add value to the individual construction project and the organizations, for example, in terms of improved work satisfaction and team spirit among organizational members.
For research, this paper firstly provides a bottom-up and micro-level perspective on practices and the daily working life of people on an operative level, which to date, has been lacking in studies of social procurement. Secondly, by having a practice lens, the tension between new and old practices that individual actors must handle is illustrated. The paper shows how individual actors must strike a balance between fulfilling their formal responsibilities and performing new practices on an ad hoc basis, to ensure that their daily life with the ER interns work. This navigation between practices is not easy, due to an imbalance of resources. Acting as practice carriers for both traditional work tasks and new employment requirement practices can lead to an ambiguity of what the scope of responsibilities and roles is. However, through a hybridization of the roles of individual practitioners, this ambiguity may be mitigated, and sustainable employment requirement practices can be established. Lastly, the paper suggests an important relation between what we call first-order, premeditated practices, and second-order, supportive and emergent practices, and how both types of practices are vital for making employment requirements work.
For practitioners who work with social procurement, this paper emphasizes the importance of widening the official responsibilities of supervisors of ER interns, as well as balancing bespoke, person-centric practices to individual ER interns and their individual abilities, with standardized and routinized practices. This could include issues regarding compensation and follow-up routines. Making such changes could enable a more effective use of employment requirements, and a positive cumulative effect for ER interns, their supervisors, and organizations.
Future research could investigate how expectations and plans for employment requirements differ between parent organizations and clients, in relation to what actually happens in practice in the projects, and how collaboration between projects, parent organizations and clients are organized. Moreover, the interviewees talked much about creating meaningful internships for the ER interns, and many ER interns are perceived as disillusioned by their previous internship experiences that never led to a permanent job. Future studies can therefore build on previous research on meaningful work, to examine what that implies for employment requirements, how it is achieved, and how the sense of meaningless work can be diminished.