“SwThises FJiorsutr nOanle o tfo S Poicciko luopg yt,h 4e8 P (h2o),n 2e0”2: F2o, r3m3s5 –of3 5R2ec ruitment for Low-Skilled Jobs 335 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs Eva Nadai* and Robin Hübscher* Abstract: Based on a qualitative study on the employability of workers without vocational qualification in five different industries, this article examines hiring practices in the labour market segment of low-skilled jobs from a sociology of conventions perspective. In the ab- sence of educational signals, employers use personal networks and trial workdays to reduce the uncertainty regarding the quality of job applicants. However, professional and technical intermediaries become more important, thus leading to the formalization of recruitment channels and valorisation. Keywords: Recruitment, low-skilled jobs, labour market intermediaries, convention theory, valorization «Der Erste, der das Telefon abnimmt»: Rekrutierungsformen für geringqualifizierte Jobs Zusammenfassung: Basierend auf einer qualitativen Studie zur Beschäftigungsfähigkeit von Ausbildungslosen analysiert der Beitrag Rekrutierungsformen für geringqualifizierte Jobs aus einer konventionentheoretischen Perspektive. Angesichts fehlender Qualifikationen nutzen Arbeitgeber persönliche Netzwerke und Probearbeitstage zur Reduktion der Ungewissheit über die Qualität von Stellensuchenden. Professionelle und technische Intermediäre gewinnen aber an Bedeutung, was zu einer Formalisierung von Rekrutierung und Bewertung führt. Schlüsselwörter: Rekrutierung, unqualifizierte Jobs, Intermediäre, Soziologie der Konventi- onen, Bewerten « Le premier qui décroche le téléphone » : Les formes de recrutement pour les emplois non qualifiés Résumé : Sur la base d’une étude qualitative sur l’employabilité des travailleur·euse·s non qualifié·e·s, cet article examine les pratiques de recrutement dans le segment des emplois non qualifiés dans une perspective théorique conventionnaliste. En l’absence de qualifications formelles, les employeur·euse·s utilisent les réseaux personnels et les périodes d’essai pour réduire l’incertitude concernant la qualité des candidat·e·s. Toutefois, les intermédiaires professionnels et techniques gagnent en importance, ce qui conduit à la formalisation du recrutement et de la valorisation. Mots-clés : Recrutement, emplois non qualifiés, intermédiaires, sociologie des conventions, valorisation * University of Applied Sciences and Arts Northwestern Switzerland, School of Social Work, CH-4600 Olten, eva.nadai@fhnw.ch, robin.huebscher@fhnw.ch DOI 10.2478/sjs-2022-0017 SJS 48 (2), 2022, 335–352 © 2022. This work is licensed under the Creative Commons Attribution-NonCommercial- NoDerivatives 4.0 License. (CC BY-NC-ND 4.0) 336 Eva Nadai and Robin Hübscher 1 Introduction Over the past decades, there has been a sharp decline in low-skilled jobs globally (Eurofound 2015; OECD 2017). Moreover, the remaining low-end jobs are being upskilled (Abel et al. 2014), whereas the demand for routine skills and physical abilities typical for such jobs is decreasing (OECD 2019). Hence, low-educated workers are experiencing increasing difficulties in labour market access. Seen from the demand side of firms, these developments create an excess supply of low-skilled workers for the fewer jobs that do not require formal qualifications (Can and Shel- don 2017). Thus, when employers need to fill a vacancy in this segment, they are confronted with a mass of job applicants lacking a crucial distinguishing feature, namely educational credentials, from which they have to pick the right person for the job at hand. The nondescript profile of low-skilled workers increases the fundamental un- certainty inherent in any recruitment. Hiring implies uncertainty insofar as “the only true measure of a worker’s effectiveness is job performance” (Doeringer and Piore 1971, cited in Larquier 2016, 13), which is unknown prior to actual employment. Employers therefore infer the productive capacities of a candidate from signals like education and from ascriptive indicators like gender, ethnicity or age (Spence 1973). Educational credentials are a crucial signal, both in a positive and a negative way: they facilitate selection by conveying standardised information on the skills and capacities of a candidate. The other way around, the missing occupational qualifica- tion constitutes a negative signal regarding productivity (Gesthuizen et al. 2011). Yet, since it is the defining characteristic of low-skilled jobs that they only require a short training on the job (Abel et al. 2014, 15), formal qualifications cannot be the decisive hiring factor in this labour market segment. Employers thus have to find other means of obtaining reliable information on the qualities of job candidates. Research on employer strategies to handle the uncertainties of recruiting and selecting workers emphasises the importance of informal channels and practices, especially social networks (Fernandez et al. 2000; Marsden and Gorman 2001; Bills et al. 2017). In the case of low-skilled workers, the use of referrals through personal contacts bypasses the problem of the missing standardised information on skills (Waldinger and Lichter 2003; Hieming et  al. 2005; Hassler et  al. 2019). Other common practices in the low-skilled segment are walk-in applications (Rieucau 2015) and trial days prior to hiring (Hieming et al. 2005; Hassler et al. 2019). The preference for informal recruiting practices for low-skilled jobs does not exclude the use of formal methods since firms usually combine several channels and meth- ods simultaneously or sequentially (Marchal and Rieucau 2010; Marchal 2015). Channels reaching beyond personal contacts include professional intermediaries like private temporary work agencies (TWAs) or public employment services and technical means like job advertisements, online job boards or work-on-demand apps. SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 337 In their overview on the role of employers in hiring, Bills et al. (2017, 303) call for contextual analyses of recruitment and argue that studying agency in hiring is key to understanding the “microfoundations of labor market processes” (Bills et al. 2017, 293). This is precisely what scholars associated with the transdisciplinary scientific movement of the economics and sociology of convention (EC/SC) started doing in the mid-1980s (Diaz-Bone 2018). The recruitment and evaluation process, the role of intermediaries and the selective social effects of different hiring methods are a main field of investigation and theorising of EC/SC (Bessy and Eymard- Duvernay 1997; Eymard-Duvernay and Marchal 1997; Larquier and Marchal 2008; Marchal and Rieucau 2010; Marchal 2015; Larquier and Marchal 2016; Larquier and Rieucau 2019). Drawing on this conventionalist theoretical framework and data from a qualitative study on the employability of low-skilled workers in the German speaking part of Switzerland, this article examines how employers recruit workers for low-skilled jobs.1 How do firms search for candidates in the “everyman” market of low-skilled jobs, and how do they assess worker quality in the absence of educational signals? How do labour market intermediaries shape the matching process? As we will show, the increasing significance of professional intermediaries and dispositives adds to the standardisation of hiring for low-skilled jobs without completely displacing direct recruitment channels and contextualised ways of testing the quality of job applicants. We will first introduce the theoretical framework (2) and describe our methods and data (3) before we present the empirical findings on employer hiring practices (4) and draw conclusions (5). 2 A Conventionalist Approach to Recruiting The theoretical and empirical contributions of EC/SC to the study of matching processes in the labour market emerge from a critical engagement with neoclassical economic theory and organisational sociology (Diaz-Bone 2018). In particular, EC/ SC questions the basic assumption that matching is mainly an issue of the efficient circulation of information and that information or the quality of job candidates are unproblematic objective facts. In contrast, the focus of the conventionalist approach is on the preconditions for the coordination of actors in the face of the uncertainty and incommensurability of qualities (Marchal and Rieucau 2010). The operation of hiring first involves finding candidates, i. e. to address the market, and, second, selecting the person that best fits the job, which implies judgments on the relative quality of candidates. Both steps rely on prior “investments in form”, which are necessary to establish the boundaries of the relevant market to create a shared 1 The research was funded by the Swiss National Science Foundation (grant 10017_175842) and conducted by the authors together with Anna Gonon and Anna John. We thank Anna Gonon, Anna John and the anonymous reviewers for their valuable feedbacks to earlier versions of this paper. SJS 48 (2), 2022, 335–352 338 Eva Nadai and Robin Hübscher language that enables the circulation of information and to define the principles of quality judgments (Eymard-Duvernay 2008; Marchal 2015; Larquier and Rieucau 2019). For instance, educational diploma help to delineate the relevant market for skilled jobs. They define the quality of candidates, provide the terminology to address them and indicate where to find them (e. g. through job ads in trade journals or on specialised internet job boards). In the absence of such prior investments, firms are confronted with a non-organised market and must create and shape the supply and quality of candidates by means of their recruitment channels and methods (Larquier and Rieucau 2019, 60). This is the case for workers without formal qualifications. EC/SC-scholars have typified a plurality of conventions of recruiting and qual- ity judgments, which each favour or disadvantage different categories of jobseekers (Eymard-Duvernay and Marchal 1997; Larquier and Marchal 2008; Marchal and Rieucau 2010; Larquier and Marchal 2016; Larquier and Rieucau 2019). Basically, conventions are categorised with respect to the “intensity of instrumentation” (Diaz-Bone 2018, 120) of the recruitment and evaluation methods. For instance, channels and evaluations of job applicants resting on proximity (personal interaction between recruiter and candidate) require less investment in forms than recruiting and evaluation through intermediaries and dispositives (see below), e. g. through job advertisements and the screening of CVs. The use of conventions varies among countries, firm size, industry sector and characteristics of jobs (function, hierarchi- cal level, qualification). Hiring for low-skilled jobs typically follows “simplified” or “contextualised” conventions (Larquier and Marchal 2008, 22ff.). The simplified method, which is mostly used by small firms in urgent need of labour, operates without intermediaries and formal evaluation and often involves only one candidate for the job.2 The contextualised pattern tests the quality of the job applicant in real or simulated work situations, for example with trial days. EC/SC accords the firm the power of valorisation (Eymard-Duvernay 2012): it is the firm that passes judgment on the quality of job applicants. At the same time, the firm is not an isolated and unified actor but rather a “plural and complex collective” that is connected to a variety of external actors (Eymard-Duvernay 2012, 12). The fragmented nature and the embeddedness of the firm in multiple external relations raises the question of the relative power of actors in the recruiting process. On the one hand, there may be various internal actors involved with different priorities and preferences resulting in contradictory judgments which must be aligned in compro- mises. The sequence of how these actors are involved in the selection process is vital: HR professionals screening applications in the first phase of the selection process may sort out applicants that a supervisor might retain (Marchal 2015, 179ff.).3 On the 2 In a more recent paper focussing on the screening process, Larquier and Marchal (2016) use the term “streamlined” for a simple approach to screening characterised by the lack of means, which they oppose to three more formalised types. 3 The preselection phase is the most crucial: the formulation or the medium of a job advertisement excludes some applicants before they even get in contact with a recruiter (Marchal 2015, 237). SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 339 other hand, conventionalist research points to the important role of external labour market intermediaries like private and public job placement agencies. Intermediaries are not just neutral mediums for economic transactions but “active entities involved in the construction of markets and the dynamics of valuation that drive them” (Bessy and Chauvin 2013, 85). In their seminal study on recruitment Eymard-Duvernay and Marchal (1997) showed how consultants translate their interpretations of the expectations of firms into selection criteria and in this way influence the defini- tion of the quality of applicants. Besides persons and organisations, EC/SC also recognises material objects as intermediaries; these are usually termed “dispositives” (Diaz-Bone 2018, 109). A pertinent example of dispositives shaping recruitment are online job boards, which transform the language of the market through a high degree of standardisation and codification (Marchal et al. 2007). 3 Data and Methods The empirical data for the analysis of recruitment forms were collected between 2018 and 2021 in a study on the employability of low-skilled workers, i. e. workers without formal occupational qualifications. This category represents 11  percent of the working age population of Switzerland (FSO 2021a). The study comprises semi-structured qualitative interviews with employers, workers and professional labour market intermediaries, complemented by documents. Firms and workers were sampled in five industries with a large share of low-skilled jobs: catering, cleaning, construction, manufacturing and retailing (FSO 2021a).4 › Employers: the sample contains 33 interviews in 27 firms operative in the above-mentioned industries. We interviewed staff who are involved in recruit- ing: HR- and line managers and direct supervisors of low-skilled workers. The interviews addressed the function of low-skilled jobs within the production process, skills requirements, recruitment and evaluation practices and the relations to external intermediaries. Vis-à-vis our interviewees, we defined the jobs we were interested in as “unskilled” in the sense of not requiring formal occupational training. › Intermediaries: as private intermediaries, we selected three temporary work agencies with a substantial share of placements in low-skilled jobs. Because of the high unemployment rates of low-skilled workers, we also included public intermediaries: a regional employment office (RAV) of the unemployment insurance and two job placement programmes. A total of 10 interviews with 4 The FSO data pertain to the educational level of workers according to industry sectors. As Murphy and Oesch (2018, 9) note, workers’ education is a good proxy for the skills requirements of jobs. SJS 48 (2), 2022, 335–352 340 Eva Nadai and Robin Hübscher intermediaries concerning the selection of workers and cooperation with cli- ent firms were conducted. › Low-skilled workers: To capture the dynamics of employability, the workers were interviewed three times over the course of three years. The first wave comprised 39 interviews, the second wave 24 and the third wave 14. The research follows the methodological principles and coding techniques of grounded theory (Strübing 2014). Sampling, data collection and data analysis were intertwined in several sequences of sampling and fieldwork phases following pre- liminary data analyses. Regarding firms, the sample is biased in favour of medium and large companies that have professional HR departments, hence those which use more “written-based” recruiting methods (Larquier and Marchal 2016). A fifth of the sample falls into the medium category; almost two thirds have at least 250 employees. Likewise, only medium and large temporary work agencies with equally formalised operative processes agreed to participate in the study. 4 Recruiting for Low-Skilled Jobs: Channels and Judgments In the Swiss labour market, personal networks, ads on the internet and unsolicited applications are the most frequently used recruitment channels for firms; public employment services and temporary work agencies play a minor role (Buchs and von Ow 2017). These findings of the Swiss Job Market Monitor are not broken down by qualification or industry, however. As we will show, professional intermediaries are important in the segment of low-skilled jobs. In the following presentation of our empirical data, we will first discuss direct recruitment channels (4.1), then methods involving different types of intermediaries (4.2–4.4) followed by the forms of judgment (4.5). 4.1 Direct Recruitment Recruiting does not always start with a job offer. It can also happen the other way around, with the starting point being an offer of labour. This is the case for unsolicited applications, which can take many forms, such as walk-in applications, sending docu- ments to a passed-around phone number of a supervisor or filling out an online form on a company website. The walk-in application as a convenient way for low-skilled jobseekers to present themselves personally to potential supervisors is on the decline in the firms in our sample. Jobseekers still try this method in stores or restaurants, but they are then requested to hand in their formal application to the personnel department or upload the documents on the company website’s application tool. Firms in the construction and manufacturing industries send job applicants to the TWA with which they usually cooperate (see 4.4). Thus, walking in does not lead SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 341 immediately to a face-to-face interaction with decision makers as a chance to make a good impression and avoid competition (Salognon 2007) – most often jobseekers cannot circumvent the impersonal at-a-distance screening.5 In this way, the once direct and low-threshold channel of walk-in applications is increasingly formalised through the use of centralised recruitment and screening methods (Rieucau 2015). In the cleaning industry, there is a particular way of direct recruitment. In maintenance cleaning, contracts for “objects” to be cleaned are regularly put out to tender. Sometimes the successful bidder then takes over the cleaners who worked on this object from the previous employer because these workers are known to be available for the required schedule and location. Furthermore, re-employment in the same company happens first in the context of seasonal temporary employment when firms ask the TWA for the same proven workers each year and second in catering and cleaning where some companies build their own labour pool for short-term demand (e. g. for one-off catering events or for holiday replacements). 4.2 Network Recruiting Much more common than walk-ins, indeed ubiquitous, is recruiting through personal contacts – either as a form of unsolicited application by jobseekers or initiated by the employer, who invites employees to actively pass around job openings in their networks. We found recruiting through personal intermediaries in all the industries we studied but operating with different degrees of formatting. The cleaning com- panies mark the unformatted side of the spectrum, where the recruiting is done primarily by direct supervisors who rely on the networks of their teams and their own judgment. In the construction and manufacturing industry, networks play an important role as well, but, with few exceptions, jobseekers will be referred to TWAs and first employed through them (see 4.4). This implies that they have to pass the document-based application procedures of these agencies. In catering and retailing, recommendations are usually made for concrete job openings, and the applications gathered through this channel are then screened by the personnel department (with the recommendation taken into account). This channel is not only low-cost but also offers employers several advantages, like expanding the pool of applicants, improving the match of applicants and jobs and facilitating the new hires’ socialisation into the organisation (Fernandez et al. 2000; Marsden and Gorman 2001). The advantage that employers most frequently mentioned is the good quality of referrals. Because the referrers act as quality warrantors, they are concerned with their own reputation, as the HR manager of a retail chain explains: “These are usually good people because [our employees] don’t recommend anyone if they can’t stand behind 5 This finding is partly due to the sample bias in favour of medium and large firms. Yet, the few small companies do not report walk-in applications as a relevant channel either. Instead they rely heavily on personal contacts (see 4.2). The main difference from larger firms is that there is no standardised prescreening. SJS 48 (2), 2022, 335–352 342 Eva Nadai and Robin Hübscher them”. Referrals improve the matching because of the extra information flowing in both directions. On the one hand, referrers inform candidates about what the job is really like, also in terms of cultural dimensions that are hard to communicate in job ads. On the other hand, employers use referrers as a source of information on the candidate. This may take the form of a quasi-job interview as the manager of a cleaning company explains: “Normally, before he [district manager] invites her [recommended worker], he has a talk with this lady or gentleman, who recommends her and asks what she has done already, how she is and why we should hire her”. The homophily of networks, entailing the risk of discrimination, is a much- debated issue in the literature on labour market matching (Fernandez et al. 2000; Bills et al. 2017). The fact that candidates referred through personal contacts are believed to be a carbon copy of the referring worker (Waldinger 1997, cited in Bills et al. 2017, 299) reduces uncertainties about applicant quality. Especially in the case of migrants, similarity pertains to the cultural milieu, for example the country or region of origin from which employers infer the attitude towards work (Waldinger and Lichter 2003). In the eyes of the firms we studied, the possibility to discrimi- nate – in the sense of making stereotypical distinctions – is actually an advantage. In manufacturing, explicit gender or ethnical stereotyping of jobs on the basis of physical strength or dexterity is common: a food processing plant, for example, prefers Asian women for fileting fruit with a precision and speed “no machine could ever do”, as the manager said. However, using such criteria in advertising might damage the reputation of the firm or even be illegal. But exploiting the homophily of networks for recruiting also has more tangible reasons. Because the low-skilled workforce is mainly composed of migrants with often limited command of the local language, network recruiting ensures linguistically homogenous teams that are able to communicate between themselves. The similarity of workers is thus functional and necessary for the coordination of production. 4.3 Job Advertisements The use of job advertisements as technical intermediaries for at-a-distance recruit- ing varies among industries: they are common in retailing and catering but are the exception in cleaning, construction and manufacturing. In the latter three industries, jobs are only made public when other channels do not yield sufficient candidates or the positions require experience (e. g. machine operator). In retailing and cater- ing, job ads address a mixed market of workers with or without formal training. There is a severe shortage of trained workers; hence, a formal qualification is not an indispensable criterion. Moreover, there is a widespread belief that crucial qualities like friendliness or the “sales gene” cannot be acquired through formal training but are innate personal traits (Nickson et al. 2017; Nadai and Gonon 2021). Yet, for jobs that are explicitly designated as unskilled entry-level jobs, ads are considered unnecessary as this HR manager of a large retail chain explains: “We have piles of SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 343 unsolicited applications from which we can draw. (…) If you place a job ad, within a day you’ll have between two and three hundred applications. It would be (laughs) almost a bit masochistic”. If firms do publish ads for low-skilled jobs at all, they place them in “semi- public” places, so to speak, in their stores or restaurants, in customer magazines or on their company websites but not on general job boards or in print newspapers.6 In-store advertisements, besides being cheap, bring the advantage of narrowing down the potential candidates to people living close by, which was often mentioned as an important criterion for jobs with irregular working hours (Rieucau 2015, 145). Furthermore, targeting customers is a way to find workers who identify with the brand or product, which is a quality demanded in interactive service jobs (Nickson et al. 2017; Monchatre 2018; Nadai and Gonon 2021). Whether in print or online, advertisements for low-skilled jobs mainly fulfil the function of making public a labour demand. The requirements entail, amongst others, availability for the exact working hours and/or a willingness to work irregular shifts, a valid work permit and a level of oral knowledge of the local language. Ads for retailing jobs are more elaborate but still generic in that the same description is used repeatedly to fill multiple open positions. This is exemplified by the job site of a clothing chain, where there is a task description and a list of requirements. The open positions can be found separately with only limited information, such as “Town X – flexible part-time sales assistant ca. 10 %–30 %, also for sickness and holiday replacement. Start: Immediately or by appointment”. Candidates applying for a job can simply select from a list with all the open positions if they simultane- ously want to apply for jobs in other shops of the company. The high reach and low costs of online advertisements come at the expense of accuracy. Although firms can decide what to put in their job advertisements, they cannot prevent unsuitable applications. “You can write whatever you want”, a produc- tion manager complained, “everyone is applying (…) nobody cares”. E-recruitment tools shield employers from excessive demand from (unsuitable) jobseekers. With application blanks for personal information and upload fields for documents, such as CVs, companies make sure that they get the information they want for screening applicants efficiently. If some information or document is declared mandatory in the tool, the application cannot be completed without it. In the retailing and catering firms in our sample, the CV is an indispensable upload, whereas this is less often the case for a letter of motivation. 6 Marchal and Rieucau (2010, 45) classify the posting of jobs within the firm’s premises as direct recruitment. However, insofar as job ads in shops or restaurants target an external labour market and require candidates to hand in written applications, they function like dispositives for at-a- distance hiring. SJS 48 (2), 2022, 335–352 344 Eva Nadai and Robin Hübscher 4.4 Professional Intermediaries Finally, professional intermediaries constitute another important recruitment chan- nel for low-skilled jobs, namely private TWAs and public employment services. In Switzerland, the share of workers employed through TWAs is growing, especially in the low-skilled sector (FSO 2021b); internationally, the use of TWAs for low-skilled jobs is high (Tardos and Pedersen 2011). The use of public employment services is mandatory in many low-skilled jobs because of a legal job registration obligation in jobs with a high unemployment rate.7 Temporary work agencies constitute the main recruiting channel in the construc- tion and manufacturing industries. The higher gross costs of using TWAs are offset by a number of advantages for employers, which counterbalance direct financial costs (Elcioglu 2010). The main reason for the firms in our sample is a fluctuating demand for workers, often on short notice, such as the start of a new construction site or a large order for a specific product. Second, by using TWAs, firms externalise the time-consuming and costly tasks of recruitment and personnel administration. Third, TWAs function like a “product warranty” for labour insofar as firms can easily replace workers who do not fulfil quality standards or who fall ill. Finally, temporary workers may serve as a pool to fill permanent positions. In this case, the temporary assignment serves as an extended trial period. The relation between TWAs and the firms using them as a recruitment channel is complex. As a supplier of labour, the TWA is an intermediary, yet it is also the employer of the workers it lends to the client firm. As an employer, it could there- fore be expected to have a substantial influence on defining the quality of workers. On the other hand, TWAs operate in a highly competitive market with many small agencies and a few big players. In particular, if they aim for long-term contracts for “bulk labour supply” (Purcell et al. 2004, 709), they have to meet the quantitative and qualitative demands of client firms as exactly as possible.8 Hence, although the TWAs are de jure employers, their power of valorisation is actually limited: we did not observe that they define the quality of labour, nor was there any evidence that TWAs negotiate job profiles with client firms. While they are powerful vis à vis jobseekers through their screening function, they have little influence on the demands of client firms. Rather, by following the specifications of the client firms, who are the de facto employers, the intermediaries extend the reach of the latter’s labour quality conventions (Larquier and Rieucau 2019, 58). TWAs deliver four kinds of services: building a labour pool, selecting individual workers for a concrete order, tailoring the quality of labour and providing personnel administration. Building a pool of workers operates like the recruitment methods 7 If the unemployment rate in a job category is higher than 5 %, employers must report job openings to the RAV. The jobs must be published on the RAV online job board for at least five days before employers are permitted to offer them to non-unemployed candidates. 8 The firms in our sample mostly cooperate with more than one TWA to ensure supply in case of short-term labour demand. SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 345 used by the client firms themselves. Word-of-mouth and online ads are by far the most important channels (Swissstaffing 2018, 13). The preferred application method is online, including the uploading of documents like CVs or references; this is fol- lowed by a personal interview. The selection of workers for this pool is guided by the anticipated market demand represented by the (ideally) long-term client firms. Hence, the requirements of TWAs are typically as generic as those formulated in the job ads mentioned above. Accordingly, selecting workers for a specific assignment is less a matter of finding exactly the right person but rather of finding a suitable worker as fast as possible. While the screening is done by matching the client firm’s specifications with the electronic database of the TWA, the actual selection from the still-large pool of possible candidates follows the first come, first served principle, as a TWA recruiter stated: “Most of the time we say, ‘the first one who picks up the phone’”. Moreover, TWAs deliver “ready-made” workers by preparing them for their tasks. Tailoring the quality of labour might entail conducting short mandatory courses on hygiene and safety standards, providing safety gear and the like. In this way, TWAs produce a stock of easily deployable workers. In addition to the screening of large numbers of applications, administrating the recruitment involves replacing workers who have been found unfit, obtaining official documents like work permits, or arranging mandatory medical tests. As long as workers are employed by the TWA, the administration extends to personnel management and payroll. From the point of view of client firms, TWAs thus save costs and risks beyond the recruitment. While some companies we interviewed use TWAs as their main channel, this is not the case for public intermediaries, which are used complementary to job ads or personal contacts. While TWAs can freely select the best candidates for their pool, the RAV must also place jobseekers carrying the stigma of unemployment (Bonoli and Hinrichs 2012; Marchal 2015). Yet, Liechti (2020) argues that recommendations by a public intermediary can have a positive impact on hiring chances. The firms in our sample are rather sceptical. They suspect that the unemployed lack motivation and only apply because they must prove that they are actively search for a job. These reservations also refer to the job registration obligation: in the eyes of the employers interviewed, the mandatory ads result in a lot of extra work but very few suitable candidates. Thus, matching on the basis of predefined criteria seems to be rather unsuccessful. Indirectly, the RAV still constitutes a significant recruitment channel albeit not for firms but for other intermediaries: TWAs rate the job board of the RAV and personal contacts of RAV counsellors as very important sources to build a pool of workers (Swissstaffing 2018, 13). Thus, recruitment chains are extended, and an additional screening step is built into the hiring process, which increases selectivity. Another public intermediary, which was used occasionally or regularly by five firms in our sample, are work placement programmes for the unemployed. As free trial periods, these placements provide subsidised labour for firms in the regular labour market because workers do not receive a salary but social insurance SJS 48 (2), 2022, 335–352 346 Eva Nadai and Robin Hübscher benefits (Nadai et al. 2018). From the perspective of unemployed jobseekers, the direct placement eliminates competition and ensures personal contact with the em- ployer. In this way, the document-based early sorting out of low-skilled applicants is avoided (Salognon 2007). 4.5 Evaluating the Quality of Job Applicants Labour quality conventions for low-skilled workers differ among industries, jobs and work organisations, but some features are consistently important (Nadai and Gonon 2021). The most crucial for the sample firms are physical fitness in terms of strength, stamina and robustness (also Hieming et al. 2005; Hassler et al. 2019) and fitting in, i. e. the willingness to work hard and to accept one’s place within the hierarchical order of the firm (Waldinger and Lichter 2003; Smith and Neuwirth 2009; Tranchant 2018). Qualities regarding specific skills like oral and written command of the local language or digital literacy are more variable among indus- tries and firms. The judgment of these qualities is rarely as straightforward as the “streamlined” process (Larquier and Marchal 2016). Most often, evaluation com- bines “written-based screening” (Larquier and Marchal 2016) involving CVs and references along with face-to-face interviews and practical tests (trial days on the job). Except in three owner-managed small cleaning enterprises, there are normally at least two actors or functions involved, and there is usually more than one (albeit brief ) interview with the candidate. In the most elaborate form, the whole process involves four main steps: 1. centralised document-based screening by HR staff, 2. short telephone interview conducted by HR staff, 3. practical trial, and 4. a face- to-face interview with the supervisor. The first screening is almost always done by HR professionals – either in the firm where the candidates will work or in the TWA that recruits them. CVs, refer- ences, motivation letters (less often), and sometimes official documents, like work permits, must be provided. This phase is selective in catering and retailing, where online ads and application tools generate large numbers of applications, which can be sorted through the use of keywords without someone actually looking through the applications (Marchal et al. 2007). As Remillon (2019, 114) notes, this kind of criteria-based automated sorting amounts to a “denial of evaluation”. After the prescreening of CVs, the remaining applications are scrutinised in more detail with regard to work experience, job changes and competences. Employers in these two industries express higher expectations, especially regarding the level of language and communication skills. In manufacturing and construction, where TWAs are involved, selectivity depends on whether candidates are referrals through personal contacts who have been sent to the TWA for administrative processing only or whether they first have to pass a screening by the agencies for their own pool of workers. In these industries and in cleaning, CVs are also demanded, but they are regarded as less informative than interviews and practical tests. SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 347 Short telephone interviews following the screening of documents are also in the hands of HR staff. One of the main functions is to test the language skills of candidates. We observed a certain distrust of written documents, especially as proof of the language level of migrants. Managers argued that CVs or letters might not have been written by the candidates personally but by someone with better skills. Likewise, they believe that “you could write anything in a reference letter” or in a CV. Therefore, the interview gives a more accurate impression of the person. The required language level varies greatly among industries, jobs and firms. Generally, it is higher in interactive service jobs (waitressing, salespersons) and very low on the assembly line, on construction sites and in cleaning. Nevertheless, some manufac- turing firms have candidates take formal language tests adapted to the needs of the firm but carried out by the TWA. The next two steps, the trial and the face-to-face interview with a line manager or supervisor, are intertwined in practice: the candidate might first have a short telephone interview and then be called in for a work trial period of between a few hours and a few days, following which they have a personal interview with the su- pervisor and/or a higher line manager. As a rule, the evaluation of direct supervisors is decisive. Yet, the “gut feeling” of recruiters emphasised by many studies (Hieming et al. 2005; Imdorf 2010; Marchal 2015; Monchatre 2018; Hassler et al. 2019) is only part of the picture. First, recruiters are able to break down their intuition in specific dimensions. For instance, they use their own methods of observing the crucial criteria of physical fitness or commitment: I simply SEE whether I can use someone. (…) You see, if someone comes and walks like this [mimics stiff gait], you can be sure that person never bends over. (…) They walk so slowly, or they get up and just leave the chair standing. You can’t use someone like this in maintenance cleaning, they won’t place the chairs for clients either. (HR manager, cleaning company) Second, in addition to physical and general behavioural criteria, like the candidate’s “drive”, physique, well-groomed appearance, or friendliness, there is a systematic assessment of technical aspects of job performance during the practical trial: How does someone receive the goods, control the goods, unpack them, secure them (…) how do they present them. There you feel a bit does someone have FLAIR, then you certainly see their pace very quickly, then decoration (…) their creativity. (Store manager, retailing) While not constituting a formal test, such assessments still are methodical and clearly pertinent to the specific requirements of the job. Moreover, some firms format them by means of standardised checklists for the trial day which the supervisor must use as a basis for judgment. We did not find quantitative measures of performance on the trial day though (but see Tranchant 2018). If there is a TWA involved, the trial SJS 48 (2), 2022, 335–352 348 Eva Nadai and Robin Hübscher period is in a way unlimited in time since the firms can always have “low-quality” workers replaced by the agency on short notice. 5 Conclusions Hiring is an inherently risky decision because it involves a prediction about future job performance on the basis of a person’s previous training, work experience and behaviour. Hiring low-skilled workers exacerbates this uncertainty because they lack the foremost labour market signal: a formal qualification. Moreover, the decline in low-skilled jobs has led to an excess supply of unqualified workers, which manifests itself for employers as “piles of applications” on their desks. At the same time, em- ployers notice that “good people are hard to find”, as the owner of a small cleaning company says. This paper has examined hiring practices for jobs that do not require formal education and training: how do employers find the “good people” in the “piles of applications” for low-skilled jobs? The most popular answer in sociological research is that firms shield themselves from the effort and costs of handling a potential mass of applications by resorting to informal methods, like network recruiting and trial days (Hieming et al. 2005; Larquier and Marchal 2008; Hassler et al. 2019). From the point of view of the sociology of conventions, “informal” does not equal “formless”. Rather, the question is to what extent and in what ways matching rests on prior form investments that shape the ways of defining and assessing qualities. The research presented here shows a complex blend of practices with differ- ent degrees of instrumentation. Overall, finding and assessing low-skilled workers is not a straightforward process, and there is a trend towards the standardisation of recruitment and screening methods (Rieucau 2015; Remillon 2019). The sig- nificance of written-based practices in our study is partly due to the sample bias in favour of medium and large firms with professional HR departments. However, formalisation is also driven by the rise of temporary employment as well as by the job registration obligation for low-skilled jobs in the industries studied. The latter makes the public employment service RAV and its digital dispositives mandatory intermediaries. Thus, the elaborate four-step hiring process described above (4.5) is the exception if we only regard the firm where labour is actually deployed but more common if intermediaries are taken into account. Apart from firm size, the formats and involvement of intermediaries vary among industries. The least formatted recruitment and quality evaluation is practised in the cleaning industry. Maintenance cleaning is a classical entry-level job for immigrants, with very low pay and skills requirements as well as regular turnover of cleaning objects, which often implies staff turnover. Recruiting must therefore be efficient and cheap. It is mainly in the hands of direct supervisors who tap into the personal networks of their teams and trust their own impressions from a face-to-face interview SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 349 and (in half of the firms) from a practical trial. Regarding written elements, CVs serve mainly administrative purposes, whereas the absence or presence of a work permit has a selective effect. The recruitment chain is longer in manufacturing and construction because of the delegation of hiring to external intermediaries, namely TWAs. Seen from the perspective of the hiring firm, the process is very simple: the responsible HR staff calls the TWA and “orders” a certain number of workers without any further screening because a bad match can easily be replaced by the TWA. Defining the job profile and the quality convention has to be done only at the beginning of the cooperation with a TWA or when the requirements change due to technological or organisa- tional transformation. However, behind the scenes the hiring process still involves elements of written-based recruiting and screening, but these steps are delegated to the TWA. Delegating recruitment and personnel administration to the TWA as the de jure employer provides firms with on-demand labour with no strings attached. In fact, through the involvement of the TWA, labour is stripped from the legal and implicit obligations of the work contract and indeed becomes a mere production factor, like machines. For the firms in our sample, this commodification of labour seems to be the more pertinent reason for delegating recruitment than just finding a sufficient number of workers. Recruiting and quality evaluation are closest to the at-a-distance hiring and written-based convention of assessment (Larquier and Marchal 2016) in retailing and catering, even though network recruiting and practical trials are also widely used. There are several reasons for the more elaborate practices: first, skills demands are higher than in the other three industries, especially regarding language and com- munication. Second, firms address an organised occupational market (Larquier and Rieucau 2019, 59), but because of the skills shortage and because essential qualities are believed to be personal traits rather than trained skills, they regularly accept applicants without a diploma.9 Third, restaurants in particular have a high staff turnover, i. e. a constant labour demand. Document-based online recruitment tools allow efficient processing by a centralised HR department. Hiring for low-skilled jobs increasingly involves professional and technical intermediaries. This leads to longer recruitment chains on the one hand and to the formalisation of valorisation on the other hand. We presume that the higher significance of at-a-distance recruitment has negative effects on low-skilled workers. EC/SC research has repeatedly shown that low-skilled workers profit from channels based on proximity and from evaluation in concrete work situations because this reduces competition and the impact of standard criteria like diplomas (Salognon 2007; Marchal and Rieucau 2010; Larquier and Marchal 2016; Laquier and Rieucau 2019). Thus, “the reduction of a plural space of valorization” (Remillon 2019, 112) 9 These two arguments pertain mainly to frontstage workers with customer contact, e. g. salespersons and waitresses, not to backstage functions, like dishwashers. SJS 48 (2), 2022, 335–352 350 Eva Nadai and Robin Hübscher to a standard model puts this category of workers at a disadvantage. Furthermore, the involvement of TWAs implies precarisation: it becomes more difficult to get a permanent job (Elcioglu 2010; Promberger 2012). Yet, as network recruitment is widespread, personal contacts still have an influence on evaluation. Recommenda- tions are factored in when referred candidates have to pass the screening by HR professionals and employers trust the judgment of their workers vouching for the quality of their relatives or friends. Professional intermediaries follow the specifications of client firms in the screen- ing of job applicants, and if an order is urgent, they simply select the first person in their labour pool who “picks up the phone”. There is no evidence, however, that TWAs define or negotiate the job profile or the quality of workers to a significant degree. Rather, involving TWAs extends the reach of the client firms’ quality con- ventions (Larquier and Rieucau 2019). The power of valorisation (Eymard-Duvernay 2012) remains with the companies that actually use the workers in production: they specify the requirements for a job and ultimately accept or reject the workers sent by the TWA. Within the firms, the power is distributed between professional HR staff, management and supervisors of low-skilled workers. HR conducts the first screening and in some cases formats the evaluation by drafting interview guides and checklists for practical trials to be used by supervisors. As a rule, it is the supervi- sors who have the final say: they judge the quality of candidates and decide which individuals stand out in the nondescript mass of unqualified workers through their physical capacities, their willingness to work hard and their fit in the social order of the firm and the team (Nadai and Gonon 2021). 6 References Abel, Jörg, Hartmut Hirsch-Kreinsen, and Peter Ittermann. 2014. Einfacharbeit in der Industrie. Struk- turen, Verbreitung und Perspektiven. Berlin: edition sigma. Bessy, Christian, and François Eymard-Duvernay. 1997. Les intermédiaires du marché du travail. Paris : Presses universitaires de France. Bessy, Christian, and Pierre-Marie Chauvin. 2013. The Power of Market Intermediaries: From Information to Valuation Processes. Valuation Studies 1(1): 83–117. Bills, David B., Valentina Di Stasio, and Klarita Gërxhani. 2017. The Demand Side of Hiring: Employers in the Labor Market. Annual Review of Sociology 43(1): 291–310. Bonoli, Giuliano, and Karl Hinrichs. 2012. Statistical Discrimination and Employers Recruitment. Practices for Low-Skilled Workers. European Societies 14(3): 338–361. Buchs, Helen, and Anna von Ow. 2017. Personalsuche und Stellenbesetzung: Kurzreport zur Unternehmens- befragung 2017. Zürich: Universität Zürich. Can, Ensar, and George Sheldon. 2017. Die Entwicklung der Beschäftigungschancen von Geringqualifizierten in der Schweiz. Zürich: Sozialdepartement der Stadt Zürich. Diaz-Bone, Rainer. 2018. Die Economie des conventions: Grundlagen und Entwicklungen der neuen fran- zösischen Wirtschaftssoziologie. 2nd ed. Wiesbaden: Springer Fachmedien. SJS 48 (2), 2022, 335–352 “The First One to Pick up the Phone”: Forms of Recruitment for Low-Skilled Jobs 351 Elcioglu, Emine Fidan. 2010. Producing Precarity: The Temporary Staffing Agency in the Labor Market. Qualitative Sociology 33(2): 117–136. Eurofound. 2015. Upgrading or Polarisation? Long-Term and Global Shifts in the Employment Structure: European Jobs Monitor 2015. Luxemburg: Publications Office of the European Union. Eymard-Duvernay, François. 2008. Justesse et justice dans les recrutements. Formation emploi 101: 55–69. Eymard-Duvernay, François. 2012. Epreuves d’évaluation et chômage. Toulouse: Octarès Editions. Eymard-Duvernay, François, and Emmanuelle Marchal. 1997. Façons de recruter : Le jugement des com- pétences sur le marché du travail. Paris: Editions Métailié. Fernandez, Roberto M., Emilio J. Castilla, and Paul Moore. 2000. Social Capital at Work: Networks and Employment at a Phone Center. American Journal of Sociology 105(5): 1288–1356. FSO (Federal Statistical Office). 2021a. Bildungsstand der Bevölkerung (SAKE). Table, https://www. bfs.admin.ch/bfs/de/home/statistiken/kataloge-datenbanken/tabellen.assetdetail.16324563.html (29.04.2021). FSO (Federal Statistical Office). 2021b. Arbeitnehmende, die ihre Stelle über ein privates Stellenver- mittlungsbüro gefunden haben (SAKE). Table, https://www.bfs.admin.ch/bfs/de/home/aktuell/ neue-veroeffentlichungen.assetdetail.16724638.html (10.06.2021). Gesthuizen, Maurice, Heike Solga, and Ralf Künster. 2011. Context Matters: Economic Marginalization of Low-Educated Workers in Cross-National Perspective. European Sociological Review 27(2): 264–280. Hassler, Benedikt, Lea Widmer, Thomas Geisen, Nathalie Amstutz, Nico Scheidegger, and Nadine Wenger. 2019. Arbeitsplätze ohne formale Qualifikationsanforderungen in der Schweiz. Rekru- tier ungsstrategien von Unternehmen und deren sozialpolitische Bedeutung. Zeitschrift für Sozialreform 65(2): 147–174. Hieming, Bettina, Karen Jaehrling, Thorsten Klaina, Achim Vanselow, and Claudia Weinkopf. 2005. Stellenbesetzung im Bereich „einfacher“ Dienstleistungen. Abschlussbericht einer Studie im Auftrag des Bundesministeriums für Wirtschaft und Arbeit, BMWA-Dokumentation Nr. 550. Berlin: BMWA. Imdorf, Christian. 2010. Emotions in the Hiring Process: How “Gut Feelings” Rationalize Personnel Selection Decisions. Pp. 84–105 in Emotionalizing Organizations and Organizing Emotions, edited by Barbara Sieben and Åsa Wettergren. New York: Palgrave Macmillan. Larquier, Guillemette de. 2016. Une approche conventionnaliste du marché du travail fondée sur le recru- tement des entreprises. Paris : Université de Paris Ouest-Nanterre la Défense. Larquier, Guillemette de, and Emanuelle Marchal. 2008. Le jugement des candidats par les entreprises lors des recrutements. Document de travail No  109. Centre d’Etudes de l’emploi, https://ceet. cnam.fr/medias/fichier/109-jugement-candidats-entreprises-recrutements_1508241790407- pdf?ID_FICHE=1050486&INLINE=FALSE (15.7.2021). Larquier, Guillemette de, and Emanuelle Marchal. 2016. Does the Formalization of Practices Enhance Equal Hiring Opportunities? An Analysis of a French Nation-Wide Employer Survey. Socio-Eco- nomic Review 14(3): 567–589. Larquier, Guillemette de, and Géraldine Rieucau. 2019. Valuation and Matching: A Conventionalist Explanation of Labor Markets by Firms Recruitment Channels. Historical Social Research 44(1): 52–72. Liechti, Fabienne. 2020. Connecting Employers and Workers: Can Recommendations from the Public Employment Service Act as a Substitute for Social Contacts? Work, Employment and Society 34(4): 587–604. Marchal, Emmanuelle. 2015. Les embarras des recruteurs. Enquête sur le marché du travail. Paris : Editions EHESS. Marchal, Emmanuelle, Kevin Mellet, and Géraldine Rieucau. 2007. Job Board Toolkits: Internet Match- making and Changes in Job Advertisements. Human Relations 60(7): 1091–1113. SJS 48 (2), 2022, 335–352 352 Eva Nadai and Robin Hübscher Marchal, Emanuelle, and Géraldine Rieucau. 2010. Le recrutement. Paris : La Découverte. Marsden, Peter V., and Elizabeth H. Gorman. 2001. Social Networks, Job Changes, and Recruitment. Pp. 467–502 in Sourcebook of Labor Markets, edited by Ivar Berg and Arne L. Kalleberg. Boston: Springer. Monchatre, Sylvie. 2018. Embaucher ou habiliter ? Recrutement et espaces de (non-) qualification dans l’hôtellerie-restauration. Travail et emploi 155–156(3): 93–114. Murphy, Emily C., and Daniel Oesch. 2018. Is Employment Polarisation Inevitable? Occupational Change in Ireland and Switzerland, 1970–2010. Work, Employment and Society 32(6): 1099–1117. Nadai, Eva and Anna Gonon. 2021. “Simple Jobs” for Disqualified Workers. Employability at the Bottom of the Labour Market. Pp. 199–221, in The Future of Work, edited by Christian Suter, Jacinto Cuvi, Philip Balsiger and Mihaela Nedelcu. Zürich: Seismo. Nadai, Eva, Anna Gonon, and Fabienne Rotzetter. 2018. Costs, Risks and Responsibility. Negotiating the Value of Disabled Workers Between Disability Insurance and Employers. Swiss Journal of Sociology 44(3): 405–422. Nickson, Dennis, Robin Price, Hazel Baxter-Reid, and Scott A. Hurrell. 2017. Skill Requirements in Retail Work: The Case of High-End Fashion Retailing. Work, Employment and Society 31(4): 692–708. OECD. 2017. OECD Employment Outlook 2017. Paris: OECD Publishing. OECD. 2019. OECD Employment Outlook 2019: The Future of Work. Paris: OECD Publishing. Promberger, Markus. 2012. Topographie der Leiharbeit. Flexibilität und Prekarität einer atypischen Be- schäftigungsform. Berlin: edition sigma. Purcell, John, Kate Purcell, and Stephanie Tailby. 2004. Temporary Work Agencies: Here Today, Gone Tomorrow? British Journal of Industrial Relations 42(4): 705–725. Remillon, Delphine. 2019. Unemployment and Turning Points in Careers. A Conventionalist Analysis. Historical Social Research 44(1): 92–118. Rieucau, Géraldine. 2015. Getting a Low-Paid Job in French and UK Supermarkets: From Walk-in to Online Application? Employee Relations 37(1): 141–156. Salognon, Marie. 2007. Reorienting Companies’ Hiring Behaviour: An Innovative “Back-to-Work” Method in France. Work, Employment and Society 21(4): 713–730. Smith, Vicki, and Esther B. Neuwirth. 2009. Temporary Help Agencies and the Making of a New Employment Practice. Academy of Management Perspectives 23(1): 56–72. Spence, Michael. 1973. Job Market Signaling. The Quarterly Journal of Economics 87(3): 355–374. Strübing, Jörg. 2014. Grounded Theory: Zur sozialtheoretischen und epistemologischen Fundierung eines pragmatistischen Forschungsstils. 3. Aufl. Wiesbaden: Springer VS. Swissstaffing. 2018. Die Temporärarbeitenden in der Schweiz 2018 – Wer sie sind. Wo sie arbeiten. Wie zufrieden sie sind. Dübendorf: Swissstaffing. Tardos, Katalin, and Jacob Jepsen Pedersen. 2011. Recruitment Channels and Strategies for Employing the Low-Skilled Workers in Europe. Review of Sociology 4: 95–112. Tranchant, Lucas. 2018. L’intérim de masse comme vecteur de disqualification professionnelle. Le cas des emplois ouvriers de la logistique. Travail et emploi 155–156: 115–140. Waldinger, Roger, and Michael I. Lichter. 2003. How the Other Half Works: Immigration and the Social Organization of Labor. Berkeley: University of California Press. SJS 48 (2), 2022, 335–352