Next Article in Journal
Linking Soil C:N Stoichiometry to Greenhouse Gas Balance: Implications for Ecosystem Sustainability in Temperate Forests
Previous Article in Journal
Balancing Cultural Values and Energy Transition: A Multi-Criteria Approach Inspired by the New European Bauhaus
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Understanding the Drivers of Temporary Agency Work in Slovenia: Implications for Sustainable Labor Practices

Faculty of Information Studies in Novo Mesto, 8000 Novo Mesto, Slovenia
Sustainability 2025, 17(24), 11261; https://doi.org/10.3390/su172411261
Submission received: 14 August 2025 / Revised: 2 December 2025 / Accepted: 11 December 2025 / Published: 16 December 2025

Abstract

Temporary agency work (TAW) has expanded globally as organizations seek flexibility amid skill shortages and demand volatility. In 2015 the United Nations recognized ‘decent work’ as Sustainable Development Goal (SDG 8), emphasizing sustainable economic growth, fair employment opportunities accessible to all without discrimination, environmental responsibility, and social inclusiveness. This study examines why user organizations (clients) adopt TAW and how these drivers materialize in stakeholder practices that align—or fail to align—with SDG-8 dimensions of decent work. Within a qualitative-dominant, explanatory sequential mixed-methods case study, documentary and statistical analyses were combined with 19 semi-structured interviews across agencies, clients, agency workers, trade unions, and relevant authorities. Inductive thematic analysis identified seven demand-side driver categories and assessed their effects using the SDG-8 pillars as an analytical lens (employment creation, rights at work, social protection, social dialogue). Findings indicate that TAW is primarily deployed to buffer volatility and labour shortages, accelerate hiring, and shift HR administration and parts of risk to agencies, with limited integration of SDG-8–consistent practices. Three cross-cutting gaps emerged: (i) social dialogue is narrow and compliance-oriented, with little strategic focus on decent-work outcomes; (ii) agency-worker voice and representation are weak, and agencies are not consistently recognised as social partners; and (iii) social-sustainability efforts are sparse and ad hoc, with few structured measures for skill development, equal treatment, or clear conversion pathways, while environmentally friendly initiatives are almost completely absent. In Slovenia, TAW fills systemic labour gaps but remains weakly integrated with SDG-8 practices. The study links demand-side drivers to specific decent-work shortfalls and proposes a multi-level policy roadmap—regulatory, industry, TAW agency, and social-dialogue platforms—to advance progress toward social sustainability and environmental responsibility.

1. Introduction

Private employment agencies have been at the forefront of labour policy discussions for more than a quarter of a century, playing an instrumental role in striking a balance between flexibility of work arrangements and job and social security within global markets [1,2]. In this paper, temporary agency work (TAW) refers to triangular arrangements in which a private employment agency both employs a worker and supplies that worker to a client organization (public or private) for a limited assignment. TAW is a form of non-standard employment, which refers to work arrangements that depart from the ‘standard’ employment relationship—i.e., full-time, open-ended, direct, and typically on-site employment with a single employer under a bilateral contract [3]. Other non-standard forms include temporary/fixed-term contracts, part-time and casual work, seasonal work, platform work, and various types of (often dependent) self-employment. Such arrangements are typically characterized by one or more of the following: limited job security or tenure, variable hours or pay, triangular or mediated employment relationships, weaker access to social protection and collective representation, and fragmented career progression [4].
Over the past two and a half decades, the use of TAW has increased across the EU [5,6,7], with evidence of effectiveness in addressing diverse socioeconomic challenges [8]. In line with the SDG 8 understanding of sustainable development as simultaneously economic, social and environmental [9], TAW is increasingly positioned not only as a flexibility instrument, but also as a potential lever in wider transition strategies. For client organizations, TAW enables rapid, demand-responsive scaling and supports project-based or seasonal operations [10], while digitalization and platform-enabled matching have further accelerated speed-to-hire and coordination efficiency [11,12]. When deployed responsibly, these same mechanisms can generate resource-related co-benefits—for example, by enabling remote or hybrid placements that reduce commuting or by optimising the use of existing facilities and infrastructure—thereby contributing to improved environmental performance alongside social and economic gains [13]. Agencies—the principal actors in TAW—reallocate labour and facilitate job transitions, a role underscored during the COVID-19 pandemic and the conflict in Ukraine, and amid subsequent EU-wide skills and labour shortages [14]. In the EU, they also help bridge labour-market gaps for vulnerable groups (persons with disabilities, young job seekers, migrants, older workers) [15], although considerable potential remains untapped [16].
At the same time, risks and criticisms are well documented. Agency workers—like many in non-standard arrangements—were among the most affected by the pandemic, experiencing job and income loss and declines in well-being [17,18,19,20]. Concerns persist regarding inadequate legislation or enforcement, unequal or unfair treatment, and lapses in health and safety [21,22,23,24,25], alongside evidence of regulatory evasion by some clients and/or agencies [24]. These challenges highlight the need to avoid reinforcing labour-market segmentation and to support decent working conditions as part of broader sustainability goals [2,25].
International and regional frameworks provide the normative baseline. The International Labour Organization’s (ILO) Private Employment Agencies Convention (C181, 1997) and the EU Temporary Agency Work Directive (2008/104/EC) promote flexicurity—combining flexibility for employers with adequate protection for agency workers [26,27]. This study adopts the UN/ILO framing of ‘decent work’ (SDG 8) in which sustainable development is understood as integrating economic, social and environmental dimensions [9,28]. Within this tri-dimensional understanding, the analysis operationalises the social dimension of SDG 8 into four assessable pillars: 1. employment creation (by guaranteeing productive job opportunities that offer fair wages), 2. social protection (which ensures safe work environments, social security and inclusion), 3. rights at work (which includes equal opportunities and treatment and also promotes personal development), and 4. social dialogue (which empowers workers to voice concerns, organize, and engage in decisions impacting their lives). In the analysis, these pillars are used to evaluate the extent to which TAW practices align with social sustainability, understood as maintaining and enhancing well-being through inclusive, fair, and participatory labor arrangements [29]. At the same time, recognising that SDG 8 is embedded in a broader economic–social–environmental agenda, the study also explores how the demand-side drivers of TAW and the resulting employment practices may have environmental co-implications, for example, through sectoral concentration in more emission- or resource-intensive branches, commuting and mobility patterns, and the organisation of work in existing facilities and infrastructures.
Recent scholarship has deepened the corporate social responsibility and the people dimension of sustainability within the TAW, identifying persistent gaps in rights awareness and unilateral contracting practices [30], deepening economic inequalities [31], constrained career development [25], and mixed effects on well-being [20,21,32,33]. Nonetheless, reputational concerns impose a degree of moderation on socially desirable outcomes, as client organizations and agencies understand the importance of a good employer image [30,34].
Policy responses vary: some EU member states favour stricter regulation via collective bargaining and/or state support [5,6,7], while Eurofound recommends regular reviews to lift unjustified restrictions [10]. Another approach, supported by research findings [16], recommends prioritizing the development and implementation of social innovation policies and initiatives to mitigate the negative aspects of TAW, particularly in the areas of skills enhancement and training, improving working conditions (both generally and for specific target groups), and addressing the gaps in effective social protection experienced by agency workers. Additionally, recent evidence from China [13] shows that digitally native companies effectively use collaborative strategies—such as sharing, openness, crowdsourcing, and value co-creation—to achieve positive outcomes in flexible employment. These flexible staffing models enable firms to swiftly form specialized teams with green skills, optimize resource usage, and reduce energy consumption through efficient workforce scheduling. Consequently, companies can rapidly adopt sustainable practices, enhancing their capacity for green innovation and alignment with environmental standards and market demands without the constraints of permanent hiring.
Despite extensive research on TAW, integrated evidence remains limited on (a) the organizational drivers behind selecting TAW over other non-standard arrangements and (b) the pathways through which these drivers materialize in stakeholder practices—for example, speed-to-hire, administrative/risk externalization, and short trial use—that support or undermine SDG-8 objectives for decent work. A comparable gap concerns how these same demand-side drivers and practices may carry environmental co-implications—for instance through the concentration of TAW in more emission- or resource-intensive branches, or through commuting and mobility patterns associated with triangular placements. Addressing these knowledge gaps, the study identifies and explains the demand-side drivers of TAW adoption and examines how they align with the SDG-8 pillars used here as analytical mechanisms, in order to generate policy-relevant recommendations that advance progress toward social sustainability and, in an exploratory way, indicate likely environmental co-benefits and risks.
The theoretical framework is anchored in the EU perspective on TAW and specifically examines the post EU Temporary Agency Work Directive period while drawing on the UN/ILO understanding of SDG 8 as integrating economic, social and environmental dimensions of sustainable development. This paper is methodically organized to address three pivotal research questions (R):
R1:
What specific labor market imbalances have client organizations sought to address through the adoption of TAW arrangements?
R2:
To what extent, and in what ways, have flexibility in work arrangements within TAW been balanced with considerations of job security and social protection of agency workers?
R3:
To what extent do implemented TAW arrangements promote decent work—through job creation, adequate social protection, respect for rights and equal treatment, and stronger worker voice and representation—and thereby advance social sustainability, and what indicative environmental co-benefits and risks are associated with these arrangements in terms of their sectoral placement and commuting/mobility patterns?
Consistent with this formulation, the analysis engages primarily with the social/decent-work dimension of SDG 8 while treating environmental aspects as supplementary and exploratory, tracing co-implications of the identified drivers through sectoral and mobility patterns rather than through dedicated environmental performance metrics.
Following mixed-methods typologies [35], the study adopts a qualitative-dominant, explanatory sequential mixed-methods case study design. Slovenia is examined as a case study to identify the key factors influencing the adoption of TAW by various stakeholders. The case study methodology was selected due to its capacity to enable a comprehensive investigation of the TAW phenomenon within its real-world context [36]. Slovenia’s significance lies in the sharp increase in demand for TAW among Slovenian client organizations, resulting in one of the highest proportions of agency workers within the EU in 2022. This trend persists despite the country’s relatively stringent labor protection laws and an industrial relations framework that supports social partnerships through multiemployer, multilevel bargaining.
An initial quantitative (descriptive) phase (Phase 1) was followed by an interpretive qualitative phase (Phase 2), with planned integration at multiple points. In Phase 1, official statistics and administrative/documentary sources were compiled and analyzed to map the prevalence and structure of TAW in Slovenia. Sources included legislation and regulatory documents; annual reports of the Inspectorate of Labour of the Republic of Slovenia; analyses by the Slovenian Employment Office; statistical data from the Statistical Office of the Republic of Slovenia (SURS), Eurostat, and the OECD. This phase also drew on prior research [37] documenting the evolution of TAW regulation since 1996 and associated enforcement challenges. The subsequent qualitative phase explained the mechanisms underlying Phase-1 patterns through in-depth semi-structured interviews (see Section 2). Integration occurred at the design stage (Phase-1 results informed the sampling frame and interview guide), the methods stage (quantitative/documentary patterns were translated into explanatory probes in interviews), and the interpretation stage (quantitative indicators, such as prevalence and sectoral concentration, were integrated with qualitative themes, for instance, limited social dialogue and weak worker voice, in the Results). Particular emphasis was placed on triangulation of methods and sources to enhance validity [38], by corroborating interview-derived themes with administrative reports and official statistics and by comparing perspectives across stakeholder groups. This design supports analytic generalization within the Slovenian case and enables theoretically informed transferability.
After presenting a comprehensive research review of global TAW facilitators in the next sub-chapter, the 1.2 sub-chapter highlights the unique labour market context of Slovenia, which is crucial for interpreting the results and their potential policy implications.

1.1. Previous Research on Motivations for Implementing TAW in Organizational Contexts

A modest corpus of early economic empirical studies have explored organizational decisions regarding the utilization of TAW arrangements. Initial research in this domain primarily sought to elucidate the client organization’s demand for the TAW, emphasizing the potential cost benefits, especially wage and benefit savings [39] and low training costs [40]. However, these studies also identified secondary factors driving TAW adoption, such as gaining access to special skills and adapting more effectively to market changes [41], managing workload fluctuations and staff absences [42], and maintaining some control over agency workers’ routines [40].
The enactment of the EU TAW Directive in 2008, which required equal pay and treatment, significantly reduced these cost benefits in most EU countries. As a result, subsequent research has shifted its focus, primarily exploring environmental and structural factors within organizations [43] in addition to the wider institutional context [44,45,46], which serves as driving forces for the adoption of the TAW. In addition, a modest volume of country-specific research [47,48,49,50,51] has emerged, concentrating on the implementation of the TAW within specific environments.
In their pivotal study among German businesses, Thommes and Weiland [43] methodically categorized characteristics pertinent to organizations’ engagement or abstention from the TAW into three distinct categories: (a) legal factors; (b) internal structural factors within client organizations, including the methodology of delivering goods and services, directional and communication systems, task complexity, and specificity of human capital requirements; and (c) functional factors, such as anticipated lower quality and flexibility, diminished staff motivation, and elevated transaction costs.
Based on the analysis of the data of the German Firm Survey Panel, which comprises 651 valid standardized computer-aided telephone interviews, Thommes and Weiland [43] concluded that both the likelihood and degree of utilizing the TAW are contingent upon various functional and structural characteristics. Moreover, nonmanipulable factors, particularly legislation and restrictions within collective agreements, exert a profound influence on the probability and scope of employing the TAW. The results show that organizations in the production industry are more likely to employ the TAW than are those in the service sector. In particular, it is the branch in which the particular organization is operating that has the strongest impact on the probability and extent of TAW use in Germany. Moreover, the existence of an institutionalized HR department at a client organization has a strong influence on the probability of using the TAW and has some influence on the extent of demand for the TAW. The usage of the TAW may appear to threaten a client’s internal HR department, as it could be seen as outsourcing HR tasks. This is because the agency assumes responsibilities such as recruitment, training, and compensation for the workers they provide. However, the findings of Thommes and Weiland [43] suggest that the benefits of using the TAW outweigh the risks. These advantages include setting benchmarks for the activities and compensation of permanent staff and the capacity to prevent excessive manpower demands during peak periods [45].
In an analysis encompassing large-scale data from the World Economic Forum and the European Company Survey—spanning 28 EU countries, Northern Macedonia, and Turkey—scholars [46] explored the relationships between business systems, client characteristics, and their interactive effects in relation to TAW utilization. The results indicated that clients’ use of the TAW is not, in general, influenced by the business system within which the workplace operates. It is rather broader institutional context and clients’ organizational specificities within which different policies and practices are implemented that interact to influence the (non)use of the TAW. The institutional context is sculpted by numerous factors [47], including (a) the labour market system, with particular attention given to wage bargaining centralization, forms of employee representation, and employment protection legislation; (b) the state, with its variable predictability and transparency in governance; (c) norms and values, directing trust in formal rules and regulations, as well as confidence in officials to enforce legislation aptly; and (d) the financial system. As a result, Allen and colleagues [46] observed that business systems with flexible labour markets and those with stricter, more rigid labour markets (having employee representatives and a company-wide wage deal) are often linked to more use of the TAW.
In 2009, following the adoption of the TWA Directive, Eurofound conducted a comprehensive study in EU member states on the forms of regulation and national variations, as well as the characteristics of the established TAW. Despite the relative novelty of the TAW at that time, the study highlighted several key growth drivers [10] (pp. 6–7) in the demand sector. First, the TAW emerged as a strategy for controlling wage costs and is particularly useful in sectors characterized by lower-skilled jobs, competitive markets, and fluctuating or unpredictable demands. Second, the TAW serves as a flexible solution for managing labour adjustments during periods of economic growth, fluctuations, and increasing competitive pressure. Third, the research noted that the TAW offers cost savings in transactions by delegating certain recruitment and administrative responsibilities to agencies. Last, it was observed that the TAW creates a pool of workers, offering a selection of candidates for potential permanent positions within client organizations.
In 2013, two EU social partners in the TAW sector, namely, the World Employment Confederation-Europe (WEC-Europe, at a time Eurociett), which is a representative organization of the private employment services industry at the European level, and the UNI Europa, which is the European trade union federation, mandated a study [8] that identified four major categories of reasons why and for what purposes client organizations implement the TAW. The overview study included data and multiple case studies across the EU. Among TAW drivers, the provision of numerical flexibility emerged as a main category. Conceptualized over forty years ago [52], numerical flexibility as a workforce management strategy involves practices that allow organizations to rapidly and easily adjust their headcount, ensuring that the number of employed workers precisely matches the organizational needs at any given time. According to the study results [8] (p. 64) client organizations create a flexible workforce by maintaining a core of skilled staff and using agency workers for rapid adjustments, thus reducing turnover and avoiding redundancy costs for permanent employees. This emerging trend has been similarly observed in specific national contexts, as indicated by research conducted in Germany [49,51], the USA [42,53,54], the Czech Republic and Poland [48], and the UK [47,50]. Eased recruitment is a second reason for TAW utilization since client organizations outsource recruitment and performance assessment to agencies, enabling them to evaluate agency workers’ suitability with minimal commitment. Client organizations also externalize the risks associated with directly employing workers, which is a third reason for TAW use. By doing so, these organizations avoid the responsibilities of standard employment relationships, such as selection, payroll administration, and performance management. Overall, such actions can result in either reduced or more predictable (and manageable) labour costs for client organizations. Agency workers in the EU often earn less than permanent staff but have similar overall costs to client organizations due to agency fees. In many EU countries client organizations are not liable for benefits such as pensions or maternity leave, holiday and sickness pay and redundancy-related costs.
As the TAW sector has become more established, agencies have strengthened their ability to market services to potential client organizations and workers. Fifteen years after the enactment of the TAW Directive and the Eurofound study [10], agencies now offer a wider array of services and cultivate longer-term relationships. Despite listed study findings, a significant research gap remains concerning the mechanisms that balance the demand for flexibility in work arrangements with the need for job security and social protection for workers.

1.2. The TAW in the Slovenian Labour Market

Slovenia was selected as a case study to investigate this gap due to its unique TAW market characteristics. The Slovenian TAW is marked by the existence of two distinct categories of agency workers—regular and student—and the concentration of agency workers within the four largest agencies. Furthermore, Slovenia has seen a significant increase in the proportion of TAW use in its workforce. This growth is particularly notable given the country’s relatively robust labor protection standards regulating TAW relationships. These standards are partially represented by the OECD Employment protection indicator for temporary employment, which assesses the rigor of regulations governing the use of TAW contracts and fixed-term contracts. In 2019, Slovenia’s score was 1.83, slightly higher than the OECD average of 1.79 [55].
Twenty years after the introduction of the TAW in Slovenia, the scope of agency workers was high compared to the average in EU member states. According to Eurostat [56] data, in 2021, 4.8 percent of workers aged 15 to 64 were employed under the TAW system in Slovenia, while the EU27 average for 2021 was 2.5 percent. Slovenian Labour Force Survey data for 2022 [57] indicated that the share of workers in a standard TAW (among all employed persons in Slovenia) was 2.2 percent, while the remainder consisted of student workers, which accounted for 2.6 percent of employed persons. In the context of the categories of agency workers, there are two distinct classifications of tripartite work relationships identified in Eurofound [10] (pp. 7–9) statistics regarding TAW use in Slovenia: (1) student work, which encompasses the employment of students engaged in university education and pupils in the secondary education system, and (2) the employment of agency workers, who engage in atypical employment through temporary work agencies and are subsequently placed within client organizations (i.e., TAW use). The student work system in Slovenia is a unique form of casual employment that is exclusively available to pupils and students enrolled in secondary or tertiary educational institutions. Despite its triangular employment structure, student work does not exhibit other typical characteristics associated with the standard TAW and is not considered to be TAW use by the Slovenian legislature.
Slovenia, similar to other EU countries, experiences significant challenges stemming from the mismatch between labor supply and demand, particularly in sectors such as trade, transport, tourism, manufacturing, and construction. According to Labour Force Survey Data [57] the utilization of TAW in Slovenia is notably high in these industries. This imbalance often increases reliance on agency workers to address short-term labor shortages, which can intensify competition for talent across neighboring labor markets. The workforce in these sectors is primarily composed of individuals with secondary vocational education or equivalent qualifications, although the proportion of workers with tertiary education is also increasing significantly. Lower-skilled agency workers predominantly perform on-site tasks at client premises, whereas higher-skilled workers frequently engage in remote or hybrid working arrangements supported by ICT within client organizations. The gender distribution in TAW shows a unique trend in Slovenia, where women are more likely to be employed as agency workers than men. This contrasts with the broader EU pattern, where in 17 out of 25 countries, men dominate TAW employment, according to Eurostat data [56]. The gender disparity in the Slovenian labor market is further reflected in previous research comparing perceptions of job security and workplace well-being. A survey of 1009 employees in both the public and private sectors revealed that men generally rated job security, access to resources for a decent life, safe working conditions, and workplace well-being higher than women [58].
In terms of nationality, Slovenian citizens constitute the majority of agency workers, followed by a substantial number from Bosnia and Herzegovina. Additionally, agency workers in Slovenia are often employed in some of the lowest-paying jobs, a trend particularly evident in the TAW sector, as highlighted by SiStat data [57]. Previous studies also suggest that compliance with service quality standards is closely linked to social security, income security, and employment security. Organizations that provide workers with strong opportunities in areas such as economic stability, workplace safety, and work–life balance are likely to see improvements in the quality of their products and services [59]. This underscores the importance of ensuring agency workers have access to secure employment conditions, which benefits both the workforce and client companies.
The supply side of the TAW market in Slovenia is divided into two significant segments. Of the 142 agencies registered with the Ministry of Labour, Family, Social Affairs, and Equal Opportunities (of which 134 are domestic and 8 are foreign business entities, as of July 2024) [60,61], a handful of agencies, all members of the Employment Agencies Association, mediate the majority of agency workers in Slovenia [62]. Additionally, many unregistered business entities mediate work illegally, as has been repeatedly noted by the Labour Inspectorate of the Republic of Slovenia [63]. The number of agencies rapidly grew between 2008 and 2014; however, since then, it has remained relatively stable [57]. Larger agencies operate across multiple Slovenian regions and seek to align local labour supply with demand, thereby lowering clients’ operating costs—through rapid, efficient placements and reduced commuting expenses—and enhancing operational productivity by engaging workers willing to adopt flexible schedules.
Since the implementation of the TAW, the Slovenian legislature has frequently made regulatory changes to ensure an appropriate balance between flexibility and security and to mitigate negative effects caused by larger gaps in supply and demand in the labour market, as well as issues related to structural unemployment. Two years after adopting the EU TAW Directive, Slovenia enacted the Labour Market Regulation Act (Official Gazette of the RS, No. 80/10; hereinafter, the LMRA). In Chapter IX of the LMRA, the directive was harmonized, and one of the more restrictive forms of TAW in the EU was established. The LMRA introduced a registry and a database at the Ministry of Labour through which agencies had to register (Article 163 of LMRA) and set forth several conditions for providing workers to clients (Article 164 of LMRA); i.e., an agency could not violate labour laws or have unsettled taxes and contributions, and it had to meet other staff, organizational, and spatial conditions. One significant limitation pertained to the involvement of work for foreigners from third-party countries. According to Article 167 of the LMRA, agencies could mediate work for citizens of Slovenia, EU member states, the European Economic Area (EEA), or the Swiss Confederation. For all other (foreign) workers, agencies could mediate work only if these individuals had previously acquired a permanent residence permit in Slovenia, had a personal work permit valid for three years, had acquired temporary protection status, or were asylum seekers. These stipulations significantly narrowed the circle of workers that agencies in Slovenia could mediate, as they were not allowed to obtain work permits for foreigners from third-party countries.
The revised Employment Relationship Act of 2013 (Official Gazette of the RS, No. 21/13; hereinafter, ERA-1) and subsequent enacted and amended legislation (LMRA and the Employment, Self-employment and Work of Foreigners Act—ESWFA, Official Gazette of the RS, No. 47/15) have intensified the trend of tightening conditions for TAW use. These legal changes have restricted the proportion of the TAW within a single-client organization (to no more than 25 percent of the workforce for agency workers with fixed-term contracts), formalized the agency registry under the parent ministry, imposed stricter conditions for agency establishment and operation, enhanced inspection oversight, and introduced stringent audit controls over agency activities. Among the latest amendments of ERA-1D, as of November 2023, a revised Article 61, paragraph 3, mandates that agency workers must continue to be paid between assignments at a rate not less than 80 percent (previously 70 percent) of their average monthly full-time salary over the last three months (previously 70 percent of the minimum wage). However, a proposed amendment to further reduce the allowed percentage of time-limited contract workers within a client organization from 25 percent to 15 percent was not passed.
In Slovenia, the employment contract between an agency and an agency worker can be concluded for either an open-ended period of time (as a rule, permanent employment) or for a definite, fixed term (fixed-term employment), as an exception, with a few conditions (Article 60, paragraph 1 and 2, ERA-1D). Since 2013, the proportion of permanent contracts among agency workers has been steadily increasing, now ranging around 70 percent. Additionally, the average duration of assignments at client organizations is up to five months but has been lengthening [62]. There are no special rules for TAW employment concerning the social security and social benefits of an agency worker, such as health insurance, invalidity, unemployment benefits and pensions, which are governed by the normal rules that apply to all organizations. Agency workers are free to join the relevant union for the sector, occupation or workplace in which they are placed. The estimate of trade union density for agency workers in Slovenia, as reported by The Association of Free Trade Unions of Slovenia, which is one of the major union confederations, is 0.18 percent [10]. Although labor legislation, particularly ERA-1, permits additional negotiations at the sectoral level, the rights of agency workers are rarely included in collective agreements, and this option has largely remained underutilized to date.
The effectiveness of Slovenian legislation in safeguarding the rights of agency workers has been examined by a limited number of domestic scholars, mainly in the field of legal studies, through legislative, comparative, and judicial decision reviews. Authors have debated the implementation of equality principles for agency workers relative to those directly employed by client organizations [64,65] and revealed concerns for agency workers, notably in areas such as work–life balance, health and well-being, and career progression, which are benefits typically reserved for directly employed staff. A consensus is emerging among academics [66,67,68,69,70] that current legislation and its application may still permit elements of precariousness in TAW relationships, although concrete evidence of this remains limited. An analysis of court cases [71] indicates a rising number of legal disputes concerning the liability of agencies and client organizations for harm experienced by agency workers at work.
Empirical research in this area is scarce, with a notable exception being a study commissioned by the Slovenian Employment Service [62]. The driving forces behind the proliferation of the TAW in Slovenia and the reasons for its relatively high adoption rate among clients are topics that have not been thoroughly investigated. There remains a significant gap in the research regarding the systemic mechanisms necessary to balance flexibility with job security and social protection for agency workers. Although several legislative amendments have been enacted, empirical assessments of their implementation remain limited, particularly with respect to everyday operational and contractual relations between agencies and client organizations and to the ways flexible arrangements shape job security, income stability, career progression, and worker well-being. These same operational choices are also likely to have environmental co-implications: for example, placement patterns and rotations influence commuting and business travel, decisions about on-site versus remote or hybrid work and shift design can affect building use and associated energy demand, and digitalized workflows may substitute for physical travel while creating distinct energy and equipment footprints. In this study, the primary focus remains on social sustainability and decent work within SDG 8, while environmental aspects are examined in a supplementary and indicative way—traced as potential co-benefits and risks of the identified demand-side drivers and practices, using sectoral profiles, commuting and mobility patterns, and respondents’ accounts rather than dedicated environmental performance metrics. Using this integrated but explicitly social-centred lens, the study examines Slovenian TAW to identify unrealized opportunities for improvement and to clarify the practical effects of recent legal changes. A comparable evidence gap exists internationally, where systematic evaluations of TAW’s contribution to the SDGs—most notably decent work and the social dimension of SDG 8, alongside climate- and resource-related targets—remain scarce.

2. Materials and Methods

The study aimed to identify and explain the demand-side drivers of TAW adoption among user organizations (private and public) and to examine how these drivers materialize in stakeholder practices, assessing their alignment with the SDG-8 pillars—used here as analytical mechanisms (employment creation, rights at work, social protection, and social dialogue)—in order to inform policy recommendations that advance progress toward social sustainability.

2.1. Sampling

The study employed a qualitative-dominant, explanatory sequential mixed-methods case design with integration at the design stage. Phase-1 documentary and statistical mapping informed Phase-2 sampling and interview prompts—focusing on sectors with high TAW prevalence and probing emerging mechanisms such as speed-to-hire and administrative/risk externalization. Based on a purposeful (criterion) sampling approach (with predetermined criteria of importance; see Table 1) [72], five respondent groups were purposively selected to capture the full triangular employment relationship and adjacent governance and representation actors:
  • TAW agencies (agencies): to inform service models, market positioning, and the operational mechanisms through which client demand is met.
  • Client organizations (users): to specify organizational drivers for adopting TAW (e.g., volatility, skills shortages, headcount controls) and to trace internal workforce strategies.
  • Agency workers: to surface lived experiences with job quality, equal treatment, mobility/skills, and representation.
  • Trade-union representatives: included due to Phase-1 evidence of low union inclusion among agency workers and indications of untapped potential of social dialogue at sectoral/organizational level.
  • State administrative body representatives (work/residence permits for foreign workers): included because a notable segment of agency workers in Slovenia are foreign nationals (predominantly from Bosnia and Herzegovina) in lower-skilled roles; these officials have direct insight into procedural dynamics around permit extensions, denials, and employer transfers, and how regulation operates in practice.
Within agency respondents’ groups, additional sampling criteria included the size of the agency (the sample included representatives of the five largest agencies and other agencies) and information about membership in the Employment Agency Association (the sample included members and non-members). For the agency stratum, the Phase-1 mapping drew on the official registry of all licensed agencies and on documentation from the Employment Agencies Association (18 member agencies) and the six largest agencies in the Slovenian market. Building on this frame, Phase-2 interviews focused on seven agencies: four leading agencies that jointly employ the majority of agency workers in Slovenia and three micro-, small- and medium-sized agencies, including both association members and a non-member (see Section 3). Within the client organization group and trade union representatives’ group, additional sampling criteria included an industry branch, where the sample was limited to five industry branches with the highest prevalence of TAW use.
Triangulation of the data sources was used to ensure the convergence of information from different origins and to enhance validity [38].
The snowball sampling method [73], in which selected research participants suggested other stakeholders they knew, was used for sampling client organizations (within each of the five largest industry branches), trade union representatives and for sampling agency workers. The selected participants represented information-rich cases. Recruitment continued until the sampling criteria were met. The goal was explanatory depth and triangulation, so sufficiency was judged by information richness, coverage of key stakeholder positions and convergence across sources. Documentary and statistical mapping in Phase 1 narrowed the scope, so fewer—but more information-rich—interviews were needed in Phase 2. Recruitment and analysis proceeded iteratively: after each interview, the codebook was updated and assessed for whether new material extended the conceptual boundaries of the emerging driver categories. Data collection ceased when additional interviews no longer produced new categories or substantive refinements to the seven driver mechanisms, indicating thematic sufficiency for the explanatory aims of the study.
The sampling strategy was therefore designed for analytic rather than statistical generalisation, consistent with qualitative case-study standards. In line with the ‘information power’ principle in qualitative research, sample adequacy was judged by the relevance and richness of the data in relation to the study aim, not by numerical thresholds alone [74,75]. Phase-1 documentary and statistical mapping provided near-population coverage of the Slovenian TAW sector, which allowed Phase-2 interviews to focus on a smaller set of information-rich cases with direct responsibility for TAW design, implementation and regulation. Within each stakeholder group, the aim was to achieve maximum variation on theoretically salient dimensions (agency size and association membership; high-prevalence vs. other branches among client organisations; different experiences and positions among workers and trade union representatives), so that the final configuration of 19 interviews across five groups would capture the key demand-side mechanisms and governance configurations relevant to the research questions.

2.2. Data Collection

Prospective participants were contacted through email (in December 2020 and again in November 2023) and provided with an information sheet that included comprehensive details regarding the researcher, study objectives, interview process, and confidentiality protocols. Individual interviews and focus groups were conducted from 8 December 2020 to 13 April 2024 using Zoom 6.0, and two participants opted to provide their responses live. Prior to each interview, the study objectives and procedures were clearly articulated, and any participant queries or concerns were satisfactorily addressed. All participants provided informed consent before the interviews. A semistructured interview guide was used to facilitate each discussion (see Supplementary File S1); this guide was informed by the comprehensive review of the literature and further refined following an informal pilot focus group. All interviews were digitally recorded, and notes were taken concurrently. Subsequently, all the audio recordings were transcribed and imported into Lumivero’s NVivo 14 software for further analysis.
The semistructured interview guide (S1) opened with questions on each participant’s role history in TAW, current responsibilities, and organizational setting. For agencies, it explored the service portfolio, market evolution, clients’ reasons for using TAW, client-specific tailoring, benefits communicated, illustrative cases, market trends, and future outlook. For client organizations, it examined primary reasons for adopting TAW, perceived benefits and challenges, fit with organizational strategy, effective use cases, and desired improvements. For agency workers, it addressed reasons for choosing TAW, experiences within the triangular employment relationship, and perceived effects on job security, career progression, and future plans. For the state administrative authority, it focused on the authority’s role in TAW administration, patterns linked to foreign workers, and observed system dynamics. For trade unions, it investigated sectoral conditions, the effects of TAW on labour relations, member challenges, the anticipated future role of TAW, and proposals to improve worker outcomes.

2.3. Data Analysis

The data obtained from the interviews were subjected to thematic analysis, a method for identifying patterns and themes in qualitative data. The analysis was conducted in two phases, combining deductive and inductive approaches. In the first phase, a preliminary coding framework was developed deductively from existing research, relevant theories and the research questions to define the scope of the study. This primary coding frame centred on (a) demand-side drivers of TAW (D1–D7) and (b) the four SDG-8 decent-work pillars (employment creation, social protection, rights at work, social dialogue). All transcripts were coded against this framework in NVivo, with segments assigned to one or more driver and pillar categories depending on their content.
In the second phase, an inductive coding approach [72] was used to refine and extend the primary frame. Within each driver and pillar category, additional codes and sub-codes were generated from the language and meanings used by respondents, allowing unanticipated patterns and mechanisms to emerge. The codebook (see Supplementary File S2) was continuously updated and revised as new interviews were coded and earlier transcripts were revisited, and this iterative process continued until no substantively new information was obtained from repeated coding.
To enhance analytic transparency and trustworthiness, an auditable NVivo workflow (coding memos and decision logs) was maintained, triangulation was applied across stakeholder accounts and administrative/statistical sources, and joint displays linking quantitative/documentary indicators to qualitative themes were reported in the Results (e.g., drivers → mechanisms → outcomes). Seven demand-side driver categories were identified (D1–D7; see Results), with definitions and exemplar quotations provided in Supplementary File S2. In addition, a coded-intensity matrix was constructed to summarise the salience of each driver across stakeholder groups; intensities were derived from conservative code frequencies, triangulated with documentary evidence, and the resulting matrix is presented as Table 2 (see Section 3).
In addition to this primary coding frame based on demand-side drivers (D1–D7) and the four SDG-8 pillars, a secondary coding pass identified segments in which respondents referred (explicitly or implicitly) to environmental aspects of TAW arrangements (e.g., commuting and mobility patterns, remote and hybrid work). These environment-related codes were then linked back to the driver and pillar categories to explore indicative environmental co-implications (co-benefits and risks) of TAW deployment within the Slovenian case.

2.4. Ethical Considerations

The study received Research Ethics Committee approval. All participants received an information sheet and provided informed consent prior to participation. Procedures conformed to internationally recognised standards for human-participants research (e.g., the Declaration of Helsinki). Personally identifiable information was not collected beyond what was necessary for recruitment and scheduling; interview data were pseudonymised at transcription, with direct and indirect identifiers removed before analysis and reporting. Audio files were stored on an encrypted drive with access restricted to the researcher; transcripts were managed in NVivo under the same access controls. No minors or legally vulnerable populations were targeted; participation was voluntary, with the right to withdraw at any time, and no adverse events were reported. Data processing complied with general data protection principles (lawfulness, purpose limitation, data minimisation, storage limitation, integrity and confidentiality); retention and secure destruction follow the institutional policy approved by the ethics committee. Because the transcript content contains context that could enable re-identification, individual-level data are not publicly shared; instead, anonymised excerpts supporting the analysis are provided in the codebook (Supplementary File S2), consistent with ethics guidance on safeguarding participant confidentiality.

3. Results

Study participants (19 respondents) encompassed a diverse group of stakeholders involved in TAW use in Slovenia: seven agency representatives, five client organizations representatives, four agency workers, two trade union representatives and a representative from an administrative body. Demographically, 12 of the 19 respondents were men, and the majority of the sample (n = 16) had accumulated three or more years of experience in the field of TAW in Slovenia.
After a closer examination of the agency representatives’ profiles, it was found that four participants represented the leading agencies in the Slovenian labor market, which employ the majority of agency workers in Slovenia [62]. The remaining three agency representatives were from micro, small, and medium-sized agencies. Among the participating agencies, only one was not affiliated with the Employment Agencies Association. The agency sample achieved maximum variation, revealing differences in governance commitments and the extent of HRM outsourcing.
Representatives from client organizations were primarily from the manufacturing industry (n = 3), followed by the tourism (n = 1) and construction (n = 1) industries. The manufacturing client organizations included three medium-sized companies, producing household appliances, automotive parts, and industrial instruments, all with a high prevalence of TAW arrangements. Two of these companies were long-established, domestically owned entities. The third company was the Slovenian branch of a foreign corporation that promotes TAW arrangements across its global network. The client organization from the tourism sector was a small branch of an international company, specializing in insurance products for the tourism industry. The client from the construction sector was a medium-sized Slovenian branch of a global company. Both clients from the tourism and construction industries had a long-standing tradition of employing agency workers.
The agency workers involved in the study came from industry sectors with the highest prevalence of TAW arrangements in Slovenia; two worked in logistics, while tourism and manufacturing each had one representative. The two trade union representatives (company- and industry-level) had over 10 years of experience with TAW arrangements and were from the manufacturing and logistics industries. Workers and trade unions offered interlinked perspectives: workers validated mechanisms such as trial use and pathways to direct hire, while two unions (company- and industry-level) provided oversight on social dialogue and equal treatment.
A single state-level informant was sufficient because regional units follow common procedures, providing system-wide visibility on foreign-worker permits and scarcity dynamics.

3.1. Drivers of Temporary Agency Work Engagement in Slovenia (R1)

Study respondents identified seven drivers behind client organizations’ demand for TAW use in Slovenia (See Figure 1). Of these drivers, three are primary and external drivers linked mainly to labour market (or one of its segments) imbalances: (1) medium-term fluctuations necessitating some flexibility (the predominant driver, D1), (2) the seasonal nature of jobs (D2), and (3) labour scarcity (D3). The remaining four drivers are more nuanced and relate to specific human resource strategies tied to the TAW. These are internal factors related to client organizations: (4) acting as an alternative to probationary or temporary employment (at client’s organization, D4); (5) ensuring a workforce that is adaptable, present, and motivated (D5); (6) providing career guidance and transitions to better-suited job conditions, guaranteeing a harmonious match between client and worker (D6); and (7) the partial outsourcing of the HRM function (which also includes managing worker quotas, D7).
The coded-intensity matrix (see Table 3) shows medium-term fluctuations (D1) and labour scarcity (D3) as universally high (coded intensity 3) across all stakeholder groups, confirming volatility buffering and hiring pressure as the shared anchors of TAW use. Seasonality (D2) is moderate (coded intensity 2) for agencies, unions, and the state but situational/low for clients and workers (coded intensity 1), indicating uneven sectoral exposure. Substitution for probation (D4) is moderate for clients and workers, low for agencies and unions, and absent for the state, suggesting a trial-use pathway primarily perceived within firms and by workers. The driven workforce driver (D5) is high for clients, moderate for agencies and workers, low for unions, and absent for the state, evidencing a client-centric emphasis with partial confirmation by supply-side actors. Match quality/transitions (D6) register moderate salience for agencies, clients, and workers but not for unions or the state, underscoring a placement-to-hire pathway experienced within the triangular relation rather than in oversight arenas. Partial HRM outsourcing (D7) is moderate for agencies and clients, low for workers and unions, and absent for the state, consistent with administrative centrality on the supply/user side and limited visibility elsewhere.
Taken together, these patterns imply policy levers that are shared (volatility and shortage management) yet stakeholder-specific: clients prioritize effort/motivation and trial-use/transition pathways; agencies emphasize seasonality management, matching, and HRM process support; workers foreground transitions and fair treatment in trials; unions focus on cyclical/structural pressures; and the state concentrates on macro volatility/shortage oversight rather than organization-level mechanisms.

3.1.1. Medium-Term Fluctuations Necessitating Partial Labour Flexibility

The study participants from all stakeholder groups highlighted the significant role of agencies in addressing the fluctuating labour needs of client organizations. Agency and client respondents anonymously agreed that eased management of medium-term fluctuations, which enables the partial flexibility of the workforce, especially by making possible rapid employment and easing layoffs while keeping them legally compliant, is a major driver for choosing TAW.
‘Larger companies certainly integrate us [agencies] into their processes, aiming to have a certain workforce flexibility, allowing them to adapt through multiple mechanisms to the changing needs for labour that may arise’.
(A2)
According to respondents from agencies (A1, A2, A4), such flexibility is particularly crucial given the relatively low level of flexibility of the workforce in Slovenia due to high labour protection. While both agency and client representatives acknowledged the strict labor regulations, data indicates that the protection of agency workers in Slovenia is only slightly above the OECD average [55].
Managing medium-term fluctuations becomes easier through what one respondent (A1) termed the ‘horizontal method of agency employment’. In this approach, agency workers are assigned to a single client for a specified duration. As emphasized by client-respondents (C1, C2, C4, C5) and confirmed by the agency participant (A2), agencies can offer time-efficient staffing solutions. These are particularly effective in addressing immediate staffing shortages, which are critical during periods of economic growth when certain roles are in high demand.
‘As an agency, we always encourage companies to think ambitiously. The fact that they can call us on a Friday and have 15 new workers by Monday has definitely allowed companies to think bigger, knowing they can get the workforce they need exactly when they need it’.
(A2)
However, the ‘horizontal method’ is also employed to fill permanent positions at client organizations, even when there is no short- or medium-term crisis, as highlighted by a trade union representative (T1) and confirmed by clients (C1, C3), who explained that this approach is a general HRM strategy used by their international headquarters.
Agencies provide a vast pool of potential candidates seeking career challenges (C2). As noted by agency-respondent (A2), two-thirds of the workers placed by agencies were previously unemployed, thereby serving as a gateway into the labor market. This group includes individuals who experienced long-term unemployment, had disabilities, were migrant workers, or belonged to other disadvantaged groups in the labor market. Agency respondents (A1, A3) emphasized that clients are more likely to engage with these workers when they are employed through an agency.
For certain positions at client organizations, agency work is the only available pathway to employment, which can later transition into fixed-term or permanent positions at the client organization, as explained by participating workers (W1, W3) and confirmed by both agency (A1, A6) and client representatives (C3, C5).
Agencies not only save clients’ time by overseeing the hiring process—which includes reviewing, screening, and responding to applications (A3)—but also, as emphasized by agency respondents (A1, A4, A2), manage layoffs. Agencies aid companies in handling notice periods and severance packages, rendering the dismissal process more flexible (C1, C4, C5). Agencies also guarantee continued employment for those workers who have demonstrated their value. As stressed by trade union representatives (T1, T2), this can lead to segmentation and unequal treatment of agency workers.
‘Unlike traditional layoffs involving a larger group of directly employed workers, when there is a reduced need for agency workers, the client organization and the agency are not legally required to establish or adhere to specific criteria for selecting surplus agency workers. They can make decisions based on their mutual interests. Moreover, agency workers and trade unions have no influence or involvement in this process.’
(T1)
In the event of a reduced need for workforce, agency workers are the first to be let go, even when their performance surpasses that of directly employed workers. Trade union representatives (T1, T2) emphasized that agency workers are generally not union members at the client organization. However, union representatives at the client organization are available to provide assistance if needed. According to trade union representative (T1), agency workers are often reluctant to join unions due to concerns that it may be viewed unfavorably by the client organization. Additionally, TAW arrangements are rarely addressed in sectoral collective bargaining, and legislation governing worker participation in management excludes agency workers, as emphasized by trade union representative (T1).
TAW enables the swift replacement of agency workers, allowing them to return to the agency while maintaining their employment status. As noted by agency-respondent (A4), approximately 70 percent of agency workers in Slovenia hold permanent contracts with the agency. However, as reported by agency workers (W1, W3), this situation results in the loss of benefits provided by the client organization, necessitates adapting to a new work environment, and often requires taking on unfamiliar tasks. Moreover, if no new work opportunities arise, the agency may terminate the employment of workers, even if they hold a permanent contract, as highlighted by trade union (T1) and agency (A1) representatives. This issue became particularly pronounced during the COVID-19 pandemic, when many agency workers were laid off, as noted by trade union representative (T1). According to the agency respondent (A1), agencies typically allow two to four weeks to find a new assignment for a worker before proceeding with termination.

3.1.2. Addressing Seasonality in Labour Demands

The second external driver—seasonality—was acknowledged by all stakeholders, rated as moderate by agencies, trade unions, and the state, but situational by client organizations and agency workers. Participating agencies (A1, A4) and one worker (W2) highlighted the role of TAW in smoothing seasonal demand and sustaining continuous employment for agency workers despite industry-specific fluctuations across the year. One agency (A1) referred to this approach as a ‘vertical employment model’, describing it as a creative solution to seasonality.
‘Our vertical employment model serves [names one of the Slovenian airports] and [names one of the Slovenian ski resorts]. The ski resort requires more staff in winter and less in summer, while the airport’s needs are reversed. We send the same employees to the airport in summer and to the ski resort in winter, enabling year-round employment. Without this approach, both companies would employ only short-term contracts, leading to multiple temporary contracts for employees’.
(A1)
One respondent, namely, an agency worker (W2) with one of the clients who has a history of employing agency workers due to seasonal demands, stressed that the integration of seasonal workers is seamless.
‘Given that the company has a longstanding tradition of employing agency workers and considering the realistic increase in seasonal needs, as well as the fact that regular employees are happy about the arrival of new workers (and must accept them for the work process to flow), there is no noticeable difference between the workers. We never even discuss who is an agency worker, although I often talk with other employees’.
(W2)
This arrangement involves a scenario where agency workers transition between two sectoral collective agreements (specifically in the logistics and tourism sectors), which do not necessarily guarantee the same rights. For instance, under the logistics collective agreement, a worker may be entitled to a salary supplement that is not provided for under the tourism collective agreement. As explained by an agency representative (A1), it is within the agency’s and both clients’ (the airport and the ski resort) purview to negotiate such supplements. Legally, there is no obligation to pay the supplement during the period when the worker is employed at the ski resort. Although agencies strive to secure the best possible combination of rights from both contracts, the final decision rests with the clients. Moreover, the additional complexity of these arrangements—including worker relocation, cross-client job matching, and the reorganisation of schedules and rosters—renders such services comparatively more costly for clients, as confirmed by an agency representative (A1).

3.1.3. Addressing Labour Scarcity

The third major driver for TAW use identified in this study by all stakeholders is labour scarcity, especially in certain professional groups that are in high demand or require highly qualified personnel. The strategies of agencies in handling labour scarcity involve actively seeking out specialized workers, offering better compensation through agencies, and retaining them in the TAW even during client downtime, ensuring their availability for future assignments. This is coupled with the active collaboration between client companies and agencies in sourcing in-demand profiles and addressing the unique hiring preferences of certain specialized roles.
The agency respondent (A1) explained a trend towards utilizing specialized staff or roles that require skills that are in short supply.
‘The third model, which is becoming stronger in Slovenia, is known as the specialized staff model. It pertains to roles for people with skills that are in short supply. These jobs are usually better paid by agencies than by clients working directly. For example, at the moment, it is challenging for a retailer to recruit a butcher, as many butchers are employed by agencies, which offer them higher wages and better working conditions’.
(A1)
The risk associated with this practice, where certain in-demand professions become ‘locked into’ TAW, is that it allows agencies to exert control over market conditions. This shift enables agencies to transition from simply facilitating the needs of clients in a highly competitive TAW market to becoming indispensable business partners.
The recruitment of workers in scare professions is different than that of other agency workers since they are usually not actively looking for employment.
‘We approach them and initiate conversations, attempting to persuade them to consider a change in employment. The most desirable professions are butchers, salespeople, technical staff, production workers, and warehouse workers. The situation with these in-demand roles is such that we must hire the candidate immediately, as by the next day, they may already have three new offers’.
(A4)
Workers from professions in high demand represent a niche group for whom agencies are willing to cover the waiting period (payment between assignments) rather than release them.
‘If one client no longer needs such specialized staff, we are obligated to pay the employee 70% of their salary (or more but no less than 70%) [now 80%] during the waiting period at home. This applies to hard-to-find profiles where it makes more financial sense for us to pay for their “downtime” or “waiting at home” rather than letting such a person enter the labour market’.
(A2)
When seeking staff from professions in high demand, clients (C2, C4) typically engage multiple agencies. For highly sought-after workers, agencies (A1, A3, A4, A5, A6) must be exceptionally proactive. The agency respondent (A3) also highlighted the unique behaviour of some specialized workers (especially in the IT and legal sectors), who prefer to be independent contractors, which means that they opt for the flexibility of working for multiple clients simultaneously. In this context, agencies (A1, A3) are inclined to negotiate with clients to accommodate agency workers’ preferences for greater flexibility, particularly concerning remote or hybrid work arrangements facilitated by the increased adoption of ICT. Specifically in the IT sector, such arrangements foster enhanced collaboration between agency workers and directly employed personnel.

3.1.4. A Substitute for Probationary Work or a Fixed-Term Employment Contract

The fourth—and subsequent—drivers for TAW use reflect human resource management strategies rather than broader labour-market conditions. With the exception of the state representative, who focused on macro-level oversight rather than organization-level mechanisms, respondents across agencies, user organizations, trade unions, and worker groups consistently reported that TAW is employed as a substitute for probationary periods or fixed-term contracts, and in some cases as a means of extending fixed-term employment.
The TAW provides a robust pool of workers from which client companies can employ as per their needs (C1–C4). The system is advocated by agencies, which suggest that clients use the TAW for a fixed period of time instead of fixed-term contracts. During this period, the agency provides the worker with a permanent contract, which, as explained by agency representatives (A1, A2), is intended to offer greater job and social security to the worker. Clients (C1, C3) supported this view by explaining that chaining fixed-term contracts provides less security for workers. However, clients are under no obligation to employ the agency worker after the assignment ends, nor is there a legal requirement to offer the worker a permanent contract. Consequently, TAW may serve as an additional career step or potentially as an obstacle on the path to securing a permanent employment contract with the client organization.
As explained by client (C1, C2, C4) and agency (A1, A2, A5, A6) representatives, and corroborated by trade union representatives (T1, T2), approximately 10–20% of agency workers are directly employed by client companies at the end of their assignments, typically on an annual basis. Client respondents (C1–C5) agreed that efficiency is the primary criterion for transitioning an agency worker to direct employment. Some agency workers remain with the agency for several years before becoming eligible for direct employment by the client, based on their performance (W4). Certain respondents (T1, T2, C5), including representatives from the public office regulator (R), noted that agencies often function as a preliminary screening stage for clients, enabling them to evaluate candidates and select those who meet their needs while easily parting ways with those who do not. This practice places agency workers in a precarious position, as there is no legal requirement to establish objective, transparent, and pre-defined criteria for transitioning workers. Consequently, client companies can make selection decisions based solely on their own preferences and interests.
TAW offers the chance for agency workers to transition to more challenging positions with better salaries while remaining under the agency’s employment, as explained by agency worker respondent (W2). However, these career advancements are not necessarily transferable to the client if the agency worker is directly employed by the client. Instead, such workers often start at the bottom of the career ladder, as explained by the trade union representative (T1).
At times, clients continue to use agency workers instead of directly employing them because of the commission they would need to pay for early termination of the contract with the agency. Such a commission effectively ties the agency worker to the agency. Agency respondents (A1–A7) expressed a unified view on the potential transition of a worker to direct employment; they see it as a positive outcome, indicating client satisfaction and successful service delivery.
‘You always know that eventually, you will lose a worker. You can’t keep them there indefinitely. I see this as a positive thing, as it means that the client was satisfied with your service. The worker will also be content if they find direct employment there. That’s really the essence of our job’.
(A5)
Agency-respondent (A6) explained that clients, who are semipublic organizations, may initially hire workers through agencies for a trial period, which basically replaces the probation period. By doing so, they try to avoid lengthy public employment procedures. After this, if the worker is deemed suitable, he or she transitions to a permanent position.

3.1.5. Ensuring a More Flexible, Present, and Motivated Workforce

The fifth driver concerns a perceived performance advantage of agency workers—described as more driven, highly motivated, flexible, and less prone to health-related absenteeism than directly employed staff. This view was frequently expressed by user organizations, moderately by agencies and workers, and was also acknowledged by trade unions.
Two client respondents (C1, C3) indicated that directly employing former agency workers often leads to an increase in sick leave and other absences, such as maternity leave. The fear of losing jobs or contracts not being extended makes agency workers more cautious about taking leave. Direct employment increases their sense of job security, leading to a perceived drop in productivity and increase in sick leave.
‘I can confidently say that sick leave is three times lower for agency workers at our company, as people genuinely strive to prove themselves. Our experience last year with hiring agency workers directly revealed that this approach was not satisfactory, resulting in less motivation and increased sick leave. Therefore, we have halted the current rehiring process’.
(C1)
Certain client participants (C2, C3) recognized that agency workers are often more motivated, largely driven by their desire for direct employment with the client company and a concern that they might be the first to be let go during layoffs.
Agency respondents (A1, A3, A6) viewed their role as extending beyond simply providing the workforce. They stressed that they often see themselves as advocates for workers, addressing their needs and rights and offering them extra attention and care that they might not receive otherwise.
‘Over 80% of our workforce demands are for less-educated workers, who often perform monotonous tasks such as shifting products from one box to another for decades. Such roles are typically underpaid, and the pool of willing workers is decreasing. Our agency prioritizes these workers, offering an extra degree of care and ensuring they feel valued’.
(A1)
The flexibility of agency workers is also evident in their willingness to work under less favourable conditions. One client respondent (C1) noted that some clients specifically request that workers be posted for night shifts. Conversely, one participating agency worker (W2) highlighted that shift timings are also flexible for agency workers at the client’s company. Many agency workers, especially parents, prefer night shifts to better balance work and childcare responsibilities.

3.1.6. Providing Career Guidance and Facilitating Transitions to More Suitable Working Conditions, Ensuring a Good Fit Between Client and Worker

The sixth driver pinpointed for the utilization of TAW in this study relates to addressing mismatches between worker qualifications and labour market demands or work conditions at clients and agency workers’ needs and desires. Structural unemployment represents a significant challenge within the Slovenian labour market, which is a trend also mirrored in the TAW arrangements. Beyond career guidance, agencies also play a pivotal role in matching user organizations with suitable workers and ensuring person–job fit; this was noted by all parties to the triangular TAW arrangement and explicitly corroborated by agency-worker respondents (W1, W2).
The roles of agencies are multifaceted, as stressed by respondents. Beyond just placement, agencies act as intermediaries, ensuring a ‘good fit’ between clients and workers (A2, A5, A7). In the rapidly changing job market, agencies are no longer just tapping into the pool of unemployed individuals. Instead, they have pivoted to ‘transitioning individuals from less suitable roles to more fitting ones’ (A1, A2), thereby broadening employment opportunities.
‘We’ve had instances where workers we placed returned to us after 2–3 years because something had changed in the company or because the job did not meet their expectations. The career path is a highly intriguing experience, and in today’s times, every individual, regardless of the position they hold, needs a broad array of competencies if they wish to succeed on their upwards career path’.
(A2)
When a worker’s education, experience, or skills are not aligned with market demand, agency respondents (A5, A7) explained that they step in to provide specialized guidance. However, as highlighted by several agencies (A2, A4, A5, A6, A7), there is a noticeable lack of motivation among workers to pursue retraining.

3.1.7. Partial Outsourcing of the HRM Function and Managing Workers’ Quotas

The seventh driver concerns the partial outsourcing of HRM processes by user organizations. This theme was frequently reported by agencies and user organizations and corroborated by workers and trade unions. Agency respondents (A1, A2, A3, A5) explained that this need can, in part, stem from organizations’ strategies to minimize their direct labour costs by reducing the number of directly employed personnel while still ensuring an adequate labour supply through the services offered by agencies. This approach is particularly common in the branches of international companies, where such strategies are implemented by their headquarters.
In addition to placing workers, agency respondents (A1–A7) reported that they offer several services related to HRM. They select and recruit appropriate candidates for direct company hires. In addition, they offer a comprehensive range of HR services, such as annual performance reviews, training workshops, HR documentation guidance, and psychological assessments for staff selection and development. Additionally, they provide outplacement programs to support employees transitioning in the job market, both individually and in groups.
Agency respondents (A1, A2, A3, A4, A6, A7) emphasized—and workers corroborated (W2, W4)—that agencies provide a comprehensive support system for their workers at client premises.
‘Imagine working in a manufacturing company with 100 employees. The HR department would be swamped with tasks if it needed to address every concern. Now, think of us as that HR department, constantly communicating with those 100 individuals daily, or as needed. On an agency worker’s first day, we check in by asking the following: How do they feel? Did they receive all the necessary equipment? Were they adequately informed during their trial period? We also get feedback from the client about their initial impressions’.
(A7)

4. Discussion

The worldwide implementation of the TAW, including in the EU, serves as a strategic approach for managing workforce needs and ensuring fair, socially secure conditions for workers [2]. Its importance is heightened in volatile labor markets and in segments of markets affected by structural imbalances, as noted in previous research [10] and confirmed by this study. Over more than two decades, Slovenia’s TAW has evolved and thrived in sectors with persistent labour market gaps and those experiencing constant or rapid changes. These gaps have widened due to various labour market shifts in specific industries and global challenges such as economic crises, pandemics, and resource shortages arising from the Ukraine conflict. In Slovenia, TAW usage is closely linked to specific sectors, such as trade, transport, tourism, manufacturing, and construction [57], which is a trend also observed in other countries [10]. However, unlike earlier findings in Germany [43], Slovenia’s TAW usage is not confined to the production industry but instead extends to the service sector [57,62].
Specializing in niche areas of the Slovenian labour market, agencies face significant competition, with only a handful dominating the market [62]. Their growth and success are due to their agility and ability to quickly respond to market demands. Slovenian TAW-related legislation is mandating relatively high labour protection standards, with employment protection for temporary agency contracts slightly exceeding the OECD average [55,76]. Allen and colleagues [46] noted that business systems with more rigid labour markets, with systems of collective arrangements and strong unions are often linked to the use of more TAW. However, the implementation and enforcement has been challenging, as highlighted by the Slovenian Labour Inspectorate, which also noted the inconsistency of case law in the field of TAW [63,77]. Agency workers in Slovenia have reported encountering unequal treatment, particularly regarding benefits [64,65].
The preceding chapter addressed first research question (R1) by analysing the specific labour-market imbalances that user organizations sought to resolve through TAW and the flexibility enabled by agency involvement. Second research question (R2) examines the mechanisms within TAW intended to balance job security and social protection for agency workers. In Section 4.1, the analysis assesses the extent of social-protection considerations within each imbalance identified in Slovenia and addressed via TAW. The subsequent sub-chapter turns to the last research question (R3), evaluating the extent to which implemented TAW arrangements promote decent work—through job creation, adequate social protection, respect for rights and equal treatment, and stronger worker voice and representation—and thereby advance social sustainability, and, in an exploratory way, what indicative environmental co-benefits and risks are associated with these arrangements in terms of their sectoral placement and commuting/mobility patterns. Figure 2 illustrates the study’s analytical phases and results, linking the seven demand-side drivers through the SDG-8 pillars (used here as analytical mechanisms) to the identified gaps, the indicative environmental co-implications, and the resulting multi-level policy recommendations.

4.1. Balancing Job Security and Social Protection Across Identified Labour Market Imbalances (R2)

The study results show that there are three main drivers for client organizations in Slovenia to opt for TAW use, which are linked to structural imbalances in the national labour market: (a) medium-term fluctuations, (b) the seasonal nature of work in specific industry branches and (c) labour scarcity. Numerical flexibility emerged as the main TAW driver in previous early research [42] and later in EU-wide studies [8,10] and in country-specific research [47,48,49,50,51,54]. As explained by the study respondents, the ‘horizontal type’ of TAW employment enables partial flexibility of the workforce and releases the tension in the client organization caused by uncertain or unbalanced production or service demands. Client organizations are prepared to incur additional costs (covering the full salary, taxes, social security contributions and benefits for agency workers, identical to those of their directly employed workers, in addition to the agency fee) to efficiently and smoothly manage fluctuations in workforce needs. This approach contrasts with earlier empirical studies [41,42,53], which highlighted cost savings in wages and benefits as the primary motivations for employing TAW arrangements. Although in many EU countries, client organizations using the TAW are not responsible for additional benefits such as pensions, maternity leave, holidays and sickness pay, and redundancy costs, as indicated in prior research [8], in Slovenia, these obligations fall on the client organization.
The impact of transitions for agency workers—from client to agency and then to another client—on job stability and social security is partly mitigated by legislation that secures their employment with the agency. Both this and previous studies [37,62] show that more than 70 percent of agency workers in Slovenia hold permanent employment contracts, primarily because it is more cost-effective for agencies. While the law requires agencies to cover downtime between assignments, workers may still be terminated if the agency is unable to secure a new assignment. The study revealed that during market crises (such as the COVID-19 pandemic or resource shortages stemming from the Ukraine conflict), agency workers are among the first to lose their employment. The findings also suggest that in the event of layoffs, neither the client organization nor the agency is legally obligated to establish or follow specific criteria for selecting surplus agency workers, and trade unions have no influence or involvement in this process. This lack of regulation exposes agency workers to potential unequal treatment.
In addition to the risk of not receiving new assignments, transitioning from one client to another may lead to the loss of benefits provided by the previous client organization. It also requires adaptation to a new work environment and often involves taking on unfamiliar tasks.
The lack of objective, transparent, and predefined criteria is also evident when there is a change of employer, such as when a client organization directly hires a former agency worker who previously worked on its premises. This allows client companies to make selection decisions based solely on their preferences and interests, placing agency workers in a vulnerable position. Such transitions can be further discouraged by agencies, as clients are often required to pay a commission for early contract termination. The study suggests that clients may continue using agency workers instead of directly employing them to avoid these additional financial obligations. This commission effectively ties the agency worker to the agency and limits their ability to secure direct employment with the client.
Furthermore, TAW may be the only available pathway to employment at some client organizations, but it can eventually lead to fixed-term or permanent positions, as indicated in the study. This approach is a common HRM strategy for some organizations, which effectively compels workers who wish to be employed there to start as agency workers. This strategy may be employed to manage workforce quotas or to use agency work as a trial period for identifying valuable employees.
The findings indicate that as a result, agency workers remain in a state of uncertainty and under constant pressure to prove themselves daily. Compared to workers who secure permanent contracts with the company, agency workers face additional benchmarks to demonstrate their value and must overcome more obstacles to achieve promotions, better pay, or more demanding roles. These benchmarks arise at each transition between clients and after each assignment ends, when the client selects agency workers for direct employment. The inability to retain promotions and benefits gained at one client organization represents a significant career obstacle for agency workers, as highlighted in this study.
Seasonality and labour scarcity are two TAW drivers that emerged in this study but are far less mentioned in previous research, although [10] indicated the potential of the TAW in these specific areas. Study respondents agreed that seasonality is especially well managed by agencies since agencies have the resources to transfer agency workers from one client to another and the knowledge to manage such transitions (especially in the field of assuring the legally compliant passage of the worker from the umbrella of rights under one collective agreement to another). However, legally, there is no requirement to apply the more favorable collective agreement; instead, the one binding for the client where the agency worker is currently employed must be followed. This can lead to workers losing (or gaining) certain rights and benefits on a semi-annual basis.
The use of the TAW also serves as a protective measure for organizations, shielding their manpower from overload during peak seasons, as indicated in earlier research [45] and confirmed in this study.
Labour scarcity in specific professions and for lower skill workers is another market gap that is successfully managed by agencies. In this segment, agencies might even opt for a strategy to permanently employ larger shares of workers from scarce professions and attract these workers by providing them with better salaries, work conditions and benefits that are assured by collective agreements and offered by client organizations. This strategy enables agencies to gain a (partial) monopoly in certain high-demand professions in the market while ensuring better conditions for agency workers. In doing so, agencies enhance their reputation as good employers, confirming that reputational concerns are prevalent in the Slovenian TAW sector, as indicated in recent studies [21,34]. However, such practices risk creating new labour market segments, as shown in previous research [25], and do not necessarily support long-term decent work conditions.
The study results also indicate that specific human resource strategies are tied to the use of the TAW by respondents; this confirms the findings presented by Thommes and Weiland [43], who noted that the existence of a well-structured HR department within clients’ organizations has a significant influence on the use of the TAW. The results of this study provide a clear indication of the nature of such influence by highlighting several strategies that clients’ HR departments use in relation to the TAW. Client respondents opted for TAW use as an alternative to probationary or temporary employment (at the client’s organization). This can lead to prolonged uncertainty for agency workers, as they may start by working as agency employees at a client’s organization, then transition to the client directly, only to receive multiple fixed-term contracts.
Ease of recruitment has previously emerged as an important reason for TAW utilization [8,10], while previous studies have linked this phenomenon to the possibility of evaluating agency workers’ suitability with minimal commitment.
Clients have opted for TAW use to boost motivation and flexibility among their workforce. With two groups of workers present in the client-organization environment, thereby allowing for comparison, the client respondents noted that directly employing agency workers often results in lower work-related motivation and an increase in sick leave, which is a phenomenon that has been previously recognized in other studies [42,78,79]. This situation also leads to problematic tactics by clients, exploiting agency workers’ willingness to work under less favourable conditions due to fears of job loss or contract nonrenewal. As indicated in other studies [42], this can severely impact the well-being of agency workers. As a direct consequence, agency respondents often view themselves as advocates for agency workers, highlighting that such worker segmentation within a client’s organization can lead to opportunities for unequal [22] and unfair treatment [24,34]. The findings also highlight agency workers’ concerns regarding the security and long-term prospects of their positions within client organizations, which may result in strategic behaviors such as avoiding trade union membership.
The TAW was also reported to address mismatches between clients and workers by decoupling them and finding better matches for both sides. Client respondents noted that their HR departments benefited from TAW use, as it allowed for the partial outsourcing of HRM functions. This approach is particularly useful in multinational companies, where it can serve as a strategy to reduce the number of directly employed workers or manage worker quotas, in line with directives from the client’s headquarters. Outsourcing is utilized not only to circumvent the responsibilities associated with standard employment relationships, such as selection and payroll administration [8], but also as a strategic approach to enhance the power of HRM departments and tactically bypass externally imposed regulations. However, this practice imposes a hidden cost on agency workers, as they often become ‘locked into’ various temporary arrangements before eventually securing a permanent position with the company.

4.2. Assessment of Decent Work (SDG-8) Outcomes, Social-Sustainability Gaps and Environmental Co-Implications (R3)

The second part of this Section evaluates the extent to which implemented TAW arrangements promote decent work and identifies gaps that impede progress toward social sustainability while also tracing, in an exploratory way, indicative environmental co-benefits and risks. The analysis is organized around the four SDG-8 mechanisms (also ‘pillars’): employment creation, rights at work, social protection, and social dialogue (Section 4.2.1, Section 4.2.2, Section 4.2.3 and Section 4.2.4). Section 4.2.5 synthesises the environmental co-implications of TAW that emerge from the data, focusing on sectoral location, commuting and mobility patterns, and work-organisation choices. Taken together, these findings provide the basis for the multi-level policy recommendations presented in the subsequent, final chapter.

4.2.1. Employment Creation

The study results suggest that the implementation of TAW arrangements in Slovenia contributes partially to employment creation, particularly by providing job opportunities for long-term unemployed individuals, workers with disabilities, migrant workers, and other disadvantaged groups. Participants reported that two-thirds of the workers placed by agencies were previously unemployed, positioning TAW as a gateway into the labor market. Additionally, the findings indicate that organizations are more inclined to hire these workers when they are employed through an agency.
Agencies have demonstrated their value in addressing the challenge of seasonality in labor demand by ensuring continuous employment for agency workers, despite the fluctuating needs of different industries throughout the year. Another labor market challenge successfully managed within TAW arrangements is labor scarcity. Agencies tackle this issue by actively recruiting specialized workers, offering competitive compensation through agency employment, and retaining workers within the TAW even during client downtime, ensuring their availability for future assignments.
However, the job opportunities provided through TAW are not necessarily productive and generally offer minimal or slightly above minimal wages. Agency workers often occupy some of the lowest-paid positions, a trend particularly evident in the Slovenian TAW sector, as reported by SiStat [57]. Nearly one in three agency workers earns only up to 60 percent of the average net salary [70] (p. 39). This is partly due to the undemanding nature of the work offered through TAW arrangements and the low qualification requirements for workers performing these tasks. The workforce is predominantly composed of individuals with secondary vocational education or equivalent qualifications [62]. Study participants reported that approximately four-fifths of labour demand mediated through TAW concerns lower-qualified workers engaged in repetitive, on-site tasks (e.g., prolonged manual transfer of products between containers). An important exception concerned high-demand occupations: agencies reported negotiating with clients to accommodate agency workers’ preferences for greater flexibility, particularly remote or hybrid arrangements enabled by wider information and communication technology (ICT) adoption.
The study also found that TAW can provide opportunities for agency workers to transition to more challenging positions with higher salaries while remaining under the agency’s employment. Agencies assist workers in moving from less suitable roles to those better aligned with their skills, thereby expanding employment opportunities. When a worker’s education, experience, or skills do not match market demand, agencies offer specialized guidance. However, the results indicate a noticeable lack of motivation among agency workers to pursue retraining.
The productivity of job opportunities facilitated through TAW is limited by the typically short duration of these engagements, often lasting only a few months. Moreover, career advancements achieved within TAW are not necessarily transferable to the client if the agency worker is eventually employed directly by the client. Such workers often have to start at the bottom of the career ladder. Additionally, when a worker transitions from one client to another, this may result in the loss of benefits provided by the previous client organization, the need to adapt to a new work environment, and the necessity of taking on unfamiliar tasks.

4.2.2. Equality of Rights at Work

Although the law, particularly the ERA-1, mandates equality of rights at work, ensuring equal opportunities and treatment for both directly employed workers and agency workers at client organizations, this study reveals several practices and strategies employed by agencies and client organizations that undermine the personal development and career advancement of agency workers:
  • Uncertainty and Constant Pressure: Agency workers are often subjected to a persistent state of uncertainty and pressure to prove their worth daily. They face additional benchmarks to demonstrate their value, encountering more obstacles when seeking promotions, higher pay, or more challenging roles. These hurdles are particularly evident during transitions between clients and after assignments end, when clients may consider agency workers for direct employment.
  • Career Obstacles: The inability to retain promotions and benefits earned at one client organization poses a significant career obstacle for agency workers. This study highlights how such limitations can hinder their long-term professional growth.
  • Prolonged Path to Permanent Employment: Agency workers frequently receive multiple fixed-term contracts, particularly during transitions to direct employment with a client at the end of a task. This practice prolongs their path to securing a permanent position and may keep them in a precarious employment situation.
  • Financial Barriers to Direct Employment: Client organizations are often required to pay a commission to the agency for early contract termination if they wish to directly employ an agency worker before the assignment ends. The study suggests that to avoid these additional financial obligations, clients may continue to use agency workers rather than offer them direct employment. This commission effectively binds the agency worker to the agency, limiting their opportunities for direct employment with the client.
  • Strategic Use of Agency Work: Especially in multinational companies, client organizations may use agency work strategically to reduce the number of directly employed workers or manage worker quotas, as directed by their headquarters. As a result, agency workers often find themselves ‘locked into’ various temporary arrangements before they can secure a permanent position within the company.
  • Market Monopolization by Agencies: Agencies may also seek to establish a (partial) monopoly in high-demand professions. While this strategy might temporarily improve conditions for agency workers, it also risks creating new labor market segments that do not necessarily support long-term, decent work conditions.

4.2.3. Social Protection

Social protection—encompassing safe work environments, social security, and inclusion—for agency workers in Slovenia has been progressively addressed through legislation and its amendments [37] and is enforced by the Labour Inspectorate of the Republic of Slovenia [63,77] and the courts [71]. Nevertheless, the study identifies several stakeholder practices that erode agency workers’ social security, compromise occupational safety, and impede their effective integration into user organizations:
  • Avoidance of Liability for Workplace Harm: An increasing number of legal disputes highlight the challenge of holding agencies and client organizations accountable for workplace injuries sustained by agency workers. Court cases [71] indicate that both agencies and clients often deny responsibility, leaving agency workers in precarious positions without clear recourse for compensation or protection.
  • Pressure to Perform and Always Be Available: Agency workers, already facing significant career obstacles, are particularly vulnerable to exploitation. Clients may take advantage of their willingness to work under less favorable conditions—such as night shifts or high-pressure environments—driven by fears of job loss or contract nonrenewal. The study indicates lower rates of health-related absenteeism among agency workers, suggesting that they may avoid taking sick leave to maintain their positions.
  • Extended Trial Periods: Agencies often serve as a preliminary screening tool for client organizations, allowing clients to assess workers before committing to permanent employment. This practice is particularly common in semi-public organizations, where agency workers are initially hired for a trial period, effectively replacing the traditional probation period. If deemed suitable, these workers may then transition to permanent positions, sometimes facing yet another trial period.
  • Vulnerability During Market Crises: The study found that agency workers are disproportionately affected during market crises, such as the COVID-19 pandemic or resource shortages linked to the Ukraine conflict. They are often the first to lose their jobs when client assignments are terminated early. If agencies cannot secure a new assignment within four weeks, these workers are typically let go.
  • Lack of Objective Criteria for Surplus Selection: In the event of mass early terminations of agency workers at client organizations, there are no objective, transparent, or predefined criteria for selecting surplus workers. This lack of regulation exposes agency workers to potential unequal treatment, as decisions are made based on the client’s discretion rather than clear, fair standards.

4.2.4. Social Dialogue

Social dialogue, which empowers workers to voice concerns, organize, and engage in decisions affecting their lives, remains underutilized and largely unexplored in the context of TAW, as indicated by the study results. Collective bargaining related to TAW and the involvement of agency workers in trade unions, as well as their participation in management, are areas particularly lacking in development. Agency workers have the freedom to join the relevant union for the sector, occupation, or workplace in which they are employed. However, the estimated trade union density for agency workers in Slovenia is just 0.18 percent [10] (p. 21), compared to 27 percent for all employees in the country [80]. Results indicate that agency workers are often reluctant to join unions due to concerns that it may be viewed unfavorably by the client organization. Trade unions have raised concerns about the rights of agency workers and reported instances where agency workers were discouraged from joining or organizing within unions [62,64].
Although ERA-1 permits extensive collective bargaining in the TAW sector [38], study participants reported that agency workers are rarely discussed during negotiations. As a result, collective agreements typically contain only minimal provisions pertaining to agency workers, primarily ensuring equal treatment and access to social security. Despite clear evidence of the potential benefits of social innovation and environmentally sustainable practices in mitigating the negative impacts of TAW and enhancing the sector’s overall sustainability [13,16], such measures remain almost entirely absent in practice. Furthermore, the current legal framework on worker participation—particularly the Worker Participation in Management Act (Official Gazette of the RS, No. 42/93, 61/00, 56/01, 26/07, 45/08)—presents an additional obstacle by excluding agency workers from participatory mechanisms, as they are not formally recognized as employees of the client organization.

4.2.5. Environmental Co-Implications of TAW

The qualitative findings suggest that the deployment of TAW in Slovenia has clear, if indirect, environmental co-implications linked to sectoral concentration, mobility and commuting patterns, and work-organisation choices.
As shown in Section 1.2, TAW in Slovenia is heavily used in trade, transport, tourism, manufacturing and construction—branches that are central to the green transition and comparatively resource- and emission-intensive. Within these sectors, agency workers are predominantly employed in lower-qualified, repetitive on-site roles, while a smaller segment of higher-skilled workers (for example in IT and legal services) perform remote or hybrid work supported by ICT. This configuration implies that most TAW-mediated labour is embedded in energy- and material-intensive production or service processes, while only a minority of assignments currently leverage the potential environmental benefits of flexible, digitally enabled work arrangements.
Mobility and commuting emerged as a second important environmental dimension. Agencies described both ‘horizontal’ placement patterns—assignments to a single client for a period—and ‘vertical’ seasonal rotations that move workers between geographically distant sites, such as an airport and a ski resort. While vertical rotations can support year-round employment and smooth seasonality, respondents noted that such arrangements often entail increased commuting and business travel, especially when workers are relocated across regions. By contrast, in specialised, high-demand occupations, agencies reported negotiating remote or hybrid work options to accommodate worker preferences and retain scarce skills. In these cases, respondents described effective hybrid collaboration between agency and directly employed staff and highlighted the potential for remote-eligibility criteria and minimum hybrid-day targets to reduce commuting and associated emissions. Overall, the environmental potential of TAW appears underutilised in lower-skilled, on-site segments but increasingly attainable in skill-scarce domains through deliberate governance of work location and mobility.
Work-organisation and staffing strategies linked to the main demand-side drivers—particularly volatility buffering (D1) and seasonality (D2)—also have environmental co-implications. Better alignment of staffing levels with actual demand can reduce idle facility time, over-staffed shifts and inefficient use of buildings and equipment, thereby lowering energy use per unit of output. However, several respondents emphasised that short assignment durations and high turnover hinder the implementation of more sustainable staffing measures, such as commute planning, proximity-based rostering and pooled low-carbon transport. These practices depend on relatively stable rosters, predictable schedules and sufficient lead time to coordinate across workers and sites; in high-churn environments, agencies and clients often judge the additional coordination effort and cost to outweigh near-term benefits and therefore do not prioritise them.
Taken together, these patterns indicate that the same organisational drivers that shape social outcomes in TAW also structure its likely environmental co-benefits and risks. However, it is important to underscore that the environmental analysis in this study is exploratory and indicative: no direct environmental performance indicators (such as energy use, emissions or life-cycle measures) were collected, and inferences are based on sectoral profiles, mobility and work-organisation patterns, and respondents’ accounts.

5. Conclusions

5.1. Policy Recommendations, Addressing the Social and Environmental Sustainability Gaps

To advance social sustainability in the TAW sector while beginning to realise its environmental co-benefits and manage related risks identified in this study, policy measures are proposed across four layers: (i) regulation, (ii) industry-level approaches, (iii) TAW agency approaches, and (iv) social-dialogue platforms. The primary emphasis remains on decent work and social protection; recommendations on environmental aspects reflect indicative co-implications (sectoral concentration, mobility patterns and work-organisation choices) rather than direct environmental measurement.
Regulatory measures should strengthen collective bargaining coverage and access for agency workers. This can be achieved through amendments to the LMRA and the ERA-1 to mandate their inclusion in social dialogue. Legislation should also require sectoral clauses on career pathways, regular audits verifying equal pay and benefits for agency and directly employed workers in like-for-like roles, clear, published criteria and timelines for conversion from agency placement to direct employment, and access to training for agency workers—shifting attention beyond illegality prevention and stricter conditions for establishing TAW agencies and explicitly linking employment-quality standards with green-transition and climate strategies in high-TAW sectors. Clear, objective ERA-1 standards for regularisation after trial use are needed to prevent cycling of temporary assignments, as trial use is an important driver of TAW. Longer and more predictable placements would also make it easier for agencies and user organisations to implement sustainable staffing measures (such as commute planning, proximity-based rostering and pooled low-carbon transport) that are currently difficult to organise under very short assignments. Joint responsibility for occupational safety and health should be reinforced with clear liability and investigation pathways, addressing enforcement gaps noted in documents and interviews. Licensing and compliance should be made more transparent by publishing easy-to-read, public dashboards for sectors with high TAW usage. These dashboards should show key indicators (such as the number of licensed agencies, inspection results, violations, and corrective actions) so inspectors and stakeholders can focus oversight on sectors where demand volatility and labour shortages are consistently high. Where feasible, these dashboards could also include simple, environmentally relevant indicators—such as the prevalence of long-distance ‘vertical’ rotations and the share of remote/hybrid placements—to inform just-transition discussions without imposing complex reporting burdens. Participation rights should ensure agency workers are represented in company-level trade unions (and, where applicable, works councils) and on health and safety committees at user organizations, directly addressing the documented lack of effective representation. Revising the Worker Participation in Management Act to allow some form of representative participation for agency workers, particularly in matters affecting working conditions, work location, mobility and green workplace practices, could foster a more inclusive and socially innovative governance model, better aligned with contemporary labour market dynamics and sustainability goals. Another possibility is to form a trade union specifically dedicated to TAW workers, following examples from other countries [10].
At the industry/user-organization level, sectoral training funds should be established to finance upskilling and reskilling and should allow agency workers to keep their training entitlements when they move between assignments. This directly addresses the gap shown in the results: agency workers value clear transition opportunities and good skill–job matching, while user organizations report ongoing skill shortages. In parallel, sectoral-level guidelines should support the use of remote and hybrid work options in appropriate high-skill assignments, with clear criteria co-negotiated by social partners, so that digital solutions reduce commuting and building use without undermining job security or equal treatment. Scheduling rules should require a minimum notice period, as agency workers frequently face short-notice changes driven by seasonality and medium-term demand fluctuations, which makes commuting, childcare, and second jobs much harder and contributes to the uncertainty highlighted in interviews. Sectoral agreements and guidance should encourage sustainable staffing practices—such as commute planning, proximity-based rostering and pooled low-carbon transport—whenever assignment duration and stability permit, thereby reducing unnecessary travel while improving predictability for workers. Publishing simple, public metrics that show how often agency placements lead to direct employment—and setting clear steps and timelines for that process—makes the route to permanent jobs transparent and prevents repeated use of ‘trial’ placements without real prospects of hire. Regular equal-treatment audits, required by sectoral collective agreements in high-TAW sectors, should compare wages, benefits, training access, and working conditions for agency versus directly employed workers in comparable roles, with corrective plans overseen by joint committees; this responds to evidence of unequal-treatment risks and shifts attention from mere legal compliance toward actual outcomes. Where relevant, these audits could also review whether agency workers are disproportionately allocated to more physically demanding or environmentally onerous tasks compared to directly employed staff.
At the TAW agency level, agencies should publish a short, standard report each year that either shows or explains the following: how long assignments usually last, how many workers move into direct jobs with user organizations, how many hours of training workers receive, how many grievances are filed and how quickly they are resolved, any health-and-safety incidents, and how the agency engages in social dialogue, as well as basic information on the spatial pattern of placements (for example, the share of local versus long-distance assignments and the availability of remote/hybrid options). Agencies should also adopt clear ethical rules that ban discriminatory screening and forbid any interference with worker representation; commercial incentives should reward agencies when workers transition to direct jobs and when equal treatment is achieved. Agencies should actively seek to minimise avoidable long-distance rotations by prioritising local matching where possible and by collaborating with user organisations on remote/hybrid solutions in suitable occupations and, where rosters are sufficiently stable, on pooled or low-carbon transport options for commuting. Finally, agencies should offer structured support between assignments—such as brief bridging arrangements, skills assessment, career guidance, and re-placement services—to make job matching and transitions work in practice.
For social-dialogue platforms, the results indicate the need for a standing national TAW forum, bringing together the ministry, the employment service, the labour inspectorate, employer associations and the association of TAW agencies, (sectoral) trade unions, and civil-society. To unlock the potential for greater social and environmental sustainability in the TAW sector, several measures linked to social dialogue could be considered in this forum. This forum should track sector trends and review the public agency registry and disclosure reports, including indicative environmental co-implications (such as commuting and rotation patterns in high-TAW sectors), and co-develop clear, practical guidance on conversion thresholds (when agency roles should become direct employments), fair scheduling, and representation of agency workers at user organizations as well as principles for sustainable staffing and green workplace practices in triangular employment arrangements. To move quickly from guidance to practice, the ministry should mandate time-bound working groups (co-led by user organizations, agencies, and unions) to draft model collective-agreement clauses and enterprise templates on portable training, responsibility/liability, and worker voice. Finally, the forum should publish an annual brief—compiled by the employment service and labour inspectorate—summarizing volatility/shortage/seasonality patterns, transition outcomes, and enforcement results, and, as data become available, simple indicators on mobility patterns, remote/hybrid uptake and other environmentally relevant aspects of TAW, so policy and bargaining agendas are updated each year and progress on closing the gaps can be measured.

5.2. Original Contribution of the Study

This study advances TAW scholarship by providing an integrated, evidence-based explanation of why user organizations adopt TAW and how these drivers manifest in stakeholder practices that align—or fail to align—with the SDG-8 dimensions of decent work, with the SDG-8 pillars used in this paper as the analytical mechanisms. Building on the UN/ILO understanding of SDG 8 as integrating economic, social and environmental dimensions, the analysis also traces how the same demand-side drivers and practices carry indicative environmental co-implications, particularly through sectoral concentration in more emission- and resource-intensive branches, commuting and mobility patterns, and work-organisation choices in triangular employment settings. Conceptually, it proposes and applies a concise analytic framework that links demand-side drivers (D1–D7) to SDG-8 pillars (job creation, social protection, rights at work, social dialogue), making explicit the pathways from organizational motives (e.g., volatility buffering, labour scarcity, trial use, partial HRM externalization) to measurable employment outcomes and governance gaps. The framework is extended through a secondary coding lens that identifies environment-related themes and links them back to the driver–pillar configuration, illustrating how TAW governance can generate environmental co-benefits or risks alongside social outcomes. Empirically, it provides country-level, multi-stakeholder evidence for Slovenia based on an explanatory sequential, qualitative-dominant mixed-methods case study, triangulating administrative/statistical sources with 19 in-depth interviews across agencies, user organizations, agency workers, trade unions, and the state. Practically, it moves beyond compliance narratives by diagnosing actionable social-sustainability deficits—most notably weak worker voice, unclear and non-transparent rules for moving from agency placement to direct employment, gaps in equal pay and treatment, and ad hoc sustainability efforts—and translating them into a multi-level policy roadmap (regulatory, industry/user-organization, agency, and social-dialogue platforms). The roadmap also highlights where modest changes in governance of mobility, work location and assignment design could yield environmental co-benefits (for example, through reduced commuting and more efficient use of facilities) without compromising decent-work objectives. Together, these contributions fill a well-noted gap in the literature, which has emphasized market structures and legal templates but has rarely coupled organizational drivers with SDG-aligned outcomes in a traceable, policy-relevant manner.

5.3. Research Study Limitations and Further Research

This qualitative-dominant, explanatory sequential case study targets legally operating TAW in Slovenia and is designed for analytic generalisation rather than statistical inference. The 19 respondents—purposefully selected across five stakeholder groups (agencies, user organizations, agency workers, trade unions, and a state authority)—were information-rich and sampled for maximum variation (e.g., large/small agencies; association members/non-members; high-prevalence branches plus others). Sufficiency was judged by information richness, coverage of key positions, and convergence across sources; Phase-1 documentary/statistical mapping narrowed the scope, and by the final interviews no new driver categories emerged, indicating thematic sufficiency for the study aims. This evaluation of sufficiency follows recent guidance that emphasises information power and conceptual sufficiency, namely that focused aims, specific and knowledgeable participants, strong theoretical scaffolding and systematic thematic analysis can justify relatively small but carefully composed qualitative samples for analytic inference [74,75].
Nonetheless, several scope limits apply: (i) unregistered or illicit operators were not included; (ii) cross-border TAW arrangements were outside the scope; (iii) student work—often counted in Eurostat tabulations—was excluded because Slovenian law does not classify it as TAW; (iv) the theoretical framework is EU-anchored, focusing on the post-Directive 2008/104/EC period, which limits generalisability beyond the EU; and (v) the qualitative sample is suitable for analytic generalisation within the Slovenian case but does not support statistical generalisation. As with all interview-based research, self-report and social-desirability bias are possible; these risks were mitigated through triangulation with administrative/statistical sources, probe design, and an auditable NVivo workflow (coding memos/decision logs). The coded-intensity matrix summarises salience across groups, not population frequencies, and the analysis centres on demand-side drivers and the SDG-8 (social) pillars.
Additional limits apply specifically to the environmental dimension. Environmental aspects are addressed only indirectly, via sectoral profiles, respondents’ accounts of mobility and commuting, work-location decisions (on-site vs. remote/hybrid) and work-organisation practices; no dedicated environmental performance indicators (such as energy use, emissions, life-cycle measures or building-level data) were collected. As a result, the environmental findings should be read as exploratory and indicative, identifying plausible co-benefits and risks rather than demonstrating causal environmental effects or permitting quantitative comparison across sectors or countries.
Further research into the global implications of TAW is essential, particularly concerning regulatory instruments that complement traditional legislation. These mechanisms, commonly referred to as self-regulation, include collective bargaining, contractual relations and voluntary agreements between stakeholders. Exploring the potential of such tools to enhance the effectiveness, legitimacy and transparency of the TAW sector could offer valuable insights. In particular, comparative studies examining how self-regulation interacts with formal legislation in fostering better working conditions and protecting agency workers’ rights across diverse legal systems would be especially valuable.
In the Slovenian context, several areas warrant closer investigation. Research should focus on the work–life balance of TAW workers, as this is crucial for aligning flexible work arrangements with overall well-being. The gender divide within TAW also remains largely underexplored, particularly in understanding how women are affected by TAW structures and practices, which could inform targeted policies promoting gender equality. Additionally, the impact of emerging technologies—especially AI-driven platforms used in the TAW sector—needs systematic analysis, as these tools are reshaping how agency work is organised and governed, with implications for hiring processes, job matching and worker surveillance.
Building on this study’s exploratory insights into environmental co-implications, future research should link detailed TAW micro-data with quantitative environmental indicators (for example, commuting distances, sectoral emission factors and organisational energy-use data) in mixed-methods or longitudinal designs, and examine how specific regulatory or collective-bargaining interventions in TAW affect both decent-work outcomes and environmental performance over time. Taken together, further research in these areas could provide a comprehensive understanding of the evolving dynamics in the TAW sector, informing not only national but also international policy frameworks.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/su172411261/s1, File S1: The interview topic guide; File S2: The codebook.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Institutional Review Board Statement

The study received Research Ethics Committee approval. The study was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The author declares no conflicts of interest.

References

  1. Forde, C.; Slater, G. Temporary agency work: Evolution, regulation and implications for performance. J. Organ. Eff. 2016, 3, 312–322. [Google Scholar] [CrossRef]
  2. International Labour Organization—ILO. Non-Standard Forms of Employment in Selected Countries in Central and Eastern Europe. A Critical Glance into Regulation and Implementation. ILO Decent Work Technical Support Team and Country Office for Central and Eastern Europe. 2021. Available online: https://www.ilo.org/wcmsp5/groups/public/---europe/---ro-geneva/---sro-budapest/documents/publication/wcms_793096.pdf (accessed on 29 October 2025).
  3. International Labour Organization—ILO. Non-Standard Employment Around the World: Understanding Challenges, Shaping Prospects. International Labour Office, Geneva, 2016. Available online: https://www.ilo.org/sites/default/files/wcmsp5/groups/public/%40dgreports/%40dcomm/%40publ/documents/publication/wcms_534326.pdf (accessed on 29 October 2025).
  4. Eurofound. New Forms of Employment–2020 Update; Publications Office of the European Union: Luxembourg, 2020; Available online: https://data.europa.eu/doi/10.2806/278670 (accessed on 29 October 2025).
  5. Spattini, S. Agency Work: A Comparative Analysis. E-J. Int. Comp. Labour Stud. 2012, 1, 169–210. [Google Scholar]
  6. Sartori, A. Temporary Agency Work in Europe: Degree of Convergence following Directive 2008/104/EU. Eur. Labour Law J. 2016, 7, 109–125. [Google Scholar] [CrossRef]
  7. Alsos, K.; Evans, C. Temporary work agencies: Triangular disorganization or multilevel regulation? Eur. J. Ind. Relat. 2018, 24, 391–407. [Google Scholar] [CrossRef]
  8. Voss, E.; Vitols, K.; Farvaque, N.; Broughton, A.; Behling, F.; Dota, F.; Leonardi, S.; Naedenoen, F. The Role of Temporary Agency Work and Labour Market Transition in Europe; Final Report for the Joint Eurociett/UNI Europa Project; Eurociett: Hamburg, Germany, 2013; pp. 1–167. Available online: https://www.weceurope.org/uploads/2019/07/2013_WECEU_Study-TAW-Role-Labour-Transitions.pdf (accessed on 29 October 2025).
  9. International Labour Organization—ILO. Decent Work and the 2030 Agenda for Sustainable Development. International Labour Organization. 2017. Available online: https://ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/publication/wcms_436923.pdf (accessed on 29 October 2025).
  10. Eurofound. Temporary Agency Work and Collective Bargaining in the EU; Eurofound: Dublin, Ireland, 2009. Available online: https://assets.eurofound.europa.eu/f/279033/8f15d95c92/ef0899en.pdf (accessed on 29 October 2025).
  11. van Oorschot, S. Digitalisation and Temporary Agencies: Impact on the Business Model and Internal Organisation. Res. Hosp. Manag. 2021, 11, 241–248. [Google Scholar] [CrossRef]
  12. Potocka-Sionek, N. The Changing Nature of Labour Intermediation: Do Algorithms Redefine Temporary Agency Work? In New Forms of Employment: Current Problems and Future Challenges; Wratny, J., Ludera-Ruszel, A., Eds.; Springer Fachmedien: Wiesbaden, Germany, 2020; pp. 169–190. [Google Scholar] [CrossRef]
  13. You, C.; Qiu, H.; Pi, Z.; Yu, M. Sustainable Enterprise Development in the Manufacturing Sector: Flexible Employment and Innovation in China. Sustainability 2023, 15, 8180. [Google Scholar] [CrossRef]
  14. World Employment Confederation—WEC-Europe. Making Better Matches: How the Private Employment Services Industry Helps to Reduce Labour and Skills Shortages in Europe. A World Employment Confederation-Europe Strategic Issue Paper. 30 August 2022. Available online: https://weceurope.org/publication-post/making-better-matches-how-the-private-employment-services-industry-helps-to-reduce-labour-and-skills-shortages-in-europe/ (accessed on 29 October 2025).
  15. World Employment Confederation—WEC-Europe. COVID-19: The Road to Recovery: Private Employment Services Stand Ready to Play Their Part. World Employment Confederation-Europe. 2020. Available online: https://wecglobal.org/uploads/2020/07/WEC-Policy-Paper-Covid-19-The-road-to-recovery-FINAL.pdf (accessed on 29 October 2025).
  16. Baiocco, S.; de Groen, W.P.; Kilhoffer, Z.; Lenaerts, K. Social Innovation in the Temporary Agency Work Industry. Centre for European Policy Studies (CEPS) and the Research Institute for Work and Society (HIVA) at KU Leuven. 2020. Available online: https://www.weceurope.org/uploads/2020/12/20201201_CEPS-HIVA_Social-innovation-TAW-research-report.pdf (accessed on 29 October 2025).
  17. International Labour Organization—ILO. The Impact of the COVID-19 Pandemic on Jobs and Incomes in G20 Economies. International Labour Organization. 2020. Available online: https://www.ilo.org/wcmsp5/groups/public/---dgreports/---cabinet/documents/publication/wcms_756331.pdf (accessed on 29 October 2025).
  18. Möhring, K.; Naumann, E.; Reifenscheid, M.; Wenz, A.; Retttig, T.; Krieger, U.; Friedel, S.; Finkel, M.; Cornesse, C.; Blom, A.G. The COVID-19 pandemic and subjective well-being: Longitudinal evidence on satisfaction with work and family. Eur. Soc. 2021, 23, S601–S617. [Google Scholar] [CrossRef]
  19. Nieuwenhuis, R.; Yerkes, M.A. Workers’ well-being in the context of the first year of the COVID-19 pandemic. Community Work Fam. 2021, 24, 226–235. [Google Scholar] [CrossRef]
  20. Wu, Q. Employment Precarity, COVID-19 Risk, and Workers’ Well-Being During the Pandemic in Europe. Work Occup. 2023, 50, 188–211. [Google Scholar] [CrossRef]
  21. Cajander, N.; Reiman, A. High performance work practices and well-being at restaurant work. Eur. J. Tour. Hosp. Recreat. 2019, 9, 38–48. [Google Scholar] [CrossRef]
  22. Akkerman, A.; Sluiter, R.; Jansen, G. Temporary work and deviant behavior: The role of workplace cohesion. Sociol. Q. 2020, 61, 678–702. [Google Scholar] [CrossRef]
  23. Strauss-Raats, P. Temporary safety: Regulating working conditions in temporary agency work. Saf. Sci. 2019, 112, 213–222. [Google Scholar] [CrossRef]
  24. Knox, A. Regulatory avoidance in the temporary work agency industry: Evidence from Australia. Econ. Labour Relat. Rev. 2018, 29, 190–206. [Google Scholar] [CrossRef]
  25. Fostervold, K.I.; Koren, P.C.; Nilsen, O.V. Defining sustainable and “decent” work for human factors and ergonomics. In Ergonomics and Human Factors for a Sustainable Future—Current Research and Future Possibilities; Thatcher, A., Yeow, P.H.P., Eds.; Palgrave Macmillan: Singapore, 2018; pp. 47–76. [Google Scholar] [CrossRef]
  26. Tangian, A. European flexicurity: Concepts, methodology and policies. Transf. Eur. Rev. Labour Res. 2007, 13, 551–573. [Google Scholar] [CrossRef]
  27. European Commission—EC. Report from the Commission to the European Parliament, the Council, the European Economic and Social Committee, and the Committee of the Regions on the Application of Directive 2008/104/EC on Temporary Agency work, COM(2014) 176 Final. Available online: http://ec.europa.eu/social/BlobServlet?docId=11459&langId=en (accessed on 29 October 2025).
  28. United Nations—UN. Transforming Our World: The 2030 Agenda for Sustainable Development. United Nations, Department of Economic and Social Affairs, New York, 2015. Available online: https://sdgs.un.org/2030agenda (accessed on 29 October 2025).
  29. Eizenberg, E.; Jabareen, Y. Social Sustainability: A New Conceptual Framework. Sustainability 2017, 9, 68. [Google Scholar] [CrossRef]
  30. Cajander, N.; Reiman, A.; Kärkkäinen, R.; Ylikarhi, K. Corporate Social Responsibility in Temporary Agency Work: A Study of Restaurant Work in Finland. Employ. Respons. Rights J. 2023, 35, 267–286. [Google Scholar] [CrossRef]
  31. Robinson, R.N.; Martins, A.; Solnet, D.; Baum, T. Sustaining precarity: Critically examining tourism and employment. J. Sustain. Tour. 2019, 27, 1008–1025. [Google Scholar] [CrossRef]
  32. Chambel, M.J.; Sobral, F. When temporary agency work is not so temporary. Econ. Ind. Democr. 2019, 40, 238–256. [Google Scholar] [CrossRef]
  33. Chambel, M.J.; Lopes, S.; Batista, J. The effects of temporary agency work contract transitions on well-being. Int. Arch. Occup. Environ. Health 2016, 89, 1215–1228. [Google Scholar] [CrossRef]
  34. Grimshaw, D.; Cartwright, J.; Keizer, A.; Rubery, J. Market Exposure and the Labour Process: The Contradictory Dynamics in Managing Subcontracted Services Work. Work Employ. Soc. 2019, 33, 76–95. [Google Scholar] [CrossRef]
  35. Creswell, J.W.; Plano Clark, V.L. Designing and Conducting Mixed Methods Research, 3rd ed.; Sage Publications, Inc.: Thousand Oaks, CA, USA, 2017. [Google Scholar]
  36. Yin, R.K. Case Study Research and Applications: Design and Methods, 6th ed.; Sage Publication, Inc.: Thousand Oaks, CA, USA, 2018; p. 319. [Google Scholar]
  37. Krapež, K. Regulation of temporary agency work and the modern labor market: A case study of Slovenia. Stanovnistvo 2024, 62, 127–151. [Google Scholar] [CrossRef]
  38. Farquhar, J.; Michels, N.; Robson, J. Triangulation in industrial qualitative case study research: Widening the scope. Ind. Mark. Manag. 2020, 87, 160–170. [Google Scholar] [CrossRef]
  39. Voudouris, I. The use of flexible employment arrangements: Some new evidence from Greek firms. Int. J. Hum. Resour. Manag. 2004, 15, 131–146. [Google Scholar] [CrossRef]
  40. Garen, J. Use of employees and alternative work arrangements in the United States: A law, economics, and organizations perspective. Labour Econ. 2006, 13, 107–141. [Google Scholar] [CrossRef]
  41. Abraham, K.G.; Taylor, S.K. Firms’ use of outside contractors: Theory and evidence. J. Labor Econ. 1996, 14, 394–424. [Google Scholar] [CrossRef]
  42. Houseman, S.N. Why employers use flexible staffing arrangements: Evidence from an establishment survey. ILR Rev. 2001, 55, 149–170. [Google Scholar] [CrossRef]
  43. Thommes, K.; Weiland, K. Explanatory factors for firms’ use of temporary agency work in Germany. Eur. Manag. J. 2010, 28, 55–67. [Google Scholar] [CrossRef]
  44. Whitley, R. Business Systems and Organizational Capabilities: The Institutional Structuring of Competitive Competences; Oxford University Press: Oxford, UK, 2007. [Google Scholar]
  45. Alewell, D.; Baehring, K.; Canis, A.; Hauff, S.; Thommes, K. Outsourcing HR Functions: Development of an Explanatory Approach to Firms’ (Non-Existent) Demand for Personnel Services. Int. Rev. Manag. Stud. 2007, 18, 271–292. [Google Scholar] [CrossRef]
  46. Allen, M.M.C.; Liu, J.; Allen, M.L.; Imran Saqib, S. Establishments’ use of temporary agency workers: The influence of institutions and establishments’ employment strategies. Int. J. Hum. Resour. Manag. 2017, 28, 2570–2593. [Google Scholar] [CrossRef]
  47. Stanworth, C.; Druker, J. Human resource solutions? Pers. Rev. 2006, 35, 175–190. [Google Scholar] [CrossRef]
  48. Coe, N.M.; Johns, J.; Ward, K. Flexibility in Action: The Temporary Staffing Industry in the Czech Republic and Poland. Environ. Plan. A 2008, 40, 1391–1415. [Google Scholar] [CrossRef]
  49. Holst, H.; Nachtwey, O.; Dorre, K. The Strategic Use of Temporary Agency Work: Functional Change of a Non-Standard Form of Employment. Int. J. Action Res. 2010, 6, 108–138. [Google Scholar]
  50. Enright, B. (Re)considering New Agents: A Review of Labour Market Intermediaries within Labour Geography. Geogr. Compass 2013, 7, 287–299. [Google Scholar] [CrossRef]
  51. Ferreira, J. The German temporary staffing industry: Growth, development, scandal and resistance. Ind. Relat. J. 2016, 47, 117–143. [Google Scholar] [CrossRef]
  52. Atkinson, J. Manpower Strategies for Flexible Organisation. Pers. Manag. 1984, 16, 28–31. [Google Scholar]
  53. Gramm, C.L.; Schnell, J.F. The use of flexible staffing arrangements in core production jobs. ILR Rev. 2001, 54, 245–258. [Google Scholar] [CrossRef]
  54. Kalleberg, A.L.; Reynolds, J.; Marsden, P.V. Externalizing employment: Flexible staffing arrangements in US organizations. Soc. Sci. Res. 2003, 32, 525–552. [Google Scholar] [CrossRef]
  55. OECD. The OECD Indicators of Employment Protection Legislation. Organisation for Economic Co-Operation and Development, Last Update February 2024. Available online: https://www.oecd.org/en/data/datasets/oecd-indicators-of-employment-protection.html (accessed on 29 October 2025).
  56. Eurostat. Temporary Employment Agency Workers by Sex, Age and NACE Rev. 2 Activity. 2023. Available online: https://ec.europa.eu/eurostat/databrowser/view/LFSA_QOE_4A6R2__custom_89620/bookmark/table?lang=en&bookmarkId=3b52ebd8-7b16-4e64-9613-83f8412d2ff1 (accessed on 29 October 2025).
  57. Statistical Office of the Republic of Slovenia—SiStat. Labour Force Survey Results, 2022. 2023. Available online: https://www.stat.si/StatWeb/en/News/Index/11047 (accessed on 27 July 2023).
  58. Kozjek, T.; Ferjan, M.; Žnidaršič, A. The Perception of Job Security—Empirical Evidence from Slovenia. J. East Eur. Manag. Stud. 2017, 22, 287–314. [Google Scholar] [CrossRef]
  59. Kozjek, T.; Ferjan, M. Organizational Flexibility, Employee Security, and Organizational Efficiency: A Case Study of Slovenian Public and Private Sector Organizations. Organizacija 2015, 48, 3–22. [Google Scholar] [CrossRef]
  60. Ministry of Labour, Family, Social Affairs, and Equal Opportunities. Register of Domestic Legal and Natural Persons for the Activity of Providing the Work of Workers to the User. 2024. Available online: https://podatki.gov.si/dataset/register-domacih-pravnih-in-fizicnih-oseb-za-zagotavljanje-dela-delavcev-uporabniku (accessed on 29 October 2025).
  61. Ministry of Labour, Family, Social Affairs, and Equal Opportunities. Records of Foreign Legal and Natural Persons for the Performance of the Activity of Providing the Work of Workers to the User. 2024. Available online: https://podatki.gov.si/dataset/evidenca-tujih-pravnih-in-fizicnih-oseb-za-opravljanje-dejavnosti-zagotavljanja-dela-delavcev-uporab# (accessed on 27 July 2023).
  62. Redek, T.; Bavdaž, M.; Cepec, J.; Domadenik, P.; Godnov, U.; Kostevc, Č. Vpliv Zasebnih Agencij za Zaposlovanje na trg Dela v Sloveniji (The Impact of Private Employment Agencies on the Labor Market in Slovenia); Zavod RS za Zaposlovanje (Employment Service of Slovenia—ESS): Ljubljana, Slovenia, 2017. Available online: https://www.ess.gov.si/fileadmin/user_upload/Trg_dela/Dokumenti_TD/Analize/Vpliv_zasebnih_agencij_za_zaposlovanje_na_trg_dela_v_Sloveniji.pdf (accessed on 29 October 2025).
  63. Labor Inspectorate of the Republic of Slovenia (Inšpektorat RS za delo). Poročilo o delu Inšpektorata RS za delo za leto 2020. GOVSI. 2020. Available online: https://www.gov.si/assets/organi-v-sestavi/IRSD/LETNA-POROCILA-IRSD/LP_IRSD_2020_www_skupno.pdf (accessed on 29 October 2025).
  64. Kuralt, T. Kritični Pogled na Opravljanje Dejavnosti Zagotavljanja dela Delavcev Uporabniku (A Critical View on the TAW Model). Podjetje Delo 2016, 3, 615–623. [Google Scholar]
  65. Kresal Šoltes, K. Razmejitev obveznosti med agencijo in podjetjem uporabnikom ter načelo enakega obravnavanja–je lahko model tudi za druge nestandardne oblike dela? Delav. Delodajalci 2017, 2, 199–220. [Google Scholar]
  66. Senčur-Peček, D. Izzivi delovnega prava v času prekarnega dela. Podjet. Delo 2019, 6–7, 1033. [Google Scholar]
  67. Senčur Peček, D. Navidezni samozaposleni in prikrita delovna razmerja (Ostensibly self-employed and concealed employment relationships). In Prekarno Delo: Multidisciplinarna Analiza; Kresal Šoltes, K., Strban, G., Domadenik, P., Eds.; Elektronska izdaja Zbirka Maksima: Ljubljana, Slovenia, 2020; pp. 46–66. Available online: https://institut-delo.com/DOC22/Prekarno_delo.pdf?LV=1 (accessed on 29 October 2025).
  68. Kresal, B. Prikrita delovna razmerja–nevarno izigravanje zakonodaje (Concealed Employment Relationships–Perilous Evasion of Legislation). Delav. Delodajalci 2014, 14, 177–200. [Google Scholar]
  69. Kresal Šoltes, K. Začasno agencijsko delo (Temporary agency work). In Prekarno Delo: Multidisciplinarna Analiza; Kresal Šoltes, K., Strban, G., Domadenik, P., Eds.; Elektronska izdaja Zbirka Maksima: Ljubljana, Slovenia, 2020; pp. 46–66. Available online: https://institut-delo.com/DOC22/Prekarno_delo.pdf?LV=1 (accessed on 29 October 2025).
  70. Breznik, M.; Čehovin Zajc, J. Prekarizacija standardnega in nestandardnega zaposlovanja v Sloveniji (2005–2019) (Precarization of Standard and Non-Standard Employment in Slovenia (2005–2019)). Teor. Praksa 2021, 58, 28–48. [Google Scholar] [CrossRef]
  71. Pustovrh Pirnat, T. Sodna praksa glede odgovornosti zaposlitvene agencije za škodo, ki jo utrpi delavec pri delu (Case law regarding the liability of the temporary employment agency for damage suffered by a posted worker at work). Delav. Delodajalci 2013, 1, 79. [Google Scholar]
  72. Palinkas, L.A.; Horwitz, S.M.; Green, C.A.; Wisdom, J.P.; Duan, N.; Hoagwood, K. Purposeful Sampling for Qualitative Data Collection and Analysis in Mixed Method Implementation Research. Adm. Policy Ment. Health Ment. Health Serv. Res. 2015, 42, 533–544. [Google Scholar] [CrossRef]
  73. Atkinson, R.; Flint, J. Accessing Hidden and Hard-to-Reach Populations: Snowball Research Strategies. Soc. Res. Updat. 2001, 33. [Google Scholar]
  74. Malterud, K.; Siersma, V.D.; Guassora, A.D. Sample Size in Qualitative Interview Studies: Guided by Information Power. Qual. Health Res. 2016, 26, 1753–1760. [Google Scholar] [CrossRef]
  75. Vasileiou, K.; Barnett, J.; Thorpe, S.; Young, T. Characterising and Justifying Sample Size Sufficiency in Interview-Based Studies: Systematic Analysis of Qualitative Health Research over a 15-Year Period. BMC Med. Res. Methodol. 2018, 18, 148. [Google Scholar] [CrossRef]
  76. OECD. The New OECD Employment Protection Legislation Indicators for Temporary Contracts. Organisation for Economic Co-Operation and Development, 2021. Available online: https://www.oecd.org/content/dam/oecd/en/data/datasets/indicators-of-employment-protection/OECD-EPLIndicators-TemporaryContracts.pdf (accessed on 29 October 2025).
  77. Labour Inspectorate of the Republic of Slovenia (Inšpektorat RS za delo RS). Poročilo o Delu Inšpektorata RS za Delo za Leto 2022. GOVSI. 2023. Available online: https://www.gov.si/assets/organi-v-sestavi/IRSD/LETNA-POROCILA-IRSD/Letno-porocilo-o-delu-IRSD-za-leto-2022.pdf (accessed on 29 October 2025).
  78. Hopkins, B. Explaining variations in absence rates: Temporary and agency workers in the food manufacturing sector. Hum. Resour. Manag. J. 2012, 24, 227–240. [Google Scholar] [CrossRef]
  79. García Mainar, I.; Green, C.P.; Navarro Paniagua, M. The Effect of Permanent Employment on Absenteeism: Evidence from Labor Reform in Spain. ILR Rev. 2018, 71, 525–549. [Google Scholar] [CrossRef]
  80. ETUI. Worker Participation in the EU: Slovenia. 2024. Available online: https://worker-participation.eu/national-industrial-relations/countries/slovenia (accessed on 29 October 2025).
Figure 1. Drivers behind the demand for TAW in Slovenia.
Figure 1. Drivers behind the demand for TAW in Slovenia.
Sustainability 17 11261 g001
Figure 2. Analytical pathway from demand-side drivers to SDG-8 assessment, gaps, and policy actions.
Figure 2. Analytical pathway from demand-side drivers to SDG-8 assessment, gaps, and policy actions.
Sustainability 17 11261 g002
Table 1. Sampling criteria.
Table 1. Sampling criteria.
CriterionDescription
GenderFemale, male
Experience with temporary agency workUp to 3 years, 3–10 years, more than 10 years
Type of stakeholderTemporary work agency, agency worker, client organization, trade union, administrative body
Size of temporary work agencyFour biggest agencies, other agencies (micro, small, medium-sized)
Membership of temporary work agency in an Employment Agencies AssociationNo, yes
Branch of client organizationBranches with the highest prevalence of TAW, other branches
Table 2. Respondent characteristics.
Table 2. Respondent characteristics.
CriterionDescription
GenderFemale (n = 7), male (n = 12)
Experience with temporary agency workUp to 3 years (n = 3), 3–10 years (n = 8,), more than 10 years (n = 8)
Type of stakeholderTemporary work agency—A* (n = 7), agency worker—W (n = 4), client organization—C (n = 5), administrative body/regulator—R (n = 1), trade union representative—T (n = 2)
Size of agency (n = 7)Four biggest agencies (n = 4) (A1–A4), other agencies (micro, small, medium-sized; n = 3) (A5-A7)
Membership of agency in The Employment Agencies Association (n = 7)Yes (n = 6) (A1–A6), no (n = 1) (A7)
Industry branch of client organizationManufacturing (n = 3) (C1, C3, C5), tourism (n = 1) (C2), construction (n = 1) (C4) (all are branches with the highest prevalence of TAW in Slovenia)
* Abbreviations, listed in the table, are used in the analysis in the following Section 3.
Table 3. Coded-intensity matrix of TAW demand-side drivers by stakeholder group.
Table 3. Coded-intensity matrix of TAW demand-side drivers by stakeholder group.
Drivers\StakeholdersAgenciesClient OrganizationsAgency WorkersTrade UnionsState Authority
D1 Medium-term fluctuations33333
D2 Seasonality21122
D3 Labour scarcity33333
D4 Substitution for probation12210
D5 Driven workforce23210
D6 Match quality & transitions22200
D7 Partial HRM outsourcing22110
Values reflect coded intensity from Phase-2 interviews; Scale: 0 = not observed, 1 = low, 2 = moderate, 3 = high. Categories per Supplementary File S2 (Codebook); interview guide and roster set per Supplementary File S1.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Krapež, K. Understanding the Drivers of Temporary Agency Work in Slovenia: Implications for Sustainable Labor Practices. Sustainability 2025, 17, 11261. https://doi.org/10.3390/su172411261

AMA Style

Krapež K. Understanding the Drivers of Temporary Agency Work in Slovenia: Implications for Sustainable Labor Practices. Sustainability. 2025; 17(24):11261. https://doi.org/10.3390/su172411261

Chicago/Turabian Style

Krapež, Katarina. 2025. "Understanding the Drivers of Temporary Agency Work in Slovenia: Implications for Sustainable Labor Practices" Sustainability 17, no. 24: 11261. https://doi.org/10.3390/su172411261

APA Style

Krapež, K. (2025). Understanding the Drivers of Temporary Agency Work in Slovenia: Implications for Sustainable Labor Practices. Sustainability, 17(24), 11261. https://doi.org/10.3390/su172411261

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop