Vertical and horizontal diffusion of labour standards in global supply chains: working hours practices of tier-1 and tier-2 textiles and garment suppliers

Abstract This paper explores the diffusion of working hours standards beyond tier-1 suppliers in the garment supply chain. It compares working hours designs and rationales of tier-1 and tier-2 garment suppliers in Jordan and garment and textiles suppliers in Egypt. Building on the idea of crossvergence, the paper extends previous insights on explanatory factors that are rooted in the supply chain and captures global, local, and firm-level rationales. The analysis followed a flexible pattern matching logic and draws on a qualitative multiple-case study encompassing 30 owner-manager interviews, 13 expert interviews, and multiple documents and field observations. Convergence to global standards was found to be higher in tier-1 compared with tier-2 garment firms and differed between the garment and the textiles industry. The study showed that labour standard demands of global buyers dissolve along the supply chain, while the price and lead time squeeze persists or increases. How suppliers navigate the antagonism of labour standard and procurement demands depends upon local level factors and firm resources. A framework of human, intellectual, physical, and financial resources was developed to capture structural differences between higher- and lower-tier suppliers. The findings underscore the need for new procurement strategies of global buyers and tier-1 suppliers, and further investments in multi-stakeholder initiatives that monitor sub-suppliers, support resource development, and address systemic barriers to labour standard compliance.


Introduction
Labour institutions and researchers find that working conditions in the global garment supply chain regularly breach international standards of decent work (Anner, 2019;White et al., 2017).The problem affects not only producers, workers, and their families but also their communities' chances for economic and social development (Barrientos et al., 2016).This article focuses on the design of working hours.Forced overtime and excessive hours are common in garment and textiles production (International Labour Organization, 2014).They not only negatively affect workers' performance but also compromise their health and well-being (Bannai & Tamakoshi, 2014) and in some cases have even led to suicide (Smyth et al., 2013).Most global labour standards prohibit forced overtime and excessive hours (ILO, 2014), and a growing body of literature discusses how they diffuse in global supply chains (Lund-Thomsen et al., 2016).
Research on labour standard compliance has especially focused on suppliers that directly export to global buyers (Gold et al., 2020).Buyers are, however, increasingly being held accountable for their entire supply chain, and in some countries this accountability is legally mandated (Amengual & Distelhorst, 2019;Villena, 2019;Wilhelm et al., 2016).Such requirements are challenging as garment supply chains are complex networks of contracting and subcontracting relationships (Alamgir & Banerjee, 2019;Grimm et al., 2016).These relationships can be differentiated based on whether they occur between vertical stages of the supply chain or between horizontal nodes within one stage (Lazzarini et al., 2001).Appelbaum and Gereffi (1994) differentiate between five vertical stages in the garment supply chain: (1) Raw materials are manufactured into (2) apparel components, especially textiles.These are (3) assembled into garments, (4) exported and (5) marketed and sold to the end consumer.Within each stage, subcontracting among producers creates different horizontal nodes.Especially in the stage of garment assembly, tier-1 suppliers who directly export to global buyers frequently subcontract a share of their production to lower-tier suppliers (Caro et al., 2020).Research is therefore called to discuss how labour standards diffuse beyond tier-1 suppliers (Gold et al., 2020;Villena & Gioia, 2018).This paper contributes to the nascent discourse on lower-tier suppliers and focuses on two horizontal nodes in the second stage of the supply chain, the assembly of garments, and one node in the third stage, the manufacturing of textiles.Production activities in these nodes and stages commonly take place in the Global South (Reinecke et al., 2019).
Previous research showed that the degree to which companies comply with global standards varies across the vertical stages and horizontal nodes (Gold et al., 2020).Horizontal, especially hidden, subcontracting commonly has a negative impact on working conditions (Gold et al., 2020;Nadvi & Raj-Reichert, 2015).Research on the vertical diffusion of standards yields mixed results.Vertical subcontracting is frequently associated with an increase of labour risks (Lebaron, 2014;Posthuma, 2010), but this assumption could not be supported empirically in the garment supply chain (Grace Annapoorani, 2017).The non-diffusion of global labour standards beyond tier-1 suppliers is explained in particular by arguments related to the dynamics in global supply chains (e.g.Gold et al., 2020;Grimm et al., 2016;Lebaron, 2014).Increased labour risks in lower nodes and stages of the chain are linked to the limited reach of global buyers' private regulatory efforts (Lebaron, 2014;Posthuma, 2010).Critical reflections on contemporary business models furthermore highlight that procurement requirements of global buyers counteract their demands for labour standard compliance (Jamali et al., 2017;Reinecke & Donaghey, 2021).The price and lead time squeeze present in many global supply chains, particularly the garment supply chain, increases labour risks in lower nodes and stages of the chain (Anner, 2019;Lebaron, 2014).Global supply chain-centred arguments explain why the diffusion of labour standards generally decreases beyond tier-1 suppliers, but they fall short of explaining exceptions to this rule.Textile suppliers were not found to operate with worse labour standards than garment suppliers (Cooke & He, 2010;Grace Annapoorani, 2017).Little research exists on factors outside the supply chain that influence labour standard adoption of lower-tier suppliers.The impact of supplier resources (Chand & Tarei, 2021) and labour skill level (Barrientos et al., 2011;Gereffi & Fernandez-Stark, 2016) has been addressed but awaits to be further specified.It is known from research on tier-1 suppliers that a multitude of global, local, and firm-level factors impact labour standard adoption (Holzberg, 2019;Lund-Thomsen et al., 2016;Toffel et al., 2015).The existence of additional factors that impact the diffusion of labour standards beyond tier-1 suppliers is therefore likely.A more comprehensive overview of influencing factors is needed to avoid misguided research conclusions and labour interventions.Practical interventions that address only one set of influences on labour standard adoption risk exacerbating other influencing factors when they remain unknown (Chand & Tarei, 2021).
This study centres on the perspective of suppliers to gain a comprehensive overview of the factors that promote or impair the diffusion of labour standards beyond tier-1 suppliers.Supplier-centred research allows to identify local and firm-level drivers of labour standard adoption in addition to global ones (Fontana & Egels-Zandén, 2019;Soundararajan et al., 2018) and to evaluate how influencing factors are evaluated and prioritised by decision-makers (Perry et al., 2015).For suppliers in the garment supply chain, decisions on working hours are usually considered to be HR and/or compliance decisions.The present study therefore bridges supply chain-focused research insights on the compliance rationale with IHRM scholarship.It is a core strength of IHRM research to compare HR practices across contexts and to link variations in international standard diffusion to global, local, and firm-level influences.Different frameworks have been developed to structure the antecedents of international HRM, some of them explicitly referring to the context of the Global South (e.g.Budhwar et al., 2016;Holzberg, 2019).The concept of crossvergence was found to be particularly suitable for the present study as it covers the consolidation of global and local influences on HR practices in the international arena (Al Ariss & Sidani, 2016a), theoretically points to the importance of firm-level factors to moderate standard adoption (Fan et al., 2016), and has previously been linked to the debate on labour standard compliance in global supply chains (Holzberg, 2019).It is therefore to used theoretically frame this study as will be further outlined in Sec.2.3.
The empirical investigation compares labour standard adoption across vertical stages and horizontal nodes of the garment supply chain and traces how the impact of global, local, and firm-level factors influence the diffusion of global standards beyond tier-1 suppliers.A comparative design was chosen because it enabled to identify which factors change in lower stages and nodes of the garment supply chain and to discuss the results in reference to the well-researched case of tier-1 suppliers.A multiple-case study approach was used to investigate the design of working hours in four research cases: direct garment exporters and subcontracting garment producers in Jordan (horizontal differentiation) and garment exporters and textiles producers in Egypt (vertical differentiation).While Egypt and Jordan have similar cultural backgrounds, they differ in their economic internationalisation, especially in the garment supply chain (Azmeh, 2014).Studying both countries helped to disentangle global and local influences on suppliers and to explain (the limits of) global standard diffusion.The focus on the Middle East and Northern Africa answers calls for more HRM research in the region (Al Ariss & Sidani, 2016a;Budhwar et al., 2019).
The study answers following research question and sub-questions: How do global working hours standards diffuse beyond tier-1 suppliers in the garment supply chain?
• How does global labour standard adoption compare across different horizontal nodes of the supply chain in Jordan (tier-1 vs. tier-2 garment production)?Which factors drive differences in labour standard adoption between tier-1 and tier-2 garment suppliers?
• How does global labour standard adoption compare across different vertical stages of the supply chain in Egypt (garment vs. textiles production)?Which factors drive differences in labour standard adoption between garment and textiles suppliers?

Theoretical background
This section (1) operationalises the concept of global labour standards, (2) reviews existing insights on the diffusion of labour standards in different nodes and stages of global supply chains, and (3) elaborates the theoretical (crossvergence) framework of this study.

Global labour standards
Global labour standards establish minimum requirements for decent employment worldwide but are not distinct.Various private and public standards exist in parallel and, for some labour issues, additionally refer to national legislation.Public labour standards are promoted by international organisations and inform national legislation and international labour initiatives (Holzberg, 2019).Global buyers, often in negotiation with civil society organisations and/or global unions, additionally define private standards and request suppliers to comply with them (Nadvi & Wältring, 2002).
For the current study, global standards were operationalised based on the regulations relevant for the research contexts.Desk research and interviews indicated that especially five private standards were requested from suppliers: Business Social Compliance Initiative (amfori BSCI, 2018), Ethical Trading Initiative (ETI, 2014), Fair Labor Association (FLA, 2011), Social Accountability 8000 (Social Accountability International, 2014), and/or Sedex (2017).Furthermore, some suppliers were subject to buyer-specific codes of conduct.Private standards directly affect business operations as buyers can blacklist non-compliant suppliers.Regarding global public standards, several ILO conventions (International Labour Organization (ILO), 1919, 1921, 1935, 1962) must be considered because they inform the work of international organisations operating in the research context.The ILO regulates regular and overtime hours but refers to national legislation for the prohibition of excessive hours and break times.
Working hours regulations in these standards are highly convergent (see Annex A): a maximum of 48 regular working hours are allowed per week.Overtime must be voluntary, temporary, and paid with a premium.Most private standards prohibit excessive hours, defined as hours exceeding 60 h/w, while only Egypt prohibits excessive hours by law.Breaks are mandatory, and Egypt requires at least 60 min/d.The study therefore works with one operationalisation of global working hours standards (see Table 1) and only distinguishes between the different sets of standards when the included working hours regulations diverge.

Diffusion of global labour standards in different nodes and stages of the supply chain
This study compares labour standard adoption across horizontal nodes and vertical stages of the garment supply chain and identifies factors that drive the (non-)diffusion of global standards beyond tier-1 suppliers.This sub-section reviews previous research insights on differences in labour standard adoption between tier-1 and lower-tier suppliers and factors that drive these differences.
For horizontal subcontracting, previous research has shown that compliance levels decrease in lower horizontal nodes of the supply chain (Alamgir & Banerjee, 2019;Anner, 2019;Nadvi & Raj-Reichert, 2015).In some cases, it is even a strategy of tier-1 suppliers to horizontally subcontract to keep labour standards low (Clean Clothes Campaign, 2005).Regarding vertical subcontracting, the empirical evidence is inconsistent.An increase of labour risks with vertical subcontracting has been demonstrated in the automotive, electronics and pharmaceutical supply chains (Villena & Gioia, 2018) but not in the garment supply chain.It could only be established, that the treatment of labour issues differs between the textiles and garment stage of production without characterizing one industry as being more or less compliant (Cooke & He, 2010;Grace Annapoorani, 2017).
Differences in labour standard adoption between tier-1 suppliers and their horizontal and vertical subcontractors are especially linked to contemporary designs of global supply chain regulation.Private labour regulation mainly reaches tier-1 suppliers as buyers have more power over them than over lower-tier suppliers (Gold et al., 2020;Grimm et al., 2016;Huq et al., 2014).Horizontal subcontracting is often hidden from buyers (Caro et al., 2020) and the complexity of the garment supply chain makes it highly challenging for global buyers to manage sustainability issues beyond their tier-1 suppliers (Grimm et al., 2016).In consequence, lower-tier suppliers are rarely audited (Posthuma, 2010) and experience less pressure from global brands to design their HR practices according to global standards.
This line of argumentation is shared by Villena (2019) and Anner (2019) who however add that the improvement of labour standards of lower-tier suppliers needs not only more effective global labour regulation but requires a fundamental change of procurement practices in global supply chains.The pressure that global buyers exert on their suppliers can be characterised by a 'structured antagonism' (Reinecke & Donaghey, 2021, p. 457) of interests.Typical business models of global garment brands feature price competitiveness and procurement flexibility and therewith put downward pressure on working conditions in the supply chain (Jamali et al., 2017;Reinecke et al., 2019).At the same time, social responsibility efforts of these brands invest in ensuring compliance with global labour standards in the supply chain (Amengual & Distelhorst, 2019).This inherent contradiction motivates incompliance with global labour standards (Jamali et al., 2017;Khan & Lund-Thomsen, 2011).Private regulatory efforts were consequently shown to be most successful if buyers' sourcing practices are aligned with their compliance demands (Amengual et al., 2020;Amengual & Distelhorst, 2020).While Villena (2019) and Anner (2019) have introduced this debate into research on lower-tier suppliers, it remains to be evaluated in how far buyer procurement pressure increases, decreases, or remains constant in lower horizontal nodes and vertical stages of the supply chain, and which factors moderate this effect.
While supply chain dynamics-as presented above and summarised in Figure 1-can explain that labour standards deteriorate in lower horizontal nodes of the supply chain, they fall short of explaining the inconsistent empirical evidence on the vertical diffusion of labour standards.Structural differences between the textiles and the garment industries seem to interact with differences in supply chain power relationships and lead to inconsistent employment outcomes (Grace Annapoorani, 2017).Which structural differences between tier-1 and lower-tier suppliers lead to differences in labour standard adoption is thereby largely unknown (Villena & Gioia, 2018).Previous research indicates that the skill level of workers (Barrientos et al., 2011;Gereffi & Fernandez-Stark, 2016) and the resources of suppliers (Chand & Tarei, 2021;Villena & Gioia, 2018) need to be considered.Low skilled, labour-intensive work, characteristic of the garment industry, is associated with a low quality of work and irregular working hours.Moderate skilled, machine-intensive work, as is common in the textiles industry, is connected to relatively better working conditions but long working hours when there is a shortage of skilled workers in the labour market (Barrientos et al., 2011;Gereffi & Fernandez-Stark, 2016).Supplier resources and capabilities furthermore differ between tier-1 and lower-tier suppliers (Chand & Tarei, 2021;Villena & Gioia, 2018) and it remains to be clarified which resources favour or hinder labour standard adoption.The list of further factors that might impact the vertical and horizontal diffusion of labour standards is long.A multitude of local-and firm-level factors influences the design of employment practices in global supply chains (Barrientos et al., 2011;Gereffi & Lee, 2016;Holzberg, 2019;Lund-Thomsen et al., 2016).Their impact on the vertical and horizontal diffusion of labour standards remains to be evaluated.

Theoretical framework of the study
Insights from the convergence-divergence-crossvergence debate frame this study theoretically.This sub-section introduces the concept of crossvergence, applies it to the context of global supply chains, and highlights its relevance for research on lower-tier suppliers.
IHRM scholars have debated for decades whether HR practices converge to international HR standards or whether context-specific divergence prevails.Globalisation and the worldwide diffusion of 'best practices' drive international convergence of HR practices (e.g.Kerr et al., 1960;Levitt, 1983;Prentice, 1990).Final convergence is however seldom achieved (Brewster et al., 2016).In the Middle East, HR systems for instance have integrated global elements but nevertheless remained distinctively different from Western ones (Budhwar et al., 2019).Divergence from global standards is driven by cross-country differences in national institutions and local culture (Jackson, 2015;Reiche et al., 2012;Wood et al., 2012;Wood & Horwitz, 2015).An increasing consensus suggests that convergence and divergence trends exist in parallel (e.g.Al Ariss & Sidani, 2016b;Paik et al., 2011).Al Ariss and Sidani (2016a) link this finding to Ralston's (2008) concept of crossvergence: Managers are influenced by global dynamics as well as their local environment and develop distinctive HR systems in response.Contemporary international business strategies are thus often 'glocal' , containing global, local and distinct/hybrid elements (Svensson, 2001) and it is necessary to examine which factors influence the relative importance of global and local influences (Al Ariss & Sidani, 2016a).In this context, factors in the company's micro-environment are discussed to explain an organisation's response to global and local influencing factors (Fan et al., 2016).
Research on labour standard adoption of tier-1 suppliers has come to similar conclusions.While the enforcement of global labour standards drives convergence of HR practices across tier-1 suppliers, labour standards are seldom fully complied with (Berliner et al., 2015).Local influences, including local institutions and culture, interact with demands for global labour standards, resulting in different levels of compliance (Lund-Thomsen et al., 2016;Toffel et al., 2015).Specific to the supply chain context is thereby that not only local factors drive divergence from labour standard adoption.As discussed in Subsec.2.2, global procurement demands counteract global regulatory efforts and therewith drive divergence from global labour standards (Anner, 2019;Reinecke & Donaghey, 2021).To describe this complex web of influences, Holzberg (2019) recurred to the concept of crossvergence.She differentiates between global influences (competition, public and private labour standards) and local influences (institutions, culture, industry) on suppliers and assumes their impact to depend upon the micro-environment of the firm.
The concept of crossvergence describes the formation of HR practices in an international context, including the design of working hours, and highlights that global and local factors simultaneously influence the adoption of global standards (Al Ariss & Sidani, 2016a;Holzberg, 2019).Their relative impact is moderated by firm-level factors (Fan et al., 2016).The concept can thus account for supply chain-related reasons that drive lower-tier suppliers to (not) adopt global labour standards but is open to further explanations.Knowledge on the reasons why labour standard compliance differs across horizontal and vertical linkages in supply chains is incomplete (Caro et al., 2020;Grimm et al., 2016).It has, however, been shown that the influence of global buyers decreases in lower nodes and stages of supply chains (Gold et al., 2020;Huq et al., 2014).Hence, tracing the process of crossvergence and identifying local and firm-level factors that affect the adoption of global labour standards is particularly important for the debate on lower-tier suppliers.

Research context: the garment supply chain in Jordan and Egypt
The garment supply chain is a prime example of a highly globalised buyer-driven production network (Fernandez-Stark et al., 2011).The lead firm, commonly a MNE from Northern America or Europe, is responsible for design, marketing and sales and has outsourced most of the manufacturing process (Gereffi & Memedovic, 2003).The collection of raw materials, the production of textiles and other apparel components, and the assembly of garments is commonly done in the Global South (Gereffi & Memedovic, 2003;Reinecke et al., 2019).This set up has traditionally been associated with the captive mode of the value chain governance typology of Gereffi et al. (2005, p. 84) (Anner, 2019;Gereffi et al., 2005), meaning that buyers have a high level of control over their suppliers.The rise of large multi-national tier-1 suppliers that can coordinate all stages of the manufacturing process has partially motivated a shift to relational governance models with lower levels of power asymmetry between buyers and suppliers (Azmeh & Nadvi, 2014;Gereffi et al., 2005).Anner (2019) however does not find this mode of governance yet to be dominating buyer-supplier relationships in the global garment supply chain.Captive governance modes dominate not only the relationship between buyers and their smaller tier-1 suppliers but also the relationship between tier-1 suppliers and their horizontal subcontractors.Especially in the stage of garment assembly, horizontal subcontracting is common and often hidden from buyers (Caro et al., 2020).
The exporting garment industry in Jordan encompasses solely the stage of garment assembly.About 50 direct exporters, 15 satellite factories (sub-units of direct exporters near residential areas), and 15 subcontractors operate in Jordan (Better Work Jordan, 2019;data: 2018) and export mainly to the United States.The context therefore allows to study the impact of horizontal subcontracting on global supply chain dynamics.A distinguishing feature of the industry in Jordan is its strong international linkages.Most companies are owned by South or Southeast Asian multinationals that have opened factories in Jordan to benefit from Jordan's favourable trade agreements with the United States (Azmeh & Nadvi, 2014).Their relationship with global buyers is therefore more relational than buyer-supplier relationships in other contexts (Azmeh & Nadvi, 2014).Asian migrants, especially from Bangladesh, India, Sri Lanka, and Nepal, compose 75% of the workforce (BWJ, 2019;data: 2018).Jordanian Labour Law requires a 25% share of Jordanian workers, which is supported by the creation of satellite factories.Following a number of labour scandals that began in 2006, working conditions have been closely monitored by international observers (Kolben, 2013).Better Work Jordan (BWJ), a program initiated by the ILO and the International Finance Corporation, started working with suppliers to improve labour standard compliance in 2009.Today, international observers report significant improvements and comparably favourable working conditions (Aissi et al., 2018;Robertson, 2019).
The garment supply chain in Egypt is vertically integrated and covers the entire production process from cotton cultivation to the manufacture of yarns, textiles, and garments (Said & Abdel-Hamid, 2018).It is therefore highly suitable for studying the vertical diffusion of global labour standards.The focus of this paper is on the production of textiles and garments, which are mainly exported to the United States and Europe, plus Turkey.Textiles manufacturing is highly capital intensive and depends on technological investments in machinery, while clothing manufacturing is predominantly labour intensive with rather inexpensive technical equipment (Fukunishi et al., 2013).A distinguishing feature of the industry is a long-standing historical tradition.Family-owned enterprises operate in Egypt and suppliers predominantly trade in captive value chains.Their employees are mainly young Egyptians with little education.In garment production, the share of female workers is high by local standards.In textiles factories, they are a minority.Although working conditions in factories are formally documented as comparably favourable (Azmeh, 2014;Marello et al., 2009), reports by local media and nongovernmental organisations about breaches of labour standards increasingly paint a more complicated picture (FLA, 2016).

Research design
The paper draws on a qualitative multiple-case study conducted in the textiles and garment industries in Jordan and Egypt.The study design was chosen as it enables an in-depth analysis of actor decision-making in their complex real-life environment (Reinecke et al., 2016;Yin, 2009).The within-case and cross-case analysis followed a pattern-matching logic (Trochim, 1989).The technique helped to guide the use of theory in the analysis and to structure it.Data were collected to derive case-specific patterns of working hours designs.Empirically observed patterns were compared with theoretically indicated patterns (derived from the literature) to assess the degree of standard adoption and to structure the presented rationales for working hours designs.Subsequently, the patterns were compared across the research cases to assess differences across the supply chain.

Data collection
The main data used for the analysis consisted of semi-structured interviews with owners, HR, Compliance and/or General Managers of supplier firms ('owners-managers') who were involved in the decisions on working hours designs in their organisation.A maximum variation logic guided the sampling of supplier firms and respondents.The method is designed to discover common themes across a wide variety of respondents that reflect the core and shared experiences of the research population (Patton, 2002).A supplier sampling frame and a respondent sampling frame were created with the help of desk research and pre-interviews with local industry and labour experts, which took place before the main field research started.Both sampling frames included criteria that previous research and local experts highlighted to influence the adoption of global labour standards.The selection of supplier factories was based on the size (number of employees), the location (different industrial zones and residential areas), the ownership (local/foreign, family/corporate), and the export focus (Europe, United States, other) of the factory.Respondents covered different genders, age groups, and nationalities.Industry experts and contacts helped facilitate access to interviewees based on the criteria included in the sampling frames.
In total, 30 owner-manager interviews (see Annex B for a list of respondents and Table 2 for an overview of interviews), conducted between September to October 2017 (Egypt) and April to May 2018 (Jordan), were analysed.Most respondent and supplier characteristics included in the sampling frames could be covered provided they existed in the research contexts.Fewer interviews were needed to cover the case of tier-1 garment suppliers in Jordan than originally anticipated as the industry was highly consolidated, supplier characteristics were similar, and interview responses were very homogeneous.Covering the case of tier-2 suppliers was a challenging task.Three Jordanian tier-2 suppliers participated in the study.Other tier-2 suppliers could not be identified or repeatedly postponed interviews before cancelling them.
Expert and document data served to fill (most of) the data gap (see Sec. 3.5).
During the interviews, respondents described their working hours and reflected on changes over time and the reasons for implementing certain practices.If respondents mentioned only selected aspects, they were prompted to provide more information.If firms had both textiles and garment operations ('integrated firms'), the operations were discussed separately.The interviews were conducted in English or German, in Egypt with Arabic translation.Interviews lasted 60-90 minutes, with working hours being only one part of the interview.Most interviews were audio-recorded and transcribed.If respondents did not give permission for recording, their interviews were documented in detail.To validate the credibility (Lincoln & Guba, 1985) of the translation, the transcription was done by native speakers who not only transcribed the English or German contents but also translated the Arabic questions and answers.All transcripts were subsequently double-checked by the author.The translators deviated during the interviews from the wording of the interviewer but accurately conveyed their meaning.To protect respondents' anonymity, each interview was assigned an interview code, specifying the context (Egypt or Jordan) and type of interview (Firm, Expert, or Worker).For example, the fifth interview in Jordan, which was the second owner-manager interview, was labelled J05F02.Saturation was achieved for all research cases except the tier-2 garment suppliers in Jordan, meaning that the last interviews conducted in tier-1 garment and textiles factories did not provide substantially more information (Bauer & Aarts, 2000).

Data analysis
The data analysis followed a pattern-matching logic (Trochim, 1989) and was supported by the qualitative research software MAXQDA.
First, theoretically indicated and empirically observed patterns were deduced, with global labour standards serving as theoretical reference for working hours designs.Respondents' rationales for working hours were expected to follow a crossvergence logic (Holzberg, 2019), including references to external global and local influences and the internal firm environment.All influencing factors discussed in Subsec.2.2 were integrated in this framework.A content analysis, guided by Schreier (2014), helped to identify empirically observed patterns.The main categories were structured according to the logic of the theoretical patterns, and sub-categories were added using a subsumption strategy (Schreier, 2014).This process ensured comparability with the theoretical patterns and accounted for all information within the data.All data were double-coded at minimum to ensure coding consistency.In the Egyptian cases, more coding rounds were frequently necessary as interview responses showed higher ambiguity, particularly due to language challenges, fuzzy working time policies, and a social desirability effect.
The content analysis showed that two cases were not homogeneous within themselves.In the Jordanian tier-1 garment industry, respondents clearly distinguished between migrant and Jordanian workers.In the Egyptian textiles industry, the interviewed firms worked either with a 2-shift or a 3-shift system and working hours differed accordingly.Both cases were divided into two sub-groups.
In the second step, theoretical and empirical patterns were matched for each research case, using a version of pattern matching that Sinkovics (2017) labelled 'flexible pattern matching' .The initial theoretical patterns were compared with the empirically observed patterns and gradually refined until they explained the empirical findings.Flexible pattern matching accommodates the circular nature of qualitative research and allows combining and exploring different theoretical approaches.In the third step of the analysis, the results were compared across the research cases by matching patterns of tier-1 and tier-2 garment exporters in Jordan (horizontal comparison) and patterns of garment and textiles exporters in Egypt (vertical comparison).
The pattern-matching procedure confirmed the structuring of working hours designs according to global labour standards.To assess the level of convergence to global standards, it was analysed whether observed working hours fully matched, breached, or even exceeded global standards.Furthermore, reported changes over time were considered and translated into a trend of increasing, decreasing, fluctuating or unchanging hours.Initial theoretical patterns of working hours rationales were significantly refined during the analysis.Although arguments referring to external global and local influences could be structured following the logic of Holzberg (2019), Osterwalder and Pigneur's (2010) differentiation of human, intellectual, physical, and financial resources better covered arguments pertinent to the micro-context of suppliers.Furthermore, an additional category was used to include references to cross-level initiatives promoting labour standards in Egypt and Jordan.Annex E presents the final pattern-matching frame.

Data triangulation
Due to the high sensitivity of the research topic, the research frame anticipated the need for data triangulation.Other perspectives on the research issue were examined (Flick, 2020) in order to validate the credibility of interview responses (Lincoln & Guba, 1985) and to enrich the information gathered during owner-manager interviews (Flick, 2020).
Thirteen key informant interviews with industry and labour experts (see Annex C), eight (group) interviews with workers (see Annex D), field notes documenting observations during factory visits, and the analysis of multiple documents helped triangulate the data.The interviewed experts represented local and international, industry and labour organisations operating in the research contexts and were known in the local industries for their expertise on work-related issues in supplier firms.Expert interviews helped not only to evaluate the credibility of interview responses but also served to evaluate the generalisability of the collected data.Documents, observations, expert, and worker interviews predominantly confirmed the information provided by owners-managers of supplier firms.In Jordan, interviewed experts reported the same working hours designs as owners-managers, highlighted similar rationales for implementing these designs, and considered the interview findings common for the industry.The Annual Report of BWJ (2019; data: 2018) furthermore confirmed the trends that were discovered in the comparison of tier-1 and tier-2 suppliers' working hours.Workers however reported exceptional incidents of tweaked overtime records and cases in which they had to fulfil their production targets although regular overtime was already used.The additional hours were not recorded.In Egypt, there are no statistics that explicitly refer to the export industry.Instead, statistics combine firms producing for local and for export markets.The data nevertheless confirmed the direction of working hours differences between the textiles and the garment industry (CAPMAS, 2018;data: 2017).Experts added that the reported working hours designs, and their justifications represented the range of designs that can be found in the export industries.Industry and labour experts, however, differed in their opinion which of the detected working hours are typical for the industry with industry experts painting a more socially desirable image of the industry than labour experts or owners-managers interviewed in the study.

Findings
Interviewed owners-managers described how they organised working hours (see Annex F for a complete overview of designs) and why they organised working time in this way.The analysis confirmed that global supply chain-related influences (global procurement practices, global labour standard demands) affect working hours designs and identified other influencing factors in the local and micro-level environment of suppliers.The micro-level influencing factors could be grouped according to a resource logic proposed by Osterwalder and Pigneur's (2010).The resource logic was further refined in the analysis and a resource framework was created to capture firm-level influences on the adoption of global working hours standards.The framework is presented in Subsec.4.1.Subsecs.4.2 to 4.5 present the findings of the within-case analysis to illustrate how global, local, and firm-level factors interact and lead to case-specific working hours designs.Figure 2 (adoption of global working hours standards) and Table 4 (working hours rationales) synthesise the descriptions and compare findings across cases.The findings of the cross-case comparisons are elaborated in Sec.4.6.

Capturing the micro-context of suppliers with a firm resource framework
Micro-level influences on the adoption of global working hours standards could be grouped according to Osterwalder and Pigneur's (2010) differentiation of human, intellectual, physical, and financial resources.Resources determine the degree to which a company pursues social and environmental objectives (Osterwalder & Pigneur, 2011) and accordingly influenced the adoption of global working hours standards in this study.The specific resource characteristics that promote or hinder global labour standard adoption of suppliers have, however, not yet been identified, and were therefore identified within the study.Table 3 presents an overview of human, intellectual, physical, and financial resources that impacted working hours designs of suppliers.Their specific impact on global standard adoption is described in the following description of within-case findings in order to account for interaction effects between global, local, and resource influences.

Case 1: working hours in the garment industry in Jordan (direct exporters)
Working hours converged to global labour standards in Jordanian tier-1 garment firms.Jordanians commonly worked 48 h/w and overtime only in exceptional cases.Their willingness to work overtime thereby increased, approaching the norm of global standards.Migrants commonly worked the maximum number of hours allowed by global standards (60 h/w).Working hours had declined over time to reach standard compliance, while the permanence and implicitness of migrant overtime still violated the criterion of temporariness and made the criterion of voluntariness of overtime a mere formality.However, a trend towards convergence was visible.
Respondents' rationales for working hours designs included references to global influences and firm resources.Buyers demanded keeping working hours below 60 h/w and monitored compliance with the help of BWJ and their own auditors.Buyer demands were characterized as 'very strict' [J18F12].Although excessive overtime was prohibited, respondents underlined that regular overtime was needed to meet buyers' procurement demands.
No factory in Jordan can go above break-even by working eight hours.It is a very plain equation, we do a lot of mathematics […] the bottom line is, if you have eight hours, you will not make money.[J18F12]  Overtime regulations were portrayed as being welcomed by migrants: '[Migrants] come here to earn a living.They do not have families.For them being in the job helps '. [J19F13].However, the needs of Jordanian workers conflicted with buyers' procurement pressures: 'Jordanians cannot work […] more than eight hours, especially the ladies.They have a family; they have their kids '. [J09F06].As Jordanians were essential human resources for tier-1 firms, owners-managers had to honour their demands.Jordanians had to be hired by law, were difficult to attract and retain, and were well-connected to local authorities and communities.
The interviews clearly revealed that suppliers faced great challenges in balancing global procurement, global labour standards, and human resource demands.Intellectual resources seemed to make the difference.Interviewed firms had developed highly professional HR and planning capabilities and operated with efficient production lines.BWJ, a local initiative with global roots, supported this progress.
We have good, experienced people, experts [in Jordan], […] and we are ready to learn.We [have learned for many years how to improve our business and HR and the] experts continue with us; we will learn more from them.[J19F13]

Case 2: working hours in the garment industry in Jordan (subcontractors)
Tier-2 garment firms commonly operated with 66-72 h/w and thus breached global labour standards.Although hours had declined over time, it remained doubtful whether working hours would eventually converge to global standards.Respondents characterised tier-2 garment business as being not profitable when working hours complied with standards.
Respondents portrayed working hours designs mainly as a response to global influences, specifically the intense procurement pressure in the industry which was passed on to them by tier-1 suppliers.Respondents acknowledged that private labour standards prohibited excessive hours, but tier-2 suppliers did not face significant repercussions for breaching this rule.J12F08 said that his goal of becoming a tier-1 supplier was his only incentive to comply with global standards.
While procurement pressure was the dominating theme across interviews, J06F03's design of working hours was additionally driven by human resource considerations.
[Workers] should not have very long hours, but they should have overtime so that they would benefit because they came from their countries to here to benefit.The other tier-2 suppliers described workers as wanting excessive hours.For example, J13F09 recounted that 15 workers from a direct exporter switched to his company because they wanted to work excessive hours to maximise their income.
Across tier-2 firms, intellectual, physical, and financial resources appeared less developed than in tier-1 firms.Only one respondent mentioned (self-)training in HR and production planning, the factories looked less organised than tier-1 factories, and respondents complained about scarce financial resources.BWJ had only recently started working with tier-2 firms and their consultants were not (yet) well-accepted by owners-managers.

Case 3: working hours in the garment industry in Egypt
The Egyptian garment industry had more within-case variation than the other cases.Egyptians commonly worked 48-54 h/w, in compliance with global standards.Observed regular hours (7-8 h/d) were often even shorter than standard regular hours.However, it seemed questionable whether overtime was always voluntary because respondents linked it to production needs rather than worker will.Despite the high compliance levels, the case cannot be labelled a good example of convergence.Working hours either fluctuated or did not change over time and were driven by local rather than global influences.
Rationales for working hours designs mainly included local and resource arguments.Respondents especially referred to local laws as the determining factor for working hours.Some respondents added that, when offered, overtime opportunities corresponded to workers' desire to increase their income [E01F01, E12F09, E27F22, E28F23].Permanent or excessive overtime did not seem possible though as firms predominantly employed local Egyptians with family and social obligations.E19F14 highlighted: 'A lot of the workforce are ladies, and they cannot work in night shifts, so we have just one 8 hours with 1 hour rest'.Owners-managers felt they needed to meet workers' demands because recruiting and retaining skilled labour were difficult.E27F22 explained that a good work environment had to be provided to compensate for meagre salaries.
Even though legal and human resource considerations dominated the interviews, half of the respondents also referred to buyer power and its impact on working hours.They mostly linked global procurement demands, which were said to be non-negotiable, with the need for longer hours: 'I do not like for the workers at all to stay after 5, I would love to [give them more free time], [but] what shall I do?It [lead time] is not enough ' [E21F16].In this context, intellectual, physical, and financial resources were described as not being well developed.E11F08 highlighted that only a few firms operated with professional HR departments and sophisticated planning and work processes that might help to reduce overtime.Other firms struggled with lack of space and unreliable delivery of raw materials.E19F14 added that after the Egyptian revolution and the devaluation of the Egyptian Pound, financial resources were tight and investing in better HR and working conditions was impossible.Global and local initiatives supporting resource development were scarce and worked only with selected suppliers in Egypt.
Global (private) labour standards were commonly characterized as impractical.Respondents complained that the standards conflicted with buyers' procurement demands, production realities in Egypt, or suppliers' own quality of work understanding.
The exactness, the adherence to specifications […] is not the same in the Egyptian context [it conflicts with Egyptian culture], on the one hand you are dealing with a supply chain that is not […] up to standard, and on the other hand you are dealing with customers [global buyers] that would not tolerate any deviations in whatever form, so it is a big challenge.[E19F14]

Case 4: working hours in the textiles industry in Egypt
Textiles firms operating with a 3-shift system (8-hour shifts, including breaks) commonly exceeded global standards.Observed regular hours, excluding break time, were shorter than standard hours.Only a few (exceptional) incidents were reported, in which workers worked two consecutive shifts, in breach of global standards.Working hours in firms operating with 2-shift systems (12-hour shifts, including breaks) continuously breached global standards and convergence was not apparent.Most respondents reported no changes in working hours over time.Some indicated that they alternated between 2-and 3-shift systems depending on the order situation and worker availability.
When asked about rationales for working hours designs, respondents explained that their main physical resources, the textiles machines, required 24 h/d operations.Owners-managers operating with 3-shift systems added that the local law required 8-hour workdays.2-shift systems were linked to the scarcity of qualified and reliable human resources and workers' wishes to maximise their income.Notably, the textiles workforce was almost exclusively male and regular wages were below living wages.I wanted to have three shifts.Immediately, we had resistance […] from the workers, why? [Workers] preferred to work the 12 hours shift, get the 4 hours [overtime pay] every day.[E23F18] Selected respondents acknowledged that the long hours were detrimental to workers' health.When asked how they would prefer to organise working time, E3125 said: 'I would put the workers to work for 8 hours a day, […] That is natural.I want to look out for my labourers'.
The interviews conveyed the impression that owners-managers genuinely wanted to take good care of their workers, but many lacked the intellectual, physical, and financial resources to implement global labour standards.The lack of resources was mentioned the most frequently when respondents were asked how labour standard compliance could be enhanced: Improving the work environment depends very much on the prosperity of the company, the more the company is prosperous the more you can give a better work environment.[E19F14] Furthermore, respondents repeatedly described workers' poor economic, social, and educational background as an obstacle to improving productivity and labour standards: The employees especially here are [unreliable and breach the rules]-I know that they are under stress and have financial problems and grievances, but work is work.[E23F18] Buyer power was not highlighted as relevant for working hours design.Only one respondent mentioned procurement demands in the export industry to be highly demanding.Buyers' labour standards were rarely addressed.

Findings of the cross-case analysis
The cross-case analysis compared working hours designs (see Figure 2) and rationales (see Table 4) of tier-1 and tier-2 garment suppliers in Jordan as well as those of garment and textiles suppliers in Egypt.In Jordan, working hours practices converged to global standards more markedly in tier-1 compared with tier-2 firms (horizontal differentiation).Buyer pressure to comply with labour standards was higher in tier-1 firms; they were more closely monitored and supported by BWJ; and their intellectual, physical, and financial resources were more developed.In Egypt, the findings revealed that different dynamics were in play in garment compared with textiles production (vertical differentiation).Buyer pressure to comply with labour standards was more apparent in the garment industry but did not significantly affect working hours.Practices were driven by legal considerations, as well as social and family obligations, especially for the female workforce.Textiles firms were equally affected by legal obligations, but only some of the interviewed firms complied with them.Scarcity of skilled labour and the earning needs of male workers incentivised constant overtime and excessive hours.

Discussion
The study analysed the diffusion of global labour standards in the garment supply chain and showed that standard adoption differs across horizontal nodes and vertical stages of the chain.The findings confirm that labour risks increase with horizontal subcontracting (Anner, 2019;Lebaron, 2014;Nadvi & Raj-Reichert, 2015) but refute that such claims can generally be made for vertical subcontracting.While some textiles suppliers showed high levels of divergence, others (almost) fully complied with global working hours standards.Average wages were furthermore significantly higher in textiles compared with garment production (CAPMAS, 2018(CAPMAS, , data: 2017)).Increased labour risks in lower vertical stages that could be found in the automotive and electronics supply chain (Villena & Gioia, 2018) must therefore be considered case specific.A closer look at the crossvergence of global, local, and firm-level factors influencing global standard adoption is imperative to make assumptions about the diffusion of global labour standards beyond tier-1 suppliers.
Given the factors underlying differences in standard adoption between lower-and higher-tier suppliers, the study confirms and extends previous research insights.As previously reported by Grimm et al. (2016) and Nadvi and Raj-Reichert (2015), buyers enforce labour standards primarily at their tier-1 suppliers.Lead firms lose power when direct exporters subcontract, especially when subcontracting is hidden (Gold et al., 2020).Going beyond previous research findings, this study evaluated how the deconstruction of buyer power to enforce labour standards interacts with other influences on labour standard adoption, particularly (1) global procurement pressures, (2) structural firm-level differences across tier-1 and tier-2 textiles and garment firms, and (3) different levels of embeddedness in cross-level support systems. Figure 3 illustrates this interaction.
1. Tracing the impact of global procurement demands showed that procurement demands diffused through the supply chain and, in the case of tier-2 suppliers in Jordan, even amplified along the chain.The extent to which procurement demands were passed on to lower horizontal nodes and vertical stages of the chain depended on the power relationships between tier-1 and lower-tier suppliers.
In Jordan, the large multi-national tier-1 suppliers dominating the industry had significant power over their smaller lower-tier suppliers and could thus reduce prices and lead times with little negotiation.In Egypt, garment and textiles production were closely intertwined and power relationships more balanced.This contributed to the finding that competitive pressure persisted but not amplified in textiles compared to garment production.The continuing rise of large multi-national tier-1 suppliers in the garment supply chain (Raj-Reichert, 2019) might therefore exacerbate labour risks for lower-tier suppliers.Looking at the conflict between labour standard and procurement pressure highlighted in previous research (Amengual & Distelhorst, 2020;Anner, 2019;Reinecke & Donaghey, 2021), it consequently shifted in favour of the procurement pressure in lower nodes and stages of the supply chain, particularly in Jordan.While labour standards were barely audited in subcontracting firms, quality inspections took place at every node and stage of the chain, and competitive prices and lead times were equally enforced.Attention therefore needs to be paid to procurement strategies of global buyers and (large) tier-1 suppliers to improve labour standard compliance throughout the garment supply chain.2. The study highlights the importance of the internal firm environment for understanding variation in labour standard adoption across tier-1 and tier-2 textiles and garment firms.A resource framework was developed to structure firm differences.Limited resources evidently contributed to labour standard non-compliance in Jordanian tier-2 factories and represented a barrier to labour standard implementation in Egypt.Resources have previously been linked to CSR implementation (Bux et al., 2020) and sustainable HRM (Muñoz-Pascual & Galende, 2020) but were rarely addressed in the debate on labour standard diffusion in global supply chains.Notable exceptions pointed out that limited managerial, technical, and financial resources of lower-tier suppliers (Chand & Tarei, 2021;Nadvi & Raj-Reichert, 2015;Villena & Gioia, 2018) and the varying skills of textiles and garment suppliers' human resources (Barrientos et al., 2011;Gereffi & Fernandez-Stark, 2016) can help to explain why labour standards dissolve in supply chains.The resources were, however, not further specified or systematised.In this study, Osterwalder and Pigneur's (2010) differentiation of human, intellectual, physical, and financial resources provided the starting point to condense the multiple differences that exist on firm level into a workable framework.As the work of Osterwalder and Pigneur (2010) is not specific to suppliers, the empirical analysis identified the human, intellectual, physical, and financial resources that impacted working hours designs of tier-1 and lower-tier suppliers (see Tables 3 and 4).Sales, HRM, leadership and manufacturing planning capabilities (intellectual resources), as well as profit margins and access to outside capital (financial resources), were less developed among tier-2 compared to tier-1 suppliers, hindering the adoption of global working hours standards.BWJ found only 33% of the tier-2 suppliers enrolled in their program to make a profit or break even (BWJ, 2017(BWJ, , data: 2016)).Resource rationales also explained why labour standards varied at different vertical stages of the supply chain and even overshadowed the influence of global dynamics.Work-life priorities of human resources differed between the male-dominated textiles and the female-dominated garment industry in Egypt and the machine-intensive work in the textiles industry required other working hours designs than the labour-intensive one in the garment industry.3. Glo-cal multi-stakeholder initiatives increasingly monitor and support sustainability issues in global supply chains (Baumann-Pauly et al., 2017).The Better Work initiative that operates in 10 garment production countries, including Jordan and Egypt, advises suppliers on how to implement global labour standards and monitors their success.The program has been shown to improve labour standard compliance in practice.(Alois, 2018;Pike & English, 2022;Rossi, 2015).Therefore, it is problematic that higher-and lower-tier suppliers profit to different degrees from such programs.The interviewed tier-1 suppliers in Jordan had participated significantly longer in the BWJ program than their subcontractors, and buyers were primarily interested in their results (BWJ, 2017).Glo-cal multi-stakeholder initiatives often depend on the support of global buyers to co-finance the program and to encourage their suppliers to enrol.Hence, tier-1 suppliers are represented to a larger extend than their subcontractors, increasing the compliance gap between higher-and lower-tier suppliers.In this study, this effect was especially visible for the case of horizontal subcontracting.

Implications for theory and practice
This study was theoretically framed by a crossvergence logic connected with existing research on the diffusion of labour standards in global supply chains.The study contributes four practical insights for improving working hours in the global garment supply chain.First, the frequently limited resources of lower-tier suppliers in global production impair compliance with global labour standards.The enforcement of global labour standards therefore needs to be supported by bottom-up/local approaches that support suppliers in improving working conditions.Multi-stakeholder initiatives that directly work with suppliers and other local stakeholders serve this purpose.This study shows that they are especially needed in lower nodes and stages of the supply chain, where they are currently rare.Initiatives are therefore called to focus on lower nodes and stages of the supply chain where their help is needed the most.The resource framework developed in this study can guide their actions, particularly regarding working hours standards.Bolstering intellectual, physical, and financial resources to improve labour standards as well as workers' capacity to fight for humane working conditions is crucial in the face of the existing limitations of global, local, public, and private labour regulation.Second, efforts to address resource limitations are likely to be insufficient on their own vis-à-vis the existing procurement pressure in the garment supply chain.The case of tier-1 suppliers in Jordan shows that full compliance with labour standards puts even direct exporters on or over the edge of competitiveness.The incentive to pass on the pressure to lower-tier suppliers is high.For smaller, less powerful sub-suppliers, the procurement squeeze becomes even more severe.As voluntary initiatives to change purchasing practices of buyers and tier-1 suppliers have yielded limited results, this study supports the need for greater legal regulation.Buyers and tier-1 suppliers who are sincere in their efforts to promote decent work in their supply chains need to rethink their procurement practices.As suggested by Amengual et al. (2020) and Amengual and Distelhorst (2020), procurement strategies ought to be adjusted to compliance ratings of suppliers.The adjustment however needs to go beyond increasing order volumes and include better price and lead time conditions and longer-term business relationships.Especially fluctuating order volumes make it difficult for suppliers to keep working hours stable.Third, contemporary private regulatory efforts of global buyers largely fail to reach lower-tier suppliers.To reduce labour risks in their multi-tier supply chains, buyers can rely on tier-1 suppliers to cascade standards down the chain (Villena, 2019), nominate sub-suppliers which they monitor themselves (Fontana et al., 2021), prohibit unauthorised subcontracting (Kashmanian, 2018), participate in glo-cal multi-stakeholder initiatives that work with sub-suppliers (Tewari, 2017), or adopt a systems approach to eliminate root causes of labour challenges in sourcing countries (Ripley, 2020).The first three solutions serve to increase global labour standard pressure beyond tier-1 suppliers.The case findings presented in this study show that the chances of success of these strategies are low if local and firm-level challenges are not addressed.Buyers are instead advised to cooperate with and to request their suppliers and su-suppliers to enrol in initiatives in their sourcing countries, particularly glo-cal multi-stakeholder initiatives that address the systemic factors that hinder global labour standard compliance.Table 4 provides an overview of factors that promote/hinder the adoption of global working hours standards.Fourth, worker and management education being repeatedly linked with substandard working conditions were unexpected.Many respondents considered education to be the key to resource development.This shows the need to integrate local technical and business schools, a currently underrepresented stakeholder group, into multi-stakeholder initiatives for sustainable change.Future research can thereby show in how far the development of local technical and business schools can facilitate decent work progress.

Limitations and future research
This study used a qualitative approach and focused on working hours designs in different vertical stages and horizontal nodes of the supply chain.This approach carries following limitations that are recommended to be addressed in future research.First, the maximum variation sampling strategy used in this study enabled to identify common characteristics of working hours designs and rationales across a wide range of (sub-)suppliers within each context.The non-representative nature of this strategy prevented analysing the distribution of working hours designs within the industry or the relative importance of identified influencing factors.Within the analysed research cases, validating and extending the presented findings with the help of quantitative research would therefore be worthwhile.Especially in the case of tier-2 suppliers in Jordan, only a few interviews could be conducted and expert knowledge was limited.In the case of Egypt, HR practices and rationales varied significantly.Creating a typology of different supplier groups within one node or stage of the chain would advance understanding of power and resource dynamics across horizontal nodes and vertical stages of global supply chains.Second, the local context was kept constant for each cross-case comparison conducted in this study.Structural differences between local contexts that focus on different nodes and stages of the supply chain could therefore not be identified.Local influences were solely identified to put global influences in context.Analysing two country contexts furthermore helped disentangling the complex web of global, local, and firm-level influences, and to gain insights into the context-specificity of the results.While working hours designs differed between Egypt and Jordan, similar working hours rationales could be found across the two contexts (see Table 4).This is especially noteworthy as the degree of internationalisation and the dominant mode of governance differed between the two contexts.Many of the detected working hours rationales have furthermore been found in previous research to also apply to other textiles and garment production sides in the Global South.The relative impact of influencing factors is however likely to differ between contexts.The skill-level of workers (Barrientos et al., 2011;Gereffi & Fernandez-Stark, 2016) played a minor role in this study's comparison of lower-and higher-tier suppliers because local labour was found to be difficult to recruit across research cases.This is, however, likely to differ in contexts where labour force participation rates of women are higher than in Egypt (22%) and Jordan (17%) (World Bank, 2022;data: 2017).The identified working hours rationales and differences between higherand lower-tier suppliers are therefore likely to be transferable (Lincoln & Guba, 1985) within the Global South, but future research is recommended to evaluate their relative impact in other countries, especially for the big production countries such as China, Bangladesh, and Vietnam.Third, this study focused on the second and third stage of the garment supply chain.It remains to be evaluated how the process of crossvergence plays out in other stages of the garment supply chain and other multi-tier global supply chains.The transferability of research findings is thereby most likely to be found in buyer-driven supply chains that operate with captive modes of governance and production sides in the Global South.Fourth, this study focused on working hours designs.Global labour standards include norms for a multitude of organisational practices and sustainability standards of global buyers address economic and environmental issues in addition to social (including labour) ones.While the categorisation of resources into human, intellectual, physical, and financial resources might apply across standards, the exact resources that promote or hinder standard adoption differ and need to be identified for the other dimensions of labour and sustainability standards.Annex F

Overview of working hours practices in each research case
note 1: The size of the bubble illustrates how many respondents reported their firm to implement a certain practice.note 2: The terms 'regular hours' , 'overtime hours' , and 'excessive hours' are used in this graphic as implied by respondents, e.g.Textiles owners-managers characterized 8 h/d plus 4 oT h/d as their regular hours and the use of an extra shift as exceptional or excessive.for the pattern matching procedure and in the description of results, they were adjusted to the definitions used in global labour standards to make them comparable.

Figure 1 .
Figure 1.supply chain-focused research insights on the diffusion of global labour standards.

Figure 2 .
Figure 2. adoption of global working hours standards.
[…] They have to relax, they have to have days off, because if you do continuous overtime [in the sense of excessive overtime], […] productivity would go down.They would start to get sick […].[J06F03]

I
think if the people are willing, the [problem is their] ability, [their] education […].[My colleague] for example he was well educated [by the] ILO.We need 120 people like him, so I think it is education.[E13F10] HR in Egypt is really weak; most factories' HR departments are not executed productively.[E06F03]

Figure 3 .
Figure 3. global influences and supplier resources within one national context.

Table 1 .
global working hours standards in the research contexts.

Table 2 .
overview of owner-manager interviews.
note: The overview shows 34 instead of 30 interviews as it includes four interviews in integrated firms (egypt) in which textiles and garment operations were discussed.

Table 3 .
supplier resources that impact the adoption of working hours standards.
In subcontracting, it is longer hours.[…] Because there are deadlines with penalties […] if you get delayed.[And there is] a price issue.We can negotiate when they have a very high season, and they need you very much.[…] I remember for one style we took in the high season $1.6 for sewing.When the high season finished, they offered us $0.35.[…] You cannot do anything with [this little amount of money].[J06F03] Reinecke et al. (2019)e how global and local influences crossverged and were moderated by firm resources.While most influencing factors were present across research cases, their impact and interplay differed between cases, leading to different working hours designs.The findings have the following theoretical implications.First, it is imperative to supplement global rationales for the (non-) diffusion of global standards with local and firm-level ones to avoid misinterpretations and misguided interventions.If local level factors had not been accounted for, the convergence to global working hours standards in the garment industry in Egypt would likely have been attributed to labour standard demands of global buyers.A unilateral enforcement of working hours standards can furthermore harm workers if industry wages are below the local living wage, as it is the case in many garment production contexts(Anner, 2019), or lead to the insolvency of tier-2 suppliers, rendering their workers unemployed.By adapting the concept of crossvergence to the context of global supply chains, this study provided a theoretical lens to capture the complex, real-world situations of lower-tier suppliers in the Global South and demonstrated its applicability in the context of global supply chains.Second, global pressure to comply with labour standards exists in parallel with buyer procurement pressure.Global influences can therefore not be considered as a single factor in the con-/di-/crossvergence debate, particularly in the context of global supply chains While labour standard pressure decreases along the supply chain, procurement pressure does not, at least in the garment supply chain.Building on the work ofReinecke et al. (2019), future research needs to identify sustainable procurement strategies that allow for decent work progress in all nodes and stages of the supply chain.For lower-tier suppliers, this concerns procurement strategies of global buyers and of tier-1 suppliers, taking into account power-relationships between buyers, tier-1 suppliers, and sub-suppliers.The rise of large multi-national tier-1 suppliers in the garment supply chain (Raj-Reichert, 2019) increases the risk that procurement pressure is passed down the supply chain to less powerful sub-suppliers.Third, a firm resource logic was put forward to capture micro-level firm differences that explain the (non-)diffusion of global labour standards beyond tier-1 suppliers.In the case of horizontal subcontracting, price, lead-time, and flexibility pressure was passed on to tier-2 suppliers with limited resources.Even if global standards would be strictly enforced at these production sites, it is highly unlikely that they could meet them.Future research can expand this study's framework of human, intellectual, physical, and financial resources for the adoption of working hours standards (see Table3) to capture more labour standards, countries, and supply chains.Fourth, the characteristics of global supply chains offer a new research perspective for IHRM.The convergence-divergence debate has previously focused on comparing HR practices of local firms and MNE subsidiaries across different contexts.Thereby, it would be interesting to learn more about HR strategies of suppliers.While supply chain scholarship commonly depicts employment practices of suppliers as a mere question of labour standard compliance, little is known about the HR strategies that suppliers develop in response to the crossverging global and local influences they encounter.