Unlocking shared value: reimagining multi-stakeholder collaboration for future-oriented post-school skills development Open Access

South Africa's skills ecosystem remains under pressure to respond to rising youth unemployment, shifting labour market demands, and rapid technological disruption. Despite a suite of post-Apartheid reforms, post-school educational institutions continue to face criticism for producing graduates whose qualifications are misaligned with industry needs (Maringe and Osman, 2016). Chelechele (2009) shows that persistent workplace inequalities also contribute to skills shortages. International studies similarly highlight the global challenge of aligning education systems with labour market demands amid rapid technological and structural change in the Future of Work (FoW) (ILO, 2020; OECD, 2019; UNESCO, 2021). These debates emphasise responsive, coordinated skills ecosystems in which multiple stakeholders reduce mismatches and promote employability by developing adaptable, work-ready capabilities (Rikala et al., 2024).

Global development agendas, such as the Sustainable Development Goals, increasingly highlight the role of multi-stakeholder partnerships (MSPs) as a mechanism to bridge systemic implementation gaps in education and training (United Nations, 2020). International scholarship emphasises their value in improving coordination, policy coherence, and leveraging cross-sectoral expertise to improve skills outcomes (Higham et al., 2024). However, global studies also note that MSP effectiveness varies significantly across contexts, with challenges often linked to governance, communication, and institutional capacity (Bryson et al., 2015; Ratner et al., 2022). The need for coordination across government, higher education, industry, and civil society is evident in Post-School Education and Training (PSET) policy documents (DHET, 2019). Yet, real-world coordination challenges persist – ranging from fragmented communication to weak institutional trust – undermining the operational effectiveness of such partnerships.

PSET refers to all education and training opportunities available to individuals after they leave school (DHET, 2019). In South Africa, this includes universities, Technical and Vocational Education and Training colleges, Community Education and Training colleges, sector education and training authorities, and various private providers. Comparable post-school systems exist internationally; for example, Australia's tertiary education landscape comprises universities, technical and further education institutes, and registered training organisations (Guthrie, 2009; Wheelahan and Moodie, 2011), while the United Kingdom's system includes universities, FE colleges, apprenticeship providers, and employer-led training schemes (Hodgson and Spours, 2019). Across contexts, PSET systems are expected to remain adaptive, employer-engaged, and future-focused to support effective youth transitions (ILO, 2020; OECD, 2021; UNESCO, 2023).

A future-fit PSET curriculum should equip graduates with relevant, adaptable, in-demand skills for a changing labour market. Global literature on the Fourth Industrial Revolution (4IR) and digital transformation highlights the necessity of embedding transversal skills, digital literacy, and problem-solving within curricula (World Economic Forum, 2025). With the rise of the 4IR, automation, and digital transformation, traditional qualifications alone are becoming less sufficient (ILO, 2020; Schwab, 2024). Curricula therefore need to integrate critical thinking, digital literacy, and lifelong learning competencies to prepare learners for emerging industries and evolving job roles (World Economic Forum, 2025). In the absence of alignment between education systems and labour market needs, skills mismatches are exacerbated, reinforcing unemployment and constraining economic growth. Although curriculum innovation is frequently promoted as a remedy, the literature consistently indicates that such reforms are unlikely to achieve meaningful impact without collaborative governance, institutional coherence, and sustained engagement with labour market actors (Bryson et al., 2015; McGrath and Powell, 2016).

Stakeholder collaboration plays a pivotal role in creating a thriving PSET environment by aligning the goals, resources, and expertise of government, higher education institutions, industry, and civil society (Bryson et al., 2015; Emerson et al., 2012). Such collaboration ensures that curricula are relevant to labour market demands, that skills development initiatives are inclusive, and that transitions from learning to employment are smoother. Research on skills ecosystems, coordinated governance, and collective impact models underscores that no single actor can resolve skills challenges alone; effective collaboration requires shared vision, consistent communication, and clarity of roles (Bryson et al., 2015; Kramer and Kania, 2011).

When stakeholders work together strategically, they can co-create responsive programmes, address systemic gaps (such as funding, infrastructure, or workplace exposure), and build trust across the education and employment ecosystem (Billett, 2014). International evidence shows that sustained cross-sector collaboration can strengthen employer engagement, improve curriculum responsiveness, and expand pathways from learning to work (OECD, 2019). In essence, stakeholder collaboration transforms the PSET system from a fragmented structure into a coordinated, impact-driven platform for social and economic development.

Although global studies highlight the value of MSPs in skills systems, there is limited empirical research exploring how such partnerships operate within the South African PSET context and how strategic communication shapes their effectiveness (McGrath and Powell, 2016; OECD, 2019; Pattberg and Widerberg, 2016). This gap provides the motivation for the present study. This study, therefore, investigates how multi-stakeholder collaboration is enacted in South Africa's post-school sector, and how strategic communication – understood as deliberate, trust-building and goal-oriented interaction – can enhance the design and implementation of such partnerships (Bryson et al., 2015; Nel, 2017). The researchers draw on stakeholder theory (Laplume et al., 2008) to interrogate power, legitimacy, and role clarity among actors, and social systems (Görke and Scholl, 2006; Parsons, 2005) theory to conceptualise the PSET ecosystem as a set of interdependent subsystems under adaptive pressure.

Through this generic qualitative study, the researchers examine the relational, communicative, and structural dimensions that underpin effective MSPs for skills development. It addresses the following research questions:

According to van Damme (2014), skills levels are critical in shaping the economic condition of a country. An unskilled society widens the inequality gap because unskilled labourers do not participate in market productivity that grows the base of a country's economy (Uprety, 2019). International scholarship echoes this dynamic, showing that skills deficits – particularly those shaped by historical inequality – continue to constrain inclusive growth in many developing and post-colonial contexts (OECD, 2019; UNESCO, 2023). South Africa represents one such case, where high levels of inequality are deeply rooted in historical processes that deliberately restricted the development of skills among the majority population.

The structural exclusion of black South Africans from quality education and skilled professions under Apartheid contributed directly to the persistent skills deficit and economic marginalisation observed in the post-Apartheid era. This historical context underpins contemporary constraints on inclusive growth. While some scholars describe Apartheid as an ideology and social construct, it was a racially based government policy of separation (Giliomee, 2012). Essentially, it was a legalised system of oppression that perpetuated the suppression of black South Africans. Even after the democratic transition, the exclusion of black people from entering various professions remained a constant (Mariotti and Fourie, 2014; van Sittert, 2015). International sources similarly highlight how legacies of exclusion – whether through colonialism, segregationist schooling, or restricted professional access – continue to shape contemporary skills gaps (ILO, 2020; OECD, 2019; UNESCO, 2023; World Bank, 2019).

The National Skills Development Act 97 of 1998 was implemented to strategically address South Africa's skills shortage (Plant and Padotan, 2017). Rasool and Botha (2011) assert how education, across all sectors of the economy, plays a pivotal role in building a skilled labour force that meets the needs of the labour market. To strengthen the legislative mandate for skills development, the Skills Development Levy Act 9 of 1999 was implemented (Plant and Padotan, 2017). The purpose of this Act was to encourage workplace skills development, within the national qualifications framework, to adequately equip the South African workforce in both the public and private sector (Paterson et al., 2014). It also sought to accelerate skills development in the public sector and incentivise private-sector training (Tsotsotso et al., 2017). Contemporary scholarship increasingly demonstrates that policy mechanisms alone are inadequate; effective skills development depends on coordinated action among education providers, employers, and intermediary organisations, reflecting patterns identified in skills ecosystem research across both advanced and emerging economies. (Buchanan et al., 2017; CEDEFOP, 2018). Scholars agree that the scale of skills needs exceeds what policy can achieve alone and requires cross-sectoral collaboration (Petersen et al., 2016). Linking this to human capital formation, investments that combine foundational learning, work-based learning, and signalling through qualifications have persistent payoffs in earnings and productivity (Becker, 1975).

The future of work (FoW) refers to workplaces shaped by technological advancements and social shifts (Gratton and Scott, 2017). Moghaddam (2020) emphasises that preparing for future jobs requires acquiring skills relevant for the FoW. According to this understanding, it is necessary for the post-school system to adequately provide upskilling and reskilling of individuals to meet the needs of the labour market of the future. Effective skills development in post-school education is crucial as the FoW demands graduates who are not just knowledgeable, but also agile, critically minded, and digitally literate. A systematic review highlighted how the Fourth Industrial Revolution (4IR) emphasises critical thinking as the most critical competency, and that student-centred, active learning strategies are essential for preparing learners for complex, tech-driven environments (Aboderin and Havenga, 2024). Global syntheses call for transversal skills, digital fluency and lifelong learning architectures to navigate rapid technological change (ILO, 2020; OECD, 2020; UNESCO, 2021; World Economic Forum, 2025). International scholarship likewise indicates that digital transformation and labour market volatility necessitate education systems that embed transversal skills, foster lifelong learning, and enhance institutional responsiveness, frequently through collaboration with employers and social partners (Buchanan et al., 2017; CEDEFOP, 2018). Universities must therefore integrate such pedagogies and work collaboratively with key stakeholders, as they cannot address system-level challenges alone.

MSPs bring together cross-sector actors to address complex societal challenges (Moghaddam, 2020). These collaborations are voluntary and built on shared visions, responsibilities, and resources (Banerjee et al., 2020). The “skills mismatch” created by an education system that fails to produce enough skilled workers for the labour market requires the contribution of cross-sector stakeholders to address this misalignment (Bhorat et al., 2016). These stakeholders include government, industry, civil society, and educational institutions, amongst others. While this challenge is acute in South Africa, it mirrors skills coordination problems evident in many emerging and advanced economies, where shifting labour-market demands require more adaptive forms of collaboration (Billett, 2014; CEDEFOP, 2018; OECD, 2021). MSPs are especially critical in contexts like South Africa, where systemic inequalities and evolving labour market demands require coordinated, inclusive, and adaptive responses to ensure that skills development is both relevant and equitable.

International research reinforces the value of MSPs in strengthening coordination, ensuring labour market alignment, and improving responsiveness in skills systems (Bryson et al., 2015; Emerson et al., 2012; Kramer and Kania, 2011; OECD, 2021). Studies from Europe highlight that partnership-based governance structures (such as skills councils, employer coalitions, and sectoral alliances) improve information sharing and enable more consistent curriculum adaptation (CEDEFOP, 2018; UNESCO, 2022). Yet, such partnerships also face common challenges, including misaligned expectations, weak communication systems, and power imbalances among actors (Emerson and Nabatchi, 2015). These recurring global challenges contextualise, rather than isolate, similar issues observed in the South African system, where collaboration remains fragmented despite clear policy mandates.

Effective strategic communication in MSPs is essential for skills development because it ensures clarity of roles, alignment of goals, and sustained trust among diverse partners (Bryson et al., 2015; Cornelissen, 2023; Nel, 2017). Given the complexity of cross-sector collaboration, strategic communication helps prevent misunderstandings, fosters accountability, and promotes shared ownership of skills initiatives. In MSPs, communication functions as a strategic capability that builds and maintains relationships for joint problem-solving (Bryson et al., 2015; Cornelissen, 2023). Nel (2017) states that trust and credibility in MSPs are fundamental to their success. Transparent communication cultivates a working environment built around trust and accountability (Jansen and Kalas, 2020; Nel, 2017). The literature increasingly positions communication not as an operational activity but as a strategic capability central to governance, coordination, and long-term partnership sustainability.

The theoretical basis of this study was founded on a combination of social systems theory and stakeholder theory. Social systems theory is described as the study of human interaction within social systems (Parsons, 2005). This theory has been used to examine the dynamic construct of society and its various social sub-systems (Brazhnikov, 2017). Educational and skills research applies systems thinking to analyse complexity, feedback and adaptation in skills ecosystems (Luhmann, 1995). Social systems are categorised by their agile nature that allows them to withstand pressures and constraints that threaten equilibrium (Görke and Scholl, 2006). The formation of MSPs to achieve skills development goals can be understood as a collaboration between interconnected parts of a larger system, working together to respond and adapt to social and economic pressures (Mayrhofer, 2004). Comparative studies show that systems-oriented reforms in VET and higher education hinge on coordination across nested institutions and intermediaries (Wheelahan and Moodie, 2017).

While social systems theory highlights the adaptive and interconnected nature of systems responding to shared pressures, stakeholder theory adds a layer of analysis by examining how different actors within those systems hold varying degrees of power, legitimacy, and influence. Together, they offer a comprehensive lens to understand how MSPs operate, not only as structural responses to social challenges but also as arenas where stakeholder dynamics shape decision-making and collective outcomes. Stakeholder theory examines how stakeholder attributes and salience determine power, legitimacy and urgency (Freeman, 1984; Lafreniere et al., 2013). This is crucial in organisational contexts where influence and accountability must be clarified (Laplume et al., 2008). It provides the most suitable framework to examine the interrelated nature of stakeholder groups in a multi-stakeholder setting (Walby, 2007). Despite its global application in organisational and governance research, stakeholder theory is underutilised in PSET-specific studies in South Africa, highlighting a gap this study addresses. In practical terms, these lenses informed our coding frame: system-level categories captured interdependencies and feedback loops, while stakeholder-salience codes captured power, legitimacy and urgency in participants' accounts.

The study adopted an interpretivist tradition. An interpretivist approach is appropriate when the aim is to understand how stakeholders construct meaning around collaboration, communication, and institutional roles within the PSET system. This stance is well suited to complex social processes and enables analysis of participants' subjective experiences in dynamic, multi-stakeholder settings. Accordingly, a qualitative design was employed to generate rich, contextualised insights into relationships, power dynamics, and institutional practices within MSPs. To situate the design in prior work, the approach extends earlier MSP and skills-governance studies by applying an interpretivist lens in a South African PSET context and by explicitly linking analysis to social systems and stakeholder theory.

The study drew on a purposive sample of 24 respondents from diverse roles across the skills development and education ecosystem in South Africa. Participants were senior professionals and executives across government, higher education, non-profit organisations, and private training providers. Maximum-variation logic guided selection to capture breadth across sectors and roles. Most interviewees held leadership roles with responsibilities in labour market intelligence, stakeholder engagement, skills development, and policy implementation. Interviews averaged 46 min, indicating strong engagement. Purposive sampling was suitable because the study required participants who possessed expert knowledge of PSET governance and partnership processes, consistent with similar qualitative MSP studies internationally (Palinkas et al., 2015). Recruitment continued until informational saturation was reached and no new codes emerged. Participants responded to targeted prompts aligned with each research question. The interview guide mirrored each research question (RQ) with 2–3 prompts and probes to ensure coverage while allowing emergent insights.

The data were collected through 24 semi-structured interviews (SSIs). Interviews were conducted both in person and online. Participants completed and returned informed consent forms to the researchers before the commencement of all interviews. Permission to conduct the study was granted by the Ethics Committee at the researchers' host institution. Permission to record interviews was acquired. Interviews were transcribed verbatim using a digital transcribing tool to ensure accuracy and support reliable analysis. SSIs were chosen because they allowed for conversational depth while maintaining alignment with the study's conceptual framing (Bryman, 2016), a method frequently used in MSP and skills ecosystem research.

The data were analysed using thematic analysis. This involved identifying and organising patterns in the data to develop codes and themes aligned with the study's contextual features. The data were analysed through a rigorous process of reading the transcripts and listening to the interview recordings. Analysis involved repeated reading of transcripts and listening to recordings to identify patterns that informed codes and themes. Initial codes were developed inductively from the data, capturing key ideas and recurring patterns, and then grouped into sub-themes and overarching themes aligned with the research questions. To enhance coherence, sub-themes were mapped to the guiding theories: system-level codes captured interdependencies and feedback (social systems), while stakeholder-salience codes captured power, legitimacy, and urgency (stakeholder theory). Two researchers independently coded a subset of transcripts and resolved discrepancies through discussion to strengthen dependability. Thematic analysis was appropriate because it supports both inductive and deductive engagement with data (Braun and Clarke, 2006) and is widely used in qualitative studies of governance and education partnerships (Braun and Clarke, 2013; Bryman, 2016; McGrath and Powell, 2016).

To enhance analytical transparency and support replicability, the coding process followed an iterative sequence: (1) familiarisation with transcripts; (2) inductive open coding; (3) clustering and refinement of codes into intermediate categories; and (4) consolidation into the final themes. Codes and categories were iteratively reviewed against the full dataset to confirm internal coherence and distinct theme boundaries (Braun and Clarke, 2006, 2013). A brief indication of the code-to-theme reduction is provided in Table 1 to show how initial codes were grouped into intermediate categories and final themes.

Trustworthiness was ensured through methodological rigour guided by credibility, dependability, confirmability, and transferability (Maher et al., 2018; Tobin and Begley, 2004). Credibility refers to the extent to which qualitative researchers truthfully and unambiguously present a study's findings and was strengthened through SSIs, which enabled open-ended discussion, clarification, and reduced misinterpretation (Ryan et al., 2007; Tuckett, 2005). Participants were assured of confidentiality, and prolonged engagement plus member-checking of preliminary themes further enhanced credibility.

Dependability denotes the extent to which a researcher provides an account of all processes followed in answering the research question and was supported by transparent reporting of the research design and analytic procedures (Koch, 2006; Nowell et al., 2017). Confirmability, which concerns the objectivity of data collection and interpretation (Koonin, 2014), was enhanced through pilot testing, refinement of the interview guide, and verbatim transcription.

Transferability indicates the extent to which a study's findings can be replicated and was addressed through thick description of context, sampling, and thematic patterns (Dean, 2018; Ryan et al., 2007). Clear research questions and detailed methodological accounts support future replication, and findings are most applicable to multi-actor skills initiatives in resource-constrained, emerging-economy PSET systems.

The themes identified resonate with existing literature emphasising the interconnectedness of actors within broader social systems (Brazhnikov, 2017; Parsons, 2005) and the need for coordinated responses to skills mismatches (Bhorat et al., 2016). This alignment reflects the global evidence base showing that systemic skills challenges require collaborative, adaptive, and multi-actor responses across education and labour market ecosystems (OECD, 2019; UNESCO, 2021). Each theme is discussed below in relation to the three research questions. Table 1 presents an integrated overview of the final themes, associated sub-themes, illustrative initial codes, alignment with the research questions, and the frequency with which each theme appeared across the 24 interviews. While the study remains qualitative and interpretivist, the inclusion of frequency counts provides additional insight into the relative prominence of themes without implying statistical generalisation.

Across the 24 interviews, stakeholder complementarity and system interdependence emerged as the most prominent theme (n = 19), followed closely by strategic communication for partnership effectiveness (n = 18). Institutional alignment and policy coherence was raised by approximately two-thirds of participants (n = 15), while partnership sustainability and accountability mechanisms received moderate emphasis (n = 12). Quality education as a shared development outcome featured prominently across interviews (n = 14), particularly in relation to early childhood development (ECD) and holistic learner outcomes.

Participants demonstrated a clear recognition that effective skills development depends on interdependence between government, industry, civil society, and education actors. About three-quarters emphasised that no single institution can drive system-level change alone. Government was seen to provide policy instruments and funding, civil society was seen to offer grassroots reach, and industry shapes demand and labour absorption. This aligns with social systems theory, which highlights the interconnectedness of societal subsystems (Parsons, 2005), and with global skills-ecosystem literature stressing coordinated pathways into work (CEDEFOP, 2018; OECD, 2021).

However, respondents also raised concerns about unclear mandates, duplication, and informal collaborations that sometimes clashed with formal governance structures. Power asymmetries were a recurring issue, with larger institutions often dominating decision-making. These concerns reflect stakeholder-theory insights on power, legitimacy, and influence (Laplume et al., 2008) and echo international research showing that weak role clarity undermines accountability (Bryson et al., 2015). Across interviews, participants stressed the need for explicit roles, decision rights, and structured escalation processes.

Collaboration was repeatedly framed as essential rather than optional. As P2 explained, “we cannot be experts in all sectors … we need experts from different industries to assist us in developing policies and channelling funding.” This illustrates the role of MSPs in bridging knowledge and resource gaps, particularly where state capacity is constrained (Bhorat et al., 2016; Moghaddam, 2020).

Participants also highlighted a sequential skills-pipeline logic in which government provides foundational training while industry refines applied competencies. Concerns about responsiveness were strong. P7 warned that without alignment to industry and community needs, “you are likely to be skilling young people for unemployment.” This sentiment mirrors global critiques of slow-moving education systems in rapidly changing labour markets (Gratton and Scott, 2017). As P7 further observed, “If you are not involving your cross-sectional stakeholders … you are not going to survive,” reinforcing arguments that structural collaboration is fundamental to addressing large-scale skills challenges. Practically, these findings suggest that partnership design should formalise role clarity early, including decision rights and escalation routes, to avoid duplication and mitigate dominance dynamics.

Participants consistently described communication as central to building trust, coordinating activities, and sustaining engagement in MSPs. They emphasised that effective partnerships require clear objectives, aligned expectations, and regular, transparent communication. This reflects Nel's (2017) view that trust and communication underpin MSP functionality, aligns with Jansen and Kalas (2020) call for transparency, and supports governance literature highlighting communication as a core capability in managing complex, cross-sector collaborations (Bryson et al., 2015; Cornelissen, 2023). Two-way, iterative communication was described as especially important where partners came from different organisational cultures and decision-making traditions.

A recurring emphasis was the role of boundary spanners or communication “champions” who maintain continuity, translate expectations, and align agendas across institutions. These roles mirror stakeholder-theory insights on actors with relational influence (Lafreniere et al., 2013) and global findings on intermediaries and skills brokers in collaborative systems (CEDEFOP, 2018). Participants also cautioned that communication must be substantive rather than symbolic; tokenistic updates risk eroding trust and reinforcing power imbalances.

Communication was also described as a cultural and systemic competency that fundamentally shapes partnership functionality. As P11 explained, “communication is very, very important … in the form of relationships … but also the messaging and clarity of roles.” This reflects its dual purpose: maintaining relationships while clarifying expectations, boundaries, and responsibilities through structured protocols, meetings, and shared tools.

Participants further highlighted the value of formalisation, including agreements that document roles, deliverables, and commitments. P2 noted that such agreements ensure accountability because “everyone is able to report on their roles … we work accountable for what we have committed on.” When paired with KPIs and review cycles, these instruments strengthen follow-through.

Finally, communication was linked to system-wide learning and strategic intelligence. P1 captured this in noting that integrated reporting supports relevance, policy alignment, and shared understanding of research outputs. This positions communication not only as coordination but as a mechanism for collective insight and adaptive decision-making. This points to communication as a governance capability, not simply an operational activity. Formal communication protocols, boundary-spanning roles, and agreed reporting cycles can strengthen partnership durability and credibility.

The third research sub-question examined how institutional and policy frameworks can better support effective MSPs. Participants repeatedly highlighted misalignment between policy intent, curriculum design, and rapidly evolving skills demands, echoing concerns in Rasool and Botha (2011) and Tsotsotso et al. (2017). Although policies such as the National Skills Development Plan outline clear priorities, implementation was seen as slow, fragmented, and poorly coordinated across the PSET system.

Participants called for more agile policymaking and responsive feedback loops so that stakeholder insights can inform curriculum updates and strategic interventions in real time. This need for adaptability aligns with 4IR and FoW literature, which shows that traditional policy cycles cannot keep pace with technological change (Aboderin and Havenga, 2024; Gratton and Scott, 2017). Internationally, such slow system responsiveness contributes to persistent skills mismatches (OECD, 2020; UNESCO, 2021).

A major concern was the weak vertical integration between basic education, higher education, and workplace training. Participants felt this fragmentation disrupts learner pathways and limits the scalability of partnership-led solutions. Stronger institutional coherence – through joint planning forums, curriculum co-design, and shared data systems – was viewed as essential for making MSPs effective.

These challenges were often framed through the lens of 4IR transitions. As P1 noted, “If we're looking into the future of work… there needs to be policy alignment in terms of where we are going.” This reflects perceptions of a reactive policy environment tied to outdated occupational categories rather than anticipatory governance.

Curriculum relevance was also a persistent theme. P4 emphasised the need to “align the training at institutions to the needs of the labour market”, reinforcing arguments for iterative curriculum co-design with industry (Bhorat et al., 2016; CEDEFOP, 2018). Participants further stressed transversal competencies – analytical thinking, communication, teamwork, problem solving, innovation – as essential for 4IR readiness, mirroring global debates (Aboderin and Havenga, 2024).

Finally, several called for a system-wide labour market intelligence function. As P11 argued, “This should be even at a policy level … determine every year how many students will go into which areas.” This underscores the need for a centralised skills observatory or forecasting mechanism (Plant and Padotan, 2017) to guide programme development and resource allocation. The findings suggest that MSPs need formal interfaces into curriculum and policy cycles, including regularised labour market intelligence inputs and co-design mechanisms that shorten the feedback loop between demand signals and programme adaptation.

A cross-cutting theme across the second and third sub-questions was concern about the long-term sustainability and governance of MSPs. Participants noted that although partnerships often begin with enthusiasm, sustaining commitment, shared accountability, and consistent engagement is far more challenging. They emphasised the need for clear expectations, structured communication, and reporting mechanisms from the outset, reinforcing Banerjee et al.’s (2020) view that shared values and mutual accountability underpin durable partnerships.

Stakeholders also highlighted the importance of integrated reporting frameworks that track performance, impact, and lessons learnt. Such feedback loops reflect social systems theory, which links system stability to ongoing communication and adaptation (Görke and Scholl, 2006). Participants agreed that shared values and mutual benefit are especially important during periods of resource strain; without them, partnerships risk fragmentation.

Threats to sustainability were also evident. P20 warned of goal drift when “intention can be lost … when others are in it for public relations (PR) purposes,” pointing to the problem of preformative partnerships. P16 noted that “not everyone has the same passion … and that might affect the whole chain,” echoing global findings that uneven commitment undermines partnership stability (Emerson and Nabatchi, 2015). Similarly, P15 cautioned that “people will use the partnership to satisfy their personal goals,” underscoring the need for clearer ethical safeguards and accountability structures.

Participants also emphasised the role of champions in sustaining momentum. For P22, “having champions run a specific project … creates the assurance that projects will be sustainable.” Champions were viewed as relational anchors who maintain trust and institutional memory, consistent with Nel (2017). Codifying these roles and planning for succession were seen as essential.

Finally, recognition and visibility were identified as important mechanisms for reinforcing commitment. As P22 observed, “sharing success stories … helps in retaining more sustainable partnerships,” while P14 highlighted that consistent reporting keeps partners engaged beyond financial incentives. These insights show that sustainability depends not only on structure, but also on transparency, recognition, and emotional engagement. These findings indicate that sustainability is supported by accountability infrastructure (reporting, review cycles, defined commitments) and by relational continuity (champions, succession planning, institutional memory).

Participants consistently stressed that the ultimate purpose of MSPs is to advance quality, holistic education. Quality was understood not as exam performance alone, but as the development of social, emotional, cognitive, and moral capabilities that support long-term social mobility, echoing Uprety's (2019) emphasis on capability formation. ECD was repeatedly identified as a critical leverage point for societal transformation, while others framed quality education as foundational for equity and poverty reduction, aligning with social systems theory's view of education as central to socio-economic development (Parsons, 2005). This theme ties together the earlier findings: stakeholder alignment, communication, and policy coherence are meaningful only insofar as they enable a transformative, learner-centred system.

Concerns about governance and accountability were prominent. P16 noted “very little accountability … our resources, the curriculum, contact time with children,” linking system weaknesses to learning outcomes. The professional ethos of teachers was also highlighted, with P16 observing that many enter the profession “as a means to an end … to find a teacher who is truly passionate …” – suggesting that educator commitment shapes the quality of learning environments.

Access was also viewed as integral to quality. As P24 explained, “Quality education provides a platform where learners can excel through access to the highest level of education.” This aligns with global debates emphasising equity and future-readiness (Aboderin and Havenga, 2024; Gratton and Scott, 2017).

Participants overwhelmingly endorsed a holistic conception of education. P16 observed that quality “is beyond academic qualifications … it deals with the whole being of the child,” prompting institutions to support caregivers and educators directly. Examples included equipping caregivers with practical resources and school-level coaching efforts to shift ECD paradigms, as P14 described.

Finally, education was framed as profoundly transformational. P15 argued that it should “liberate and transform the community” and prepare future leaders, while P14 emphasised its role in poverty eradication. Collectively, these accounts position quality education as the shared development outcome that legitimises partnership work and underscores education's role in building societal resilience (Uprety, 2019). Operational implications include strengthening teacher capability, early learning ecosystems, work-integrated learning pathways, and simple metrics for access, progression, and learner well-being. The data position “quality” as a unifying outcome that can align partners across sectors, supporting MSP objectives that connect skills pipelines to broader educational foundations.

The study demonstrates that MSPs can significantly strengthen South Africa's skills ecosystem, but only when intentionally structured around clear roles, sustained communication, and coherent institutional frameworks. In line with international analyses (Bhorat et al., 2016; Moghaddam, 2020), the findings show that MSPs operate as adaptive subsystems within broader socio-economic environments, where interdependence, trust, and shared purpose are essential for system stability (Brazhnikov, 2017; Parsons, 2005). While stakeholders agreed that no single actor can address skills shortages alone, they also highlighted fragmented governance, unclear mandates, uneven commitment, and superficial engagement as persistent barriers, echoing concerns about power asymmetries and unbalanced participation (Jansen and Kalas, 2020; Nel, 2017).

Strategic communication emerged as central for expectation alignment, accountability, and trust-building, yet current communication systems were often inconsistent, weakening partnership cohesion. Similarly, policy and curriculum reform – critical to responding to rapid shifts in the FoW and 4IR contexts (Aboderin and Havenga, 2024; Gratton and Scott, 2017) – remains slow, poorly coordinated, and insufficiently informed by labour market intelligence, limiting system responsiveness.

Participants repeatedly framed quality education – holistic, equitable, and future-ready – as the shared outcome that should orient partnership efforts, consistent with global scholarship on transformative and capability-oriented systems (Uprety, 2019; van Damme, 2014). Strengthening MSPs therefore requires intentional governance practices that support long-term, cross-sector collaboration.

Practically, this entails institutionalised communication infrastructures such as integrated reporting, feedback loops, and boundary-spanning roles to support coordination and shared learning; clearer stakeholder roles and accountability frameworks to reduce duplication and mitigate dominance; and stronger policy-curriculum-industry alignment to ensure agility in meeting labour market needs. Sustained co-investment in partnership infrastructure – particularly for under-resourced organisations – would bolster durability and equity. Finally, MSPs should advance a systemic view of quality education that spans early childhood to lifelong learning and integrates cognitive, psychosocial, and ethical development.

Taken together, these improvements position MSPs as structured, communicative, and purpose-driven mechanisms capable of enhancing labour market alignment and enabling meaningful educational transformation in South Africa.

The study offers valuable, practice-based insights but is constrained by its single-country, single-case focus and a purposive sample weighted toward senior stakeholders. This may have excluded important perspectives from frontline educators, students, and community members. The reliance on self-reported interview data also raises the possibility of social desirability bias, a common limitation in qualitative, interpretivist research. Future studies should therefore triangulate interview findings with document analysis or limited observation to strengthen validity.

Further research is needed to examine MSPs longitudinally, tracing how communication, governance, and role dynamics evolve over time. Comparative work across emerging economies would deepen understanding of contextual differences and enhance international relevance. Including youth and student voices is particularly important, given their position as primary beneficiaries of skills development initiatives. Finally, mixed-methods research incorporating quantitative impact measures would complement the qualitative insights presented here and provide stronger evidence for policy and institutional reform.