Complex adaptive dynamics in a resource-constrained entrepreneurial ecosystem: a mixed-methods study of Nelson Mandela Bay

Purpose Complex adaptive systems (CAS) theory offers a useful lens for understanding entrepreneurial ecosystems (EEs), yet little is known about how CAS mechanisms manifest in resource-constrained metropolitan ecosystems in developing economies. Responding to calls for place-sensitive ecosystem research that studies EEs as a CAS, this study aims to examine how CAS mechanisms enable or constrain EE development in Nelson Mandela Bay, South Africa. Design/methodology/approach A mixed-methods design was used. Survey data from 300 economically active participants were analysed using multiple linear regression to identify salient ecosystem components, followed by 15 semi-structured interviews with key stakeholders analysed through thematic analysis and integrated using joint displays. CAS concepts – emergence, self-organisation, nonlinearity and feedback loops – were used as interpretive lenses across both phases. Findings Culture and city planning emerged as significant predictors of ecosystem perceptions, while business environment constraints inhibited development. Qualitative analysis revealed maladaptive self-organisation, weak institutional coordination and infrastructure failures, reinforcing negative feedback loops. The findings highlight constraint-dependent emergence, whereby constrained ecosystems generate partial adaptive responses that compensate for, rather than transform, systemic dysfunction. Originality/value The study provides an empirical example of CAS mechanisms operating in a resource-constrained metropolitan EE – a context underexplored in CAS–EE scholarship, focused primarily on well-resourced or institutionally stable ecosystems. The primary contribution is the theorisation of constraint-dependent emergence – a novel CAS concept identifying the specific boundary conditions under which constrained ecosystems generate partial adaptive responses. This extends path dependency theory, institutional voids scholarship and evolutionary EE frameworks. Methodologically, the study advances mixed-methods integration through theory-driven thematic analysis and causal loop diagramming of feedback mechanisms.