Innovation Ecosystems and Sustainable High Innovation Performance: Evidence from the Guangdong–Hong Kong–Macao Greater Bay Area

Abstract

The innovation ecosystem has a significant impact on regional innovation development. Based on the “actor-resource-environment” analytical framework, this study takes nine cities in mainland China within the Guangdong–Hong Kong–Macao Greater Bay Area (GBA) from 2012 to 2022 as case studies. By applying a comprehensive evaluation model and kernel density estimation, the study analyzes the development level and dynamic evolution of the innovation ecosystem in the GBA. Furthermore, using a mixed-method approach of dynamic QCA and NCA, the study explores the pathways to achieving high innovation performance in the GBA. The key findings are as follows: (1) The comprehensive score of the GBA innovation ecosystem shows a rising trend, with a pattern of multi-polarization, “top-tier effect”, and the persistence of strong performance within the region. (2) Technological innovation actors and the digital economy environment have become necessary conditions for achieving sustained high innovation performance in the GBA. Inter-group results highlight the significant temporal effects of technological innovation actors, R&D personnel input, and a public service environment, which exhibit S-shaped, W-shaped, and U-shaped patterns, respectively. Intra-group results reveal notable regional differences in the necessity level of seven conditional variables across GBA cities. (3) There are two types of innovation ecosystems driving high innovation performance in the GBA: actor-driven and resource–environment synergy-driven ecosystems. While no significant cross-time or cross-case effects exist between these two types, inter-group consistency changes indicate that the innovation ecosystem in the GBA has become more mature and stable since the establishment of the GBA in 2017.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.