Sustainable Development and Inorganic Growth: A Bibliometric and NLP Driven Review of ESG in Mergers and Acquisitions (1982–2024)

ABSTRACT

Achieving the UN Sustainable Development Goals by 2030 requires firms to embed sustainability into core strategic decisions. Mergers and acquisitions (M&A), a primary vehicle for inorganic growth, are increasingly shaped by Environmental, Social, and Governance (ESG) considerations. However, the ESG–M&A intersection remains fragmented and theoretically underdeveloped. This study systematically maps the intellectual architecture of this emerging field, identifies latent thematic frontiers, and derives theoretically grounded directions for future research. Following a PRISMA‐guided selection protocol, 1225 articles from the combined Web of Science and Scopus databases (1982–2024) were analysed using a three‐method triangulation design. Bibliometric analysis—including citation, co‐citation, bibliographic coupling, and keyword co‐occurrence—identified dominant scholars, journals, countries, and research themes. Latent Dirichlet Allocation (LDA, k  = 11) and BERTopic ( k  = 13) were applied to extract latent semantic structures and uncover nuanced thematic clusters beyond the reach of conventional co‐word methods. Existing literature on M&A as strategy revealed dominant streams such as valuation, corporate governance, and performance. However, ESG integration remained underexplored. NLP‐driven models provided novel insights, uncovering emerging clusters on green logistics, skills and social development policy, gender and biodiversity, ethical consumption, and psychological drivers of green finance—topics overlooked by conventional methods. Combining bibliometrics with NLP techniques offers methodological novelty, exposes new thematic frontiers, and informs future scholarship and practice in sustainability‐driven corporate strategies. This study presents the largest assembled corpus at the ESG–M&A intersection to date, uniquely combining classical science mapping with transformer‐based semantic clustering to map four decades of interdisciplinary scholarship.