Data-driven approach to find the best partner for merger and acquisitions in banking industry

Purpose: Merger and acquisitions (M&A) is a process of restructuring two or more companies into one, a process that occurs frequently in many companies. Previous studies on M&A mainly paid attention to the potential gains from a merger, while ignored the problem of how to select the partners to merge. This paper aims to select the best partner from different candidates for a given company to merge. Design/methodology/approach: Each company's historical data are used to identify each company's own production technology. With resources change, each company's new operation is restricted by its own production technology. Then, a 0–1 integer programming is proposed to select the best partner for M&A. Findings: The banking industry involving 27 China's commercial banks is given to verify the applicability of our proposed model. The study shows the best partner selection for each bank company. Originality/value: On the theoretical side, the study uses each company's own historical data to construct its own production technology to compressively reflect the production change after M&A. On the practical side, the study uses the proposed model to help the 27 commercial banks in China to select their best merger partner.