Which Relatedness Drives Synergies Creation? Evidence from Stock Return Correlations Using Machine-learning Tools

Abstract

We find that bidder-target stock return correlations positively predict merger synergies, and the predictive power is enhanced when we focus more on idiosyncratic correlations. A measure of idiosyncratic correlation (IDC) between bidder and target firms that controls for all other stocks, constructed using machine learning tools, is the strongest predictor of merger synergies among the correlation measures. Case analyses show that IDC captures bidder and target firms’ unique product relatedness. Using IDC and the existing measures of product relatedness, we find that, consistent with theory, unique relatedness has a much stronger effect on merger synergies than general relatedness.

Keywords: Distinct Relatedness, Synergies, Mergers and Acquisitions, Conditional Dependence, Lasso, Stock return comovement

JEL Classification: G34, G30