It is widely recognized that most economic activities show a tendency to agglomerate in space and, therefore, economic interactions among firms are likely to be influenced by their geographical proximity. Recently, the literature (Boschma, 2005) has remarked that other dimensions of proximity among economic agents, in addition to the spatial one, can play a relevant role in shaping their relationships. Thus, a growing number of contributions are investigating how proximities among agents - in terms of cognitive and technological bases, cultural and institutional similarity and organizational background - affect their exchanges. In this study we focus on a specific form of firms' inter-relationship - Mergers and Acquisitions (M&A) - which represents one of the most effective tools used by firms to achieve their strategic goals (increasing market shares, entering new markets, acquiring new competencies). M&A deals represent an interesting set-up to assess the decisive role of proximities. The acquisition of a potential target involves a prolonged and costly search process, often conditioned by the limited amount of managerial resources available to the acquiring firm because of physical, cognitive and monetary constraints. Thus, on average, only the small proportion of the most proximate and similar targets - in terms of geographical location, technological relatedness and organizational factors - are evaluated by the acquirer in the search process. In this paper we investigates the effect of geographical, industrial, organizational and institutional proximity on the probability that any two firms located in Italy engage in a M&A deal. Within a logistic rare event framework, we investigate 4,261 actual deals completed over the period 2000-2011 and around 3.8 million potential deals. We find robust evidence that all forms of proximity have a positive effect, especially industrial relatedness. Moreover, we find evidence that proximities generate asymmetric effects on M&A deals, depending on the location of bidders and targets and on whether some specific individual characteristics are featured by the acquirer or by the target firm. More specifically, we find that the probably decreases when the acquirer is located in the North of Italy and the target in the South, while the opposite case does not affect the estimated baseline probability; this is reasonably due to the much lower level of social capital featured by Southern Italian regions with respect to the Northern ones, which increases uncertainty and transactions costs, making more difficult the search process for Northern acquirers targeting Southern firms. An additional positive (negative) effect is associated to the acquirer (target) belonging to the financial sector and the target (acquirer) to the manufacturing one; finally, an additional effect is found when the target is listed and the acquirer is a private firm, whereas the reverse status pattern does not yield a significant effect.
