![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
Abstract
Colombo L. V. A. and Turati G. Why do acquiring banks in mergers concentrate in well-developed areas? Regional development and mergers and acquisitions (M&As) in banking, Regional Studies. The paper focuses on all mergers and acquisitions (M&As) transactions of regional relevance that occurred in the Italian banking sector between 1995 and 2006. A strong direct effect of regional economic and social characteristics on the concentration of the banking industry in Italy is found, as well as on the agglomeration of acquiring banks in well-developed regions. This effect survives to a number of robustness checks, including controls for banks’ profitability and efficiency, and for their institutional characteristics, indicated by the banking literature as the key factors driving concentration in banking. Possible theoretical explanations supporting the empirical findings are also investigated, and their policy implications are discussed.
Colombo L. V. A. and Turati G. 为什么银行业并购集中在发达地区？区域发展与银行业并购. 区域研究. 本文研究1995-2006年发生在意大利银行业的所有具有区域意义的并购活动，发现区域经济和社会特征对意大利银行业的集中具有很强的直接影响，对银行收购活动集聚在发达地区也有直接影响。经过稳健性检验这种影响仍然存在，包括控制银行的利润率和效率，以及控制银行的制度特征，这两点被银行研究文献视为推动银行业集中的关键因素。本文还探讨了支撑这些经验性研究发现的可能理论解释，及其政策含义。
Colombo L. V. A. et Turati G. Pourquoi les banques acquéreuses se concentrent-elles au moment des fusions dans les régions plus développées? L'aménagement du territoire et les fusions-acquisitions du secteur bancaire, Regional Studies. L'article porte sur toute opération de fusion-acquisition d'envergure régionale qui a eu lieu dans le secteur bancaire italien entre 1995 et 2006. Il s'avère que les caractéristiques économiques et sociales régionales influent sensiblement sur la tendance à la concentration du secteur bancaire italien, aussi bien que sur l'agglomération des banques acquéreuses dans les régions plus développées. L'effet satisfait plusieurs tests de robustesse, y compris des tests de la rentabilité et de l'efficacité des banques, et de leurs caractéristiques institutionnelles, indiquées dans la documentation relative aux banques comme les forces motrices de la concentration bancaire. On examine aussi des explications théoriques éventuelles qui soutiennent les résultats empiriques, et on discute de leurs conséquences pour la politique.
Colombo L. V. A. und Turati G. Warum konzentrieren sich die erwerbenden Banken bei Fusionen in gut entwickelten Gebieten? Regionalentwicklung und Fusionen und Übernahmen im Bankwesen, Regional Studies. Im Mittelpunkt dieses Beitrags stehen sämtliche Transaktionen im Zusammenhang mit Fusionen und Übernahmen von regionaler Relevanz im italienischen Banksektor zwischen 1995 und 2006. Wir stellen eine starke unmittelbare Auswirkung der regionalen wirtschaftlichen und sozialen Merkmale auf die Konzentration der Bankbranche in Italien sowie auf die Ballung der erwerbenden Banken in gut entwickelten Regionen fest. Diese Auswirkung bleibt auch nach einer Reihe von Robustheitsprüfungen bestehen, darunter eine Kontrolle der Rentabilität und Effizienz der Banken sowie ihrer institutionellen Merkmale, die in der Bankenliteratur als zentrale Faktoren für die Konzentration des Bankwesens genannt werden. Ebenso erörtern wir mögliche theoretische Erklärungen für die empirischen Ergebnisse sowie deren politischen Auswirkungen.
Colombo L. V. A. y Turati G. ¿Por qué los bancos adquirentes en fusiones se concentran en zonas bien desarrolladas? Desarrollo regional y fusiones y adquisiciones en la banca, Regional Studies. El tema de este artículo son las transacciones de fusiones y adquisiciones de relevancia regional en el sector bancario italiano entre 1995 y 2006. Observamos un fuerte efecto directo de las características económicas y sociales a nivel regional en la concentración de la industria bancaria en Italia, así como en la aglomeración de bancos adquirentes en regiones bien desarrolladas. Este efecto sobrevive a una serie de comprobaciones robustas, incluyendo los controles de la rentabilidad y la eficacia de los bancos, así como de sus características institucionales, indicadas por la bibliografía bancaria como los factores principales que llevan a la concentración en la banca. También analizamos las posibles explicaciones teóricas que apoyan los resultados empíricos y debatimos sus implicaciones políticas.
Acknowledgements
The authors thank the Editor and three anonymous referees for very useful comments, which greatly helped to improve the paper. Earlier versions of this paper were presented at the 35th Conference of the European Association for Research in Industrial Economics – Toulouse, France (2008) and at the 65th Congress of the International Institute of Public Finance – Cape Town, South Africa (2009). The authors are grateful to seminar and conference participants for their useful comments. The usual disclaimers apply.
Notes
1. The analysis presented here is related to Colombo and Turati (Citation2004, Citation2008), in which the determinants of M&As in the Italian banking industry are investigated, focusing on a shorter time period (1995–2000) and, more importantly, not controlling for regional social characteristics. As a consequence, these previous papers cannot properly identify the role played by local socio-economic conditions on the consolidation process of the banking industry, which is the main contribution of the present paper.
2. For an analysis of the Italian case, see, for example, Focarelli et al. Citation(2002); for a comprehensive survey, see Berger et al. Citation(1999).
3. All this may contribute to exacerbate the differences in socio-economic development between different geographical areas, in a cumulative process of circular causation (for example, Gärtner, Citation2009). Not surprisingly, several economists have explored the links between financial development and economic growth (for example, Levine, Citation1997), and the effects of social capital on financial development and particularly on banks’ efficiency (for example, Pastor and Tortosa-Ausina, Citation2008).
4. A conceptually equivalent exercise can obviously be conducted on target banks, obtaining qualitatively identical results. Details are available from the authors upon request.
5. The vast majority of acquiring banks have been concentrated in a relatively small subset of regions: seventy in Lombardia, fifty in Emilia Romagna and forty-three in Veneto. In all other regions there have been fewer than ten active banks over the entire period considered. In some regions (Valle d'Aosta, Trentino-Alto Adige, Abruzzo, Molise, Campania, Basilicata and Calabria) there have been no acquiring banks at all.
6. It is worth noting that the results are fully consistent with those that would be obtained by resorting to other count data models, such as the two-stage Heckman selection model. The choice to concentrate on the ZIP model was taken based on its better empirical performance according to the Akaike information criterion. On the use of this criterion for the choice among count data models, see Cameron and Trivedi Citation(1998).
7. Excluded from the analysis are all intra-group operations and all operations involving banks, the activity of which (before the merger or the acquisition) had a national extent, as the focus is on the effects of local economic conditions on banking consolidation. This last limitation concerns, however, very few operations (only thirteen) over the sample period, and it bears no impact on the results, even though it is certainly relevant in terms of intermediated resources.
8. Several papers have emphasized the importance of tacit knowledge in informing actions to explain the relevance of proximity (for example, Storper and Venables, Citation2004, Gertler, Citation2003). As noted by Engelen and Faulconbridge Citation(2009), ‘learning always involves the development of capabilities that remain tacit and which are developed through complex processes of socialization in everyday human action’ (p. 591). Along the same lines, Hall and Appleyard Citation(2009) – focusing on investment banks operating in London – show that bankers’ practices are influenced by the educational spaces and the communities in which they learn to do finance. Similarly, Faulconbridge and Muzio Citation(2009) show that the restructuring of large law firms to enhance their financial performance is influenced by the geographical context and institutional legacies.
9. A similar argument applies to the measure of average profits in the market for loans (MKUP) and to that of average profits in the market for deposits (MKDWN), providing more precise information on the degree of competition in the loans and deposits markets, respectively. The traditional mark-up index, MKUP, is defined as the difference between the average market rate on loans and the average rate on short-term government bonds. Correspondingly, the mark-down index, MKDWN, represents the difference between the average rate on short-term government bonds and the average market rate on deposits.
10. The variable DIFF was further split into its components, namely loans (LOANS) and deposits (DEP) measured in per capita terms, explicitly to account for the (potentially) different roles of loans and deposits as determinants of banking consolidation.
11. It is found that the coefficient of VCI is negative and statistically significant. Interestingly, Garmaise and Moskovitz Citation(2006) reach the opposite conclusion, arguing that M&As in banking determine an increase in crime rates by reducing the supply of loans at the local level.
12. The results of the auxiliary regressions are not reported here for brevity. They are available from the authors upon request.
13. Note that although the signs of all variables remain unchanged between the two approaches, the coefficients obtained in the panel specifications are not immediately comparable with those emerging from the probit models in , as the latter measure marginal effects. It is also worth noting that all the results obtained for the random effects panel probit models hold true for fixed-effects panel logit specifications as well.
14. Observe that the signs of all coefficients are reversed with respect to those of the probit models. This is due to the fact that the selection model in the ZIP specification identifies the probability of observing a regime in which the phenomenon under study (in this case, an M&A) cannot occur.
15. The authors also experimented with the same set of variables used for the selection equation, finding, however, that they are ill-suited to solve the problem at hand.
16. Census data are the only direct and publicly available Italian regional data on the size distribution of firms. Although it is possible to derive indicators providing specific information for all the years included in the sample, the authors abstract from doing so as the purpose here is just to explore a possible causal mechanism behind the main point of highlighting the existence of a direct effect of the local business environment.
17. According to practitioners, the price of target banks depends positively on deposits, as they represent both an opportunity to enlarge the customer base and to sell new products, and as a source of funds at a possibly low marginal cost. For an analysis of the motivations behind takeovers in the Italian banking industry, see, for example, de Vincenzo et al. (Citation2005).
18. More generally, there is a debate in regional finance (for example, Klagge and Martin, Citation2005) asking whether the spatial organization of financial markets has an impact on the flows of capital to small firms across regions.