Banking Industry Consolidation and Macroeconomic Di ... VI/VI.L... · PDF fileBanking...
Embed Size (px)
Transcript of Banking Industry Consolidation and Macroeconomic Di ... VI/VI.L... · PDF fileBanking...
Banking Industry Consolidation and Macroeconomic Differences∗
Insights from the Italian Case
Luca Colombo† Gilberto Turati‡
We use probit and count data (ZIP) models to study the consolidation process of the banking industry in Italy, a country characterized by wide macroeconomic differences across regions. Our empirical analysis high- lights three main findings. First, we document an important role for the interplay between real and financial variables in driving the consolidation process. Second, we emphasize the importance of competition in local banking markets, suggesting that more competition renders more likely the presence of acquiring banks. Third, we show that an excess of loans over deposits collected in a given region has an impact on the presence of acquiring banks. This last result suggests that, besides the benefits in terms of efficiency stemming from banking consolidation, there could also be some costs often overlooked.
Keywords: Banking consolidation, local development, count data models.
JEL codes: G21, G34, L16
∗We would like to thank the Bank of Italy for making available data on bank branches and Serena Caroppo for invaluable research assistance. We are grateful to participants at the 59th IIPF Congress (Prague), the XII International “Tor Vergata” Conference on Bank- ing and Finance (Roma), the XVI SIEP Conference (Pavia), and to Fabio Bagliano, Angelo Baglioni, Sijbren Cnossen, Mario Pagliero, Giovanni Petrella, Massimiliano Piacenza, Alessan- dro Sembenelli and Virginia Tirri for their comments on previous versions of the paper. Usual disclaimers apply.
†Università Cattolica, Largo A. Gemelli 1, I-20123 Milano. Phone: +39.02.7234.2637, fax +39.02.7234.2781, e-mail: email@example.com
‡Università degli Studi di Torino, Corso Unione Sovietica 218bis, I-10134 Torino. Phone: +39.011.670.6046, fax +39.011.670.6062, e-mail: firstname.lastname@example.org
In the last decades, the banking industry has undergone an unprecedented wave of M&As in many advanced economies. The process has started at different times and proceeded at different paces, beginning in the US in the Eighties, and reaching Europe in the Nineties. As a consequence of consolidation a reduction in the total number of banks and an increase in their average size has been observed almost anywhere. The Italian case is particularly interesting, however, in that the country is characterized by widespread differences in macroeconomic conditions across regions, that influence both the efficiency and the profitability of banks operating in different areas. Most of the literature dealing with banks’ consolidation has stressed the im-
portance of financial variables at the bank level as the main driver of observed M&As, with more efficient banks taking over those that are less efficient (see Focarelli et al., 2002, for the Italian case, and Berger et al., 1999, for a survey). However, this conclusion may hinder the role of other variables as engines of the consolidation process. In particular, the higher level of efficiency of acquiring banks might be due not only to better managerial choices, but also to a more favorable macroeconomic environment. In this paper, we argue that geographical factors – in particular the existence of differences in economic conditions at the local level – as well as the interplay of real and financial variables are major driving forces of the consolidation process of banks operating in distant markets. We study the factors lying behind the restructuring process by using a unique
data set on acquiring banks (active) and target banks (passive), and we show that the level of GDP, the degree of concentration of the banking industry, and the demand of financial resources within a certain area are all important factors in determining the pattern and direction of the consolidation process of the banking industry in different areas of the country. While the M&As wave that shaped the current Italian banking market is an
example of a consolidation process guided by differences in real and financial variables and market conditions within a country, we argue that our claims are of a greater generality, as our analysis can be extended to encompass situations in which the differences in economic conditions are across countries.1 In this respect, important case studies are those of the Central - Eastern European markets, where acquiring banks from more developed countries gained market shares (both in the deposits and loans markets) at the expenses of local (resident) banks. For instance, Gros (2003) reports that “by 2001 foreign banks had more than half of deposits in all of these countries, and in some of the larger ones (e.g. Poland, Hungary, Czech Republic) the share of foreign banks is around two thirds”. Along the same lines, Naaborg et al. (2003) show that in 2000, as a percentage of total
1In this sense, the present paper is close in spirit to the literature on FDI in the banking industry; see e.g. Focarelli and Pozzolo (2004).
banks assets, foreign banks assets reach an average of 64.4% (starting from 7,5% in 1994) in a sample of East European and Baltic countries.2 The same holds true for many Latin American banking industries; De Haas and van Lelyveld (2002), for example, show that in 1999 36% of total loans in Brazil were originated by foreign banks, and this percentage increases to 58% in Argentina. There is little doubt that this process can be positive from an efficiency point
of view. As foreign banks are shown to be more efficient than local banks, it is generally argued that improvements in management and credit policies following acquisition would induce efficiency gains and boost economic development.3 How- ever, it can also raise concerns for policy makers. In particular, M&As between banks operating in distant markets may originate processes of savings realloca- tion (at the expenses of credit availability at the local level), possibly determining credit rationing phenomena. This poses a serious policy issue, eventually calling for tools improving (coeteris paribus) credit availability at the local level (when evaluating consolidation proposals in the banking industry) as a way to reduce the likelihood of deposit “siphoning” and credit rationing. The remainder of the paper is structured as follows. In Section 2 we outline the
observed characteristics of the consolidation of the Italian banking industry and highlight some of its major implications. In Section 3 we present our empirical methodology, the data and our main results. Section 4 concludes.
2 The consolidation of the Italian banking in- dustry
2.1 The M&As geographical pattern: the role of regional differences
The Italian banking industry consolidation process, started in the early Nineties, is characterized by one evident feature: the great majority of active banks are located in the northern regions, while passive (target) banks are located both in the North and in the South (see Table 1). Thus, in the northern part of the country, in-market M&As are usually observed (i.e. those for which both the acquiring and the target firms are from the same area), while out-of-market operations are typical of southern regions, with a bank from the North taking over a bank located in the South. The out-of-market share of the M&As process is quite important. According to Bank of Italy data, between 1990 and 2001 the number of banks headquartered in the South reduced by more than one half, from
2For example, in 2000, Estonia foreign banks assets represent 97% of the total (from 2% in 1995). The same ratio equals 87% in Croatia (from 1% in 1996), 69% in Poland (from 3% in 1994), and 67% in Hungary (from 14% in 1994).
3See, among others, Claessens et al. (2001) and Clarke et al. (2003).
100 to 48; out of these 48 banks, 26 were owned by credit institutions located in the central-northern regions of the country (see e.g. Panetta, 2003). Furthermore, in its latest annual report, the Bank of Italy shows that approximately 70% of the loans originated in the South are from banks headquartered in Northern regions. Out-of-market M&As, dealing with geographically distant markets, are par-
ticularly interesting because of the significant differences between the degrees of development and the structural characteristics of local economies in the two areas of the country. There have always been wide macroeconomic differences between the Northern and Southern regions, with no significant tendency toward closing the gap. For instance, in 2003, GDP per capita was around 26,000 euro in the Center - North and 15,600 euro in the South. In the same year, the unemploy- ment rate was 4% in the North, 6.9% in the Center and 16.1% in the South. Such differences in the conditions of regional economies had certainly an impact on the observed dynamics of the consolidation process. Given the size of the phenomenon and its policy implications, it is important
to assess the determinants and the consequences originating from the decisions of northern banks to acquire credit institutions located in the South. The interplay of real and financial variables in explaining growth and development has been em- phasized in the literature (for instance, Levine, 1997 and 2004). We argue that a similar interplay had an important role in shaping the consolidation of the Italian banking industry, characterized by marked differences in the conditions of local (regional) economies, both in terms of real and financial variables. In order to isolate the main differences between regional economies