Move a Little Closer? Information Sharing and the Spatial Clustering of Bank Branches

Abstract

We study how information sharing between banks influences the geographical clustering of branches. We construct a spatial oligopoly model with price competition that explains why bank branches cluster and how the introduction of information sharing impacts clustering. Dynamic data on 59,333 branches operated by 676 banks in 22 countries between 1995 and 2012 allow us to test the hypotheses derived from this model. Consistent with our model, we find that information sharing spurs banks to open branches in localities that are new to them but that are already relatively well served by other banks. Information sharing also allows firms to borrow from more distant banks.

The structure and relations of banking systems: the UK experience and the challenges of ‘levelling-up’

“…And while fintech may have increased the efficiency of collecting and processing hard, codified information, it may have come at the expense of soft, tacit information. The public availability of hard information, such as algorithm-based credit ratings, tends to lead to the spatial clustering of banks in well-served locations, while other often smaller locations lose out as access to credit deteriorates even further (Qi et al, 2018). These results are robust after controlling for issues of endogeneity using instrumental variable and placebo regressions (Qi et al, 2018).…”

“…The aftermath of the 2008 global financial crisis led to a significant geographical restructuring of UK banking. In the UK, of the 600 bank branch closures between April 2015 and April 2016, over 90 per cent were in areas with a below median household income (Qi et al, 2018), whereas two-thirds of all bank branch openings were in wealthier neighbourhoods (Reuters, 2016). Of course, as well as bank lending, there are also other sources of investment finance, such as venture capital and angel investing, but again there is a clear divide between the nature, scale, and operation of these markets in and around London versus the rest of the country (Mason and Harrison, 2003;Mason and Pierrakis, 2013;McCann, 2016).…”

De‐branching, too‐much‐branching, and cost of debt of SMEs in Slovakia

“…As discussed in Section 3, the redesign of bank networks in Slovakia tended to follow a simple pattern by selecting the regional centres (NUTS3 level centres with socioeconomic potential), where bank branches were concentrated and drawn from other neighbouring regions. Hence, geographical aspects might have dominated the question of local competition, creating stronger spatial auto-correlation, as predicted in Qi et al (2018). The effect of the spatial dimension on the localization of bank branches is captured by the statistically significant and negative spatial autocorrelation coefficient ρ.…”

“…Additionally, multimarket banks (often foreign banks) that do not possess soft information at a scale similar to that of local banks (domestic banks) are more likely to engage in a strategy of expanding their network size over multiple markets, while local banks tend to concentrate their activities in individual local centres (Qi et al, 2018). As relationship lending is often associated with improving the probability of credit approval, yet at the expense of increasing its price (Agarwal & Hauswald, 2010; Howorth & Moro, 2012), markets populated with local banks might face an increase on average cost of debt, while expecting to be compensated by the provision of less expensive additional services (Degryse & Cayseele, 2000) or services for which proprietary soft information is less needed (Gobbi & Lotti, 2004).…”

Financial and Economic Development: impacts of de-branching on the rural population in the south of Brazil

“…There are in the recent literature many studies trying to analyse these effects. Most of them focused on the determinants of the spatial location of branches (Okeahalam, 2009;Acedanski & Karkowska, 2022;Wang & Guan, 2017;Qi et al, 2019), while others focused on the impacts in specific agents as SMEs (Rafaj & Siranova, 2022;Flogel & Gartner, 2018;Zhao & Jones-Evans, 2017;Hasan et al, 2017) or rural population (Ansong et al, 2015;BGFRS, 2019).…”