بررسی مجدد تاثیر مشارکت بانک خارجی بر حاشیه بهره در بازارهای نوظهور

Over the past two decades, emerging markets have witnessed a considerable increase in foreign bank participation, with the anticipation that foreign entry would lower financial intermediation costs. We re-examine the impact of foreign participation on bank interest margins using data on 11 Central and Eastern European countries (CEECs), where the increase in foreign bank participation was the strongest. Using the modified dealership model of Maudos and Fernandez de Guevara (2004) as a baseline specification, we show that augmenting the model by proxies for direct and indirect impact of foreign bank participation does not produce a significant outcome. We explain our results by the fact that the dealership model fully accounts for the mechanisms through which foreign bank presence is hypothesized to affect interest margins in theoretical models on international banking. We call for re-examination of some of the previous findings showing a significant own effect of foreign bank participation on interest margins in emerging markets within the framework of the dealership model.

One of the most remarkable developments in international banking during the last two decades has been the sharp increase of foreign bank participation in emerging markets. According to IMF (2000), Central and Eastern European Countries (CEECs) have been the leaders in terms of foreign bank participation among emerging markets, followed by the Latin American countries. Fig. 1 shows that the average market share of foreign banks in 11 CEECs has risen from 14% in 1995 to 80% in 2006. At the moment, foreign banks play a dominant role in banking sectors of all CEECs, except for Slovenia. The observed massive increase of foreign bank participation inevitably raises the question to what extent foreign entry influences banking institutions and their customers in CEECs. Full-size image (30 K) Fig. 1. Market share of foreign banks in CEECs. Figure options There is a multiplicity of dimensions through which the impact of foreign entry on bank customers could be analyzed. Examples from the literature include stability of domestic banking systems (Gorton and Winton, 1998), banking system concentration and competition (Barajas et al., 2000, Martinez Peria and Mody, 2004 and Sengupta, 2007), supply and accessibility of credit (Giannetti and Ongena, 2005, Clarke et al., 2001, De Haas and Naaborg, 2006 and De Haas and Van Lelyveld, 2006), banking efficiency (Fries and Taci, 2005, Bonin et al., 2005, Poghosyan and Borovicka, 2007 and Havrylchyk and Jurzyk, 2008). This paper addresses one particular aspect of foreign bank participation — its impact on bank net interest margins in host countries. The importance of the net interest margin as a measure of financial intermediation costs in CEECs arises from the fact that in the absence of developed capital markets firms in CEECs largely depend on bank financing (Berglof and Bolton, 2002). Intuitively, foreign bank entry can have a direct impact on interest margins due to the differences in the level of interest margins set by the foreign-owned and incumbent domestic banks. Foreign bank entry may also have an indirect impact through increased competition and spillover effects. The main problem in evaluating these impacts is the lack of a consistent theoretical framework, which would enable us to disentangle the impact of bank ownership from the impact of other important bank- and country-specific characteristics (for instance, market structure). The existing theoretical literature, widely represented by different forms of the bank dealership model proposed by Ho and Saunders (1981), does not provide a consistent empirical framework to test the impact of foreign bank participation. In the absence of a theoretical framework, the empirical literature relies on various theoretical models on international banking and evaluates the impact of foreign ownership by augmenting the empirical specification of the dealership model by various indicators of foreign bank participation (e.g., number of foreign banks or foreign bank dummy variables) in addition to the theoretically-motivated determinants. The results from these studies are mixed and, among other things, depend on the extent to which all the relevant variables suggested by the dealership model are taken into account. The aim of this paper is to re-examine the impact of foreign bank participation on interest margins in emerging markets using the CEECs as a laboratory example.1 Unlike most previous studies, we start our analysis using the contemporary version of the dealership model proposed by Maudos and Fernandez de Guevara (2004), in which the interest margin is modeled as a function of theoretically-motivated determinants and the impact of bank ownership is not accounted for. Using this model as a baseline specification, we augment it by introducing different proxies for direct and indirect impact of foreign bank participation. Our estimations suggest that after controlling for the theoretically-motivated determinants, indicators of foreign bank participation do not elicit a significant impact on interest margins. Intuitively, this result suggests that both direct and indirect channels, through which the impact of foreign bank participation on margins is expected to materialize (e.g. market structure), are already taken into account by the theoretically-motivated determinants in the dealership model. Our findings call for re-examination of some of the previous studies, in which foreign bank participation was found to have a significant direct influence on interest margins in emerging markets. The remainder of the paper is structured as follows. Section 2 reviews the related theoretical and empirical literature. Section 3 describes the empirical methodology and data. Section 4 presents the estimation results and their discussion. The last section concludes.

This paper re-examines the impact of foreign bank participation on interest margins in emerging markets using the recent sharp increase of foreign bank presence in CEECs as a laboratory experiment. This topic has been widely debated by the academicians and policymakers in the aftermath of the global financial crisis, during which emerging markets having broader exposure to foreign entry have suffered the most from the capital outflow and the increase in financial intermediation costs. We start by observing that the dealership model widely used in empirical work to provide a quantitative assessment of factors driving the margin in emerging markets does not allow for the impact of foreign bank participation to be explicitly tested. The mechanisms through which foreign bank participation may influence bank behavior and ultimately the margin are analyzed by other models in a framework different from the dealership model. However, the majority of these mechanisms, like market concentration, riskiness of bank portfolio and operational costs, are already taken into account by the margin determinants inspired by the dealership model. This raises the question of whether the foreign bank participation has its own direct and/or indirect impact on interest margins. Previous empirical studies that have attempted to address this question have produced mixed results. Most of the studies report a negative effect, suggesting that foreign participation helps to decrease the margin due to spillover effects and portfolio mix of foreign banks (see for example Martinez Peria and Mody, 2004), while others did not find any significant impact, or even reported a positive impact (see for example Schwaiger and Liebeg, 2008). The mixed results in these studies can be explained by differences in the coverage theoretical determinants inspired by the dealership model entering the specification. Using data on domestic and foreign-owned banks in 11 CEECs, we show that after fully accounting for all interest margin determinants inspired by the dealership model, foreign bank participation does not have any significant impact on interest margins in CEECs. The impact remains insignificant when we differentiate between proxies for indirect (foreign bank market share) and direct (dummy variables for Greenfield and acquired foreign banks) effects of foreign bank presence. We explain this finding by the fact that the variables inspired by the dealership model already account for the main mechanisms through which the impact of foreign bank participation on the margins in emerging economies may be materialized. The main policy implication of these findings is that to contribute to the reduction of financial intermediation costs, policymakers in emerging economies need to focus on policies aimed at reducing credit and market risks, increasing bank efficiency, and improving competitive environment. Opening the borders for foreign entry would not have immediate own impact on interest margins, but can contribute to its reduction if foreign entry boosts market confidence, enhances competition, and improves performance of banks in emerging markets.