Foreign Ownership and Market Power in Banking: Evidence from a World Sample

Abstract

Ownership and competition in the banking sector are policy concerns around the world that are rarely comprehensively examined. For 131 countries and 13 years we match bank ownership with over 50,000 bank-year estimates of individual bank market power. At the individual bank level, ownership does not explain market power. At the country level, on the other hand, foreign bank ownership has a positive and significant impact on bank market power because foreign banks enter through mergers or acquisitions and not through greenfield investments. We also find that the positive effect of foreign bank presence on market power is considerably weaker in countries with well-capitalized banks.

Competition and risk‐taking in investment banking

“…Finally, following Delis et al. (), we test for potential non‐linear effects between competition, EALER, and risk. We did not find any evidence of such a linear effect…”

“…For recent applications of Love and Zicchino ()’s model to the banking sector see e.g. Delis, Hasan, and Kazakis (), Head et al., (); Imbierowicz and Rauch ().…”

Bank Market Power and Firm Performance

“…A popular approach has been to estimate a translog cost function and take its derivative to obtain the marginal cost. Some recent work has shown that it is possible to improve on this methodology with semiparametric or nonparametric methods that allow for more flexibility in the functional form (Delis, Iosifidi, and Tsionas, 2013;Delis, Kokas, and Ongena, 2015). As we follow the exact same approach as in Delis, Kokas, and…”

“…The variable z, which is the so-called smoothing parameter, is crucial for the identification of the model and must be a variable that is highly correlated with a 2 and considerably varies by bank-year. Delis, Kokas, and Ongena (2015) propose using , which is intuitive given the high potential correlation of input prices with the output elasticity of costs. We use the same approach and we also verify that using each input price separately yields similar results.…”

“…This procedure yields a unique multi-level sample that contains many instances of multiple loans made by a single syndicate lead bank and/or obtained by a particular borrowing firm. We estimate the market power of banks at the bank-year level using the Lerner index (1934), which measures deviations of prices from marginal cost (Beck, De Jonghe, and Schepens, 2013;Delis, Kokas, and Ongena, 2015), and capture firm performance with several variables, including the return on assets (ROA), return on income (ROI) and Tobin's q.…”