Are Foreign Banks Better Performers Than Domestic Counterparts? Examining the Financialization Hypothesis in Mexico

ABSTRACT

This article analyzes the relationship between “financialization” and the performance of credit institutions in Mexico over the period 1999–2013. On average, foreign banks have higher cost-to-income ratio and return on assets than domestically owned banks. Non-interest income to gross revenue positively contributes to foreign bank profits, which highlights income other than conventional banking transactions. When considering references in the literature to non-interest income as a proxy for financialization, our results imply that domestic banks are more dependent on traditional lending, such as interest-earning assets, whereas foreign banks seem to engage in greater diversification to support profitability. Overall, the banking system reveals tendencies (“financialization”) that can possibly work against credit facilitation in Mexico.

RESUMEN

Este estudio analiza la relación entre la “financialización” y el desempeño de las instituciones crediticias en México durante el período 1999–2013. El promedio de la banca extranjera tiene una razón costo-ingreso más alta sobre los activos que la banca doméstica. Además, también se observa una contribución positiva de los rendimientos no financieros en relación a los ingresos netos en la utilidad de la banca extranjera, que destaca el ingreso en lugar de las transacciones convencionales de la banca. Al considerar las referencias encontradas en la literatura sobre el ingreso no financiero como representativo de la financialización, nuestros resultados implican que la banca doméstica depende mucho más de las operaciones de préstamo tradicionales, tales como activos que devengan intereses, mientras que la banca extranjera parece dedicarse a ampliar su diversificación para respaldar su rentabilidad. En general, el sistema bancario muestra tendencias (“financialización”) que pueden funcionar contra la facilitación de crédito en México.

RESUMO

O presente trabalho analisa a relação entre “financeirização” e o desempenho das instituições de crédito no México, no período 1999–2013. Em média, os bancos estrangeiros apresentam maior relação custo-receita e retorno sobre os ativos do que os bancos domésticos. Observa-se uma contribuição positiva dos rendimentos não decorrentes de juros em relação à receita bruta no lucro dos bancos estrangeiros, o que confere mais destaque à receita do que as transações bancárias convencionais. Quanto à apreciação das referências da literatura sobre os rendimentos não decorrentes de juros como um aliado da financeirização, os nossos resultados indicam que os bancos domésticos dependem mais de empréstimos tradicionais como ativos com rendimentos decorrentes de juros, enquanto os bancos estrangeiros empregam, aparentemente, uma diversificação maior como respaldo para a rentabilidade. De forma geral o sistema bancário mostra tendências (“financeirização”), que possivelmente podem trabalhar contra a facilitação de crédito no México.

KEYWORDS:

• foreign banks in latin america
• financialization hypothesis
• banking and economic development
• productive finance
• financial sector in emerging markets

ACKNOWLEDGMENTS

We are thankful to the anonymous peer reviewers who helped contribute to the later versions of this paper with their productive feedback and suggestions for improvement.

Notes

For an extensive review of literature on bank performance determinants as well as seminal studies on developed and developing countries, see Gilbert (1984); Bourke (1989); Molyneux and Thornton (1992); Berger (1995); Goldberg and Rai (1996); Demirguc-Kunt and Huizinga (1999, 2000); Molyneux, Goddard, & Wilson (2004); and Goddard, Molyneux, Wilson, & Tavakoli (2007), to name a few.

For an extensive analysis of banking sector performance in Latin America along multiple dimensions, such as cost efficiency, profitability, net interest margins and volatility, see Barajas, Steiner, & Salazar (2000); Brock and Suarez (2000a); Martinez-Peria and Mody (2004); Carvallo and Kasman (2005); Chortareas et al. (2011, 2012); Garza-Garcia (2012a, 2012b); and most recently Guillen et al. (2014), Jara-Bertin et al. (2014), to name a few. For an additional review of this literature, see also Doyran and Erdogan (2013).

See especially Barajas et al. (2000), Saunders and Schumacher (2000), Brock and Suarez (2000a; 2000b); Afanasieff, Lhacer, & Nakane (2002); Gelos and Roldos (2002); Martinez-Peria and Mody (2004); Gelos (2006); Williams, Cardim de Carvalho, & De Paula (2009); Garza-Garcia (2012b).

For example, social cost of market power (“loss of economic welfare”) was 0.15% of GDP, while cost and profit inefficiency of bank management was 0.021% and 0.075% of GDP respectively (Maudos & Solis, 2008, p. 467). Furthermore, Garza-Garcia's (2012b) analysis validates the Relative Market Power (RMP) hypothesis for the 18 largest banks in Mexico during the period 2001–2009. His findings indicate that Mexican banks enjoy “profit persistence” due to high market shares, high capitalization, and monopolistic competition. The persistence in profits is associated with “less competitive industry with very slow adjustment to normal long-run profits” and “important barriers to competition” (Garza-Garcia, 2012b, p. 29).

For an extensive discussion on “relationship lending,” banking credit-growth nexus and informational asymmetries in bank lending in developing countries, see especially, Clarke, Cull, & Martinez-Peria (2006); Focarelli and Pozzolo (2000); and Berger and Udell (2001).

As Figure 2 shows, Mexican banks’ holdings of non-interest income as a share of gross revenue (NINGR) increased from 34.89% in 1999 to 38.95% in 2013. Within domestic banks, NINGR increased from 27.54% in 1999 to 34.91% in 2013. Within foreign banks, NINGR increased from 40.61% to 44.68% during the same period.

This is a global electronic publishing maintained by Fitch/IBCA/Bureau Van Dijk (Bankscope, 2014).

Models with heteroscedasticity (non-robust standard errors) were provided to editors.

A widely used proxy variable, an industry concentration, was omitted due to difficulty of finding a suitable proxy for competition/concentration (and potential endogeneity problems with Herfindahl-Hirschman Index (HHI) when the sum of squared market shares of each bank's assets are regressed against Return on Average Assets).