The effects of bank market power in short-term and long-term firm credit availability and investment

ABSTRACT

This article investigates the short-term and long-term effects of bank market power on the availability of credit for companies, and on firm investment. Our results suggest that an increase in bank market power reduces firms’ credit availability and investment in the short-term, but firm investment recovers in the long-term. The economic significance of these relationships is found to be larger for SMEs than for other (larger) firms.

RESUMEN

Este artículo investiga los efectos del poder de mercado bancario a corto y largo plazo sobre la disponibilidad de crédito para las empresas, y sobre la inversión corporativa. Nuestros resultados sugieren que un incremento del poder de mercado bancario reduce la disponibilidad de crédito y la inversión de las empresas a corto plazo, pero a largo plazo la inversión empresarial se recupera. Se descubre que la importancia económica de estos resultados es mayor para las pequeñas y medianas empresas que las (otras) grandes empresas.

KEYWORDS:

• Bank loans
• bank market power
• Euler equation
• firm investment rate
• risk premium

PALABRAS CLAVE:

• Crédito bancario
• poder de mercado bancario
• ecuación de Euler
• tasa de inversión empresarial
• prima de riesgo

SUBJECT CLASSIFICATION CODES:

• G21
• G31
• D40

Acknowledgements

Financial support from MICINN-FEDER ECO2015-67656-P, ECO2014-59584 and Junta de Andalucía P12.SEJ.2463 (Excellence Groups) is gratefully acknowledged. We would like to express our gratitude for the comments provided by Wolfgang Drobetz, Nikolaos Georgantzis, Robert M. Feinberg, Miguel Ángel Martínez Sedano, Juliette Milgram, Scott Miller, Philip Molyneux, Jonathan Williams, and John Wilson, as well as from participants at the XX annual International Industrial Organization Conference held in Arlington, Virginia, the EFMA Annual Meeting 2012 held in Barcelona (Spain), the FMA Annual Meeting 2012 held in Atlanta (Georgia), and the Jornadas de Economía Industrial held in Murcia (Spain). We also thank two anonymous referees for his/her thorough review, comments, and suggestions, which significantly contributed to improving the quality of this paper. José Manuel Mansilla-Fernández acknowledges the Postdoctoral UniCredit Fellowship.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Berger, Demirgüç-Kunt, Levine, and Haubrich (2004) offer an extensive overview of the effects of bank concentration on firm financing, particularly for the case of SME financing, and offer a future research agenda as well.

2. Recent empirical papers have shown that the comparative advantage of large banks in hard information technologies do not appear to be monotonically increasing with firm size (see Berger & Black, 2011; Berger, Espinosa-Vega, Frame, & Miller, 2005; Berger, Frame, & Miller, 2005; Frame, Srinivasan, & Woosley, 2001).

3. See also Goddard and Wilson (2009) and Goddard, Molyneux, Wilson, and Tavakoli (2007), 2011) for a complete overview of New Industrial Organization approaches as profit hypotheses, as well as different methodological aspects.

4. The acronyms correspond to the Spanish denominations: Asociación Española de Banca (AEB), Confederación Española de Cajas de Ahorros (CECA), and Unión Nacional de Cooperativas de Crédito (UNACC).

5. We thank this suggestion to an anonymous referee and the editor.

6. Recall that substituting (I/K)_it = (I/K)_it−1 + Δ(I/K)_it, and Lerner_ijt = Lerner_ijt−1 + ΔLerner_ijt yields the error-correction mechanism equivalent to the ADL model proposed in model (3). Additionally, in the error correction mechanism is the adjustment of (I/K) to equilibrium deviations in the previous period, (I/K)_it−1 − βLerner_ijt−1.

7. See Huang, Shi, and Zhang (2011).

8. This treatment eliminates the most common source of endogeneity, although it would be not completely eliminated if errors are correlated over time. We include Sargan test as well as the AR(2) and the AR(3) tests to confirm the validity of the instruments used in our estimations.

9. Endogeneity test is computed, like the C statistic, as the difference of two Sargan–Hansen statistics in which the first model is treated as an equation with a smaller set of instruments in which the suspect regressor/s is treated as endogenous, and the second model is treated as a large set in which the suspect regressor/s is treated as exogenous.

Additional information

Funding

This work was supported by the Consejería de Economía, Innovación, Ciencia y Empleo, Junta de Andalucía: [Grant Number P08-SEJ-03781]; Ministerio de Sanidad, Servicios Sociales e Igualdad(MICINN)- FEDER [Grant Number ECO2011-27954].