Bank Size, Information Sharing and Financial Access in Africa
International Journal Managerial Finance, 14 (2), pp.188-209 (April, 2018)
AGDI Working Paper, WP/17/044
30 Pages Posted: 23 Oct 2017 Last revised: 1 May 2018
Date Written: October 22, 2017
Purpose - This study investigates how bank size affects the role of information asymmetry on financial access in a panel of 162 banks in 39 African countries for the period 2001-2011.
Design/methodology/approach - The empirical evidence is based on instrumental variable Fixed Effects regressions with overlapping and non-overlapping bank size thresholds to control for the QLH (Quiet Life Hypothesis). The QLH postulates that managers of large banks will use their privileges for private gains at the expense of making financial services more accessible to the general public. Financial access is measured with loan price and loan quantity whereas information asymmetry is implicit in the activities of public credit registries and private credit bureaus.
Findings - The findings with non-overlapping thresholds are broadly consistent with those that are conditional on overlapping thresholds. First, public credit registries have a decreasing effect on the price of loans with the magnitude of reduction comparable across all bank size thresholds. Second, both public credit registries and private credit bureaus enhance the quantity of loans. Third, compared with public credit registries, private credit bureaus have a greater influence in increasing financial access because they have a significantly higher favourable effect on the quantity and price of loans Fourth, the QLH is not apparent because large banks are not associated with lower levels of financial access compared to small banks.
Originality/value - Studies of public credit registries and private credit bureaus in Africa are sparse. This is one of the few to assess linkages between bank size, information asymmetry and financial access.
Keywords: Public goods; Financial access; Bank size; Information sharing
JEL Classification: G20; G29; L96; O40; O55
Suggested Citation: Suggested Citation