بحران بانکی و نظم و انضباط در بازار: شواهد بین المللی

This paper analyzes the effect of banking crises on market discipline in an international sample of banks. We also evaluate how bank regulation, supervision, institutions, and crisis intervention policies shape the effect of banking crises on market discipline. We control for unobservable bank, country, and time specific effects using a panel data set of banks from 66 countries around 79 banking crises. The results suggest that on average market discipline weakens after a banking crisis. This weakening is higher in countries where bank regulation, supervision, and institutions promoted market discipline before the banking crisis, and where a more accommodative approach is adopted to resolve it.

Market discipline is one of the three pillars generally accepted by regulators and scholars to limit the bank risk-shifting incentives that are exacerbated by financial safety nets. Basel II explicitly emphasizes the strengthening of market discipline (Pillar 3) as well as official supervision (Pillar 2) and capital requirements (Pillar 1) as tools to improve bank stability. The current global financial crisis and adoption of accommodative interventions, however, prompt a debate on the incentives depositors have to discipline bank risk-taking after a crisis. We address this issue by asking three questions: How does a banking crisis change the market discipline that depositors impose to control bank risk-taking? To what extent do variations in market discipline after a banking crisis depend on a country’s bank regulation, supervision, and institutions? Do the types of intervention and resolution policies that governments adopt during a banking crisis change market discipline?

Our analysis of the effect of banking crises on market discipline in a sample of 79 banking crises over 1989–2007 uses an international bank dataset to evaluate how changes in market discipline depend on a country’s regulation, official supervision, institutions, and crisis intervention policies. We apply the GMM difference estimator to an international panel data set of 2593 banks in 66 countries to control for potential endogeneity of the explanatory variables and unobservable bank-, country-, and time-specific effects.