Ebook Bank failure: Evidence from the Colombian financial crisis
During the late 1990s and early 2000s, Colombia’s financial system experienced a period of financial stress, characterized by the failure of many banks and other financial institutions, as well as by the severe deterioration of the whole system’s financial health. The capitalization ratio of the system fell dramatically, as did profitability and liquidity. As a consequence of the crisis, the number of institutions, 110 in June 1998, dropped to only 57 in December 2001, after failures, mergers and acquisitions. Total assets of the financial system experienced a real contraction of more than 20 percent during the same period, making that episode of financial stress the deepest financial crisis experienced by the country in the last century.
The literature on the financial crisis of Colombia has concentrated in explaining its causes and consequences. See Arias et al (1999), Arbeláez et al. (2003), Carrasquilla and Zárate (2002), Parra and Salazar (2000), Uribe and Vargas (2002), Urrutia (1999) and Villar et al (2005). There have been no micro-level studies of the role of specific financial variables in determining failure and time to failure of banks. This paper uses duration models to characterize the failure rates of financial institutions in Colombia and to identify key financial variables associated with these failure rates. Duration models use hazard functions rather than densities to specify the distribution of observables (and thus the likelihood function). For the method, see Kiefer (1988) and Lancaster (1990). Although early economic applications of hazard functions or duration analysis were in labor economics, they have been applied to bank failures. Lane et al (1986), Weelock and Wilson (1995), and Whalen (1991), use duration models to explain bank failure in the United States. Other studies have used duration models to explain time to failure after particular episodes of financial stress in under-developed countries. For example, Gonzalez- Hermosillo et al (1996) use them to explain bank failure after the Mexican crisis of 1994, and Carree (2003) does a hazard rate analysis of Russian commercial banks in the period 1994-1997.
There are theoretical as well as practical reasons to consider that the capitalization ratio plays a special role for financial institutions. The literature on capital crunch shows that, under capital regulations, this ratio is important for financial institutions when they are taking decisions on portfolio composition. See Peek and Rosengren (1995), Estrella et al (2000), and Van den Heuvel (2004). In the practical world, following the Basel accord (Basel Committee on Banking Supervision, 2004), financial institutions and supervisors now follow closely the capital ratio of the institutions they regulate, and impose minimum requirements. Thus, capitalization plays a special role for financial institutions in determining their overall financial health and thus the degree of trouble that they might experience in episodes of financial stress. We focus on the capitalization variable and identify a nonlinear effect. As might be expected, increasing the capitalization ratio decreases the probability of default at a decreasing rate. Although capitalization is sometimes one of several significant variables, previous studies have not identified this nonlinearity, possibly due to the relative lack of data information in datasets with few failures. The data set used here is unusually rich, in two senses: First, survival is measured on a monthly scale, which helps identify more precisely the moment of failure of financial institutions. Most of the previous studies use quarterly data, which is the frequency in which financial institutions report their balances to the supervisors in many countries. Second, due to Colombia’s financial crisis, there are enough failures to identify and measure significant effects of financial variables. We expect that our qualitative results are likely to be applicable to modern banking systems generally, though we would hesitate to extrapolate numerical values of coefficients outside of our application.
Regarding the literature on the financial crisis of Colombia, this paper contributes by providing microeconomic evidence on the main variables determining bank failure, and by providing a model which can be used as an early warning tool that can be an alternative to the costly on-site visits made by supervisors to institutions considered at risk. It also provides the supervisors a basic guideline about which financial variables are important to follow during moments of stress.
Section 2 briefly describes what happened during the episode of financial crisis in Colombia. Section 3 presents the description of the data. Section 4 presents the techniques used to construct a model for the failure of financial institutions. Section 5 presents the results of the estimation, as well as empirical tests to check the validity of the model. Section 6 presents some empirical evidence using time varying regressors. Finally, section 7 concludes.
Download
PDF Ebook Bank failure: Evidence from the Colombian financial crisis
Posted in :
