Leverage Pro-cyclicality and Bank Balance Sheet in Colombia

Leverage Pro-cyclicality and Bank Balance Sheet in Colombia* Prociclicidad del apalancamiento y balances de los bancos en Colombia Franz Hamanna,**, Rafael Hernándezb, Luisa Silvac, and Fernando Tenjo G.d * The views expressed in this document are those of the authors and not necessarily those of the Banco de la República. Of course, any mistake in this paper is our responsibility. a Advisor to the Governor, Banco de la República, Bogotá, Colombia b Research Assistant, Centro de Estudios sobre Desarrollo Económico (CEDE), Universidad de los Andes, Bogotá, Colombia c Head of the Financial Sector Unit, Banco de la República, Bogotá, Colombia d Ex-Director, Monetary Policy Board, Banco de la República, Bogotá, Colombia ** Corresponding author. E-mail address: (F. Hamann). History of the article: Received May 16, 2013 Accepted January 28, 2014 ABSTRACT The recent financial crisis has renewed the interest of economists, both at the theoretical and empirical level, in developing a better understanding of credit and its mechanisms. A rapidly growing strand of the literature views banks as facing funding restrictions that condition their borrowing to a risk-based capital constraint which, in turn, affects bank lending. This work explores the way banks in Colombia manage their balance sheet and sheds light into the dynamics of credit and leverage over the business cycle. Using a sample of monthly bank balance sheets for the period 1994-2011, we find not only that leverage is predominantly pro-cyclical in the Colombian banking sector, but also that heterogeneity matters, and thus, an aggregate measure of bank leverage can mask a fragile financial sector. In addition, although some banks display great dynamics on the right-hand side of their balance sheet during the upward phase of the leverage cycle, changes in the composition of liabilities between core and non-core do not seem to have a clear pattern. Still, more attention should be paid on this by policy makers, as these dynamics could convey information about the phase of the cycle of the economy and the financial vulnerability of the system as a whole. Keywords: Banks, Credit, Leverage, Non-core liabilities, Balance sheet, Business cycle, Colombia. JEL Classification: E44, E52, G21, E32, G21, G32. RESUMEN La reciente crisis financiera ha renovado el interés de los economistas, tanto a nivel teórico como empírico, en E32 desarrollar un mejor entendimiento del crédito y sus mecanismos. Una rama de rápido crecimiento en la G21 literatura considera que los bancos enfrentan restricciones de financiamiento que condicionan su endeudamiento a una restricción de capital basada en el riesgo que asumen, y que a su vez, afecta la oferta de crédito. Este trabajo explora la forma en que los bancos en Colombia manejan sus balances y arroja luz sobre las dinámicas del crédito y el apalancamiento durante el ciclo económico. Utilizando una muestra de balances mensuales de los bancos para el período 1994-2011, encontramos no sólo que el apalancamiento del sector bancario colombiano es predominantemente procíclico, sino también, que la heterogeneidad importa, y por Pasivos non-core tanto, una medida agregada de apalancamiento puede estar ocultando un frágil sistema financiero. Además, Hoja de balances aunque algunos bancos muestran una gran dinámica en el lado derecho de sus balances durante la fase Ciclo económico ascendente del ciclo de apalancamiento, los cambios en la composición de los pasivos entre core y non-core no parecen tener un patrón claro. Aun así, se debe prestar más atención a este tema por parte de los hacedores de política, ya que estas dinámicas pueden proveer información sobre la fase del ciclo de la economía y la vulnerabilidad financiera del sistema como un todo. Palabras clave: Bancos, Crédito, Apalancamiento, Pasivos non-core, Hoja de balances, Ciclo económico, Colombia. Clasificación JEL: E32, G21, G32. 1. Introduction This paper explores the way banks in Colombia manage their balance sheet and sheds light into the dynamics of credit. The idea is to see whether the link between credit dynamics, leverage and liability composition explains credit supply decisions. Evidence in this direction would help to understand credit fluctuations, identify possible signs of pro-cyclicality, and advance in the elaboration of a more appropriate view of the banking sector. This view is centered in the structural relationships between the two sides of the balance sheet, and would be a considerable improvement from the traditional interpretation that goes, mechanically, from money to credit. By using bank-level data, this work pays particular attention to the importance of heterogeneity within the banking system and the role it plays in the evolution of credit in the economy. The recent financial crisis has renewed the interest of economists, both at the theoretical and empirical level, in developing a better understanding of credit and its mechanisms. A growing number of studies, drawing on a tradition that underscores the inherent instability of credit systems, show that credit lies at the heart of financial crises and that the latter may be the endogenous outcome of how credit is created in the context of decisions of numerous and heterogeneous agents (see Aikman et al., 2011, Jorda et al., 2011, and Taylor, 2012). Very briefly, the inherent instability of credit results from the feedback between credit fluctuations and changes in collateral prices. This relationship is best approached in terms of what is known in the literature as leverage cycles (Geneakoplos, 2009; Jorda et al., 2011). Various strings and lines of research can be pursued in order to test the validity of these ideas and try to find empirical support for them under different scenarios. Among such lines, the present paper follows recent strands of the literature that directly connect the dynamics of credit to the behavior of banks, going beyond mere quantities and looking at their entire balance sheet, including assets, liabilities, their composition, and leverage (Shularick and Taylor, 2010). More specifically, recent interesting work builds on a formal model in which financial intermediaries manage their balance sheet in a way that is consistent with, and responds to, their credit supply decisions (see Adrian and Shin, 2010, Adrian and Boyarchenko, 2012, Adrian and Shin, 2011, and Adrian and Shin, 2012). The model in question is, briefly put, a model of credit supply and credit risk, where a bank maximizes profits subject to a value-at-risk constraint. This means that banks, and financial intermediaries in general, face funding restrictions that condition their borrowing to a risk-based capital constraint which, in turn, affects bank lending. Changes in the size and composition of balance sheets are derived from credit decisions taken by banks. Thus, there is a "lending" or "balance sheet capacity (...truncated)