An Assessment of the Competition in the Banking Industry: Empirical Evidence from Argentina with Data at Bank Level

In the last years, the financial markets have been affected by many events: deregulation, liberalization, globalization and many financial and technological innovations. Each of these events has influenced the condition of competition faced by the banks. In addition, as market power has influenced the profitability of the banks, the competitive behavior has also had impact on the efficiency and stability of the financial system. A healthy competition and a solid market structure are important factors for social welfare, which is reflected in a lower interest rate and a fluid financing for consumers and firms, especially small and medium firms. Market imperfections generated a deficient resources allocation and also a reduction in the benefits that the society derived from the use of the banking system. The literature on the measurement of competition in the banking sector has two branches: a structural approach and a non-structural approach. The structural approach included the Structure-Conduct-Performance paradigm and the efficiency hypothesis, and also many formal approaches with roots in the Industrial Organization theory. These approaches focus on whether a high concentrated market generates a collusive behavior in the big banks that results in a superior market performance or if it is the efficiency of the big banks what increases their performance. In response to the theoretical and empirical deficiencies of the structural models, the non-structural approach for evaluating the competition was developed. Among these models, the main ones are: Iwata’s model (Iwata, 1974), the Bresnahan’s model (Bresnahan, 1982, 1989) and the Panzar-Rose model (Panzar y Rose, 1974). The approach of the New Empirical Industrial Organization tests the competitive behavior and the use of market power by the banks in absence of structural measures. The loan market is related to the traditional activity of banks, lending, which is related to assets of the balance sheet. Taking into account the data requirement of the Bresnahan’s model, the election of loans is also determined by the availability of information on prices and quantities in that market: the amount of loans and interest rates. The Argentine economy between 1991 and 2001 worked under the incentives provided by a fixed exchange rate regime with convertibility of the domestic money known as Plan de Convertibilidad. This regime collapsed at the end of 2001, submerging the Argentine economy into a deep economic and financial crisis. The GDP per capita dropped by 10.8%, and loans and deposits –measured as percentage of GDP– dropped by 0.75% and 2.83% between 2001 and 2002, respectively. The financial sector began to improve at the beginning of 2003 when total deposits, in constant values, began to increase, while total loans, in constant values, began to rise only in 2004. As a consequence of the crisis in the period 2002-2007, fourteen banks went bankrupt, which represented the 14% of the total banks in operation. However the concentration –measured by the Herfindahl-Hirschmann Index (HHI)– in the loan markets was reduced. In this paper we used the methodology suggested by Bresnahan (Bresnahan, 1982) for evaluating if the reduction in the number of banks in the period 2002- 2007 affected the degree of competition of the banking industry in the loans market. With this purpose, we estimated a system of equations of supply and demand for loans for a set of banks, private and public, that operated in Argentina between 2002 and 2007. It must be taken into account that the results from the estimation show both the market structure and the average conditions of competition in the period analyzed. The results of the estimation allow us to say that banks used the market power they had to set their active interest rates. In the loan markets, the coefficient estimated value that shows the use of the market power by standard banks indicated that the banks set their active interest rate at 29.1% over their marginal costs in the period 2002-2007. This value is smaller than the value reported by Delfino (Delfino, 2002) for the period 1993- 2000 –40.3%–. This points out that banks reduced the use of their market power in the loan markets.