بهره وری هزینه در سیستم بانکداری آمریکای لاتین و کارائیب

A common cost frontier with country-specific environmental variables was estimated for a panel of 481 banks from 16 Latin American countries. A stochastic frontier model was used to estimate cost inefficiency and scale and scope economies. The results suggest that there is a wide range of inefficiency levels across countries. Very small and very large banks are significantly more inefficient than large banks. Underperforming banks tend to be smaller, to be undercapitalized, to present a poor profit performance, to be more dependent on non-interest income, to be more risky, to have a less stable deposit base and to intermediate less.

In recent years, the liberalization of financial markets at a global scale, the increasing use of advanced technology, and the information revolution have put competitive pressure on banking firms both domestically and internationally. This competitive pressure is particularly important for emerging markets, where banks constitute the main financial intermediaries to channel savings and investment. Due to these developments in the banking sectors, banks are trying to find ways to enhance their efficiency, and to exploit scale and scope economies. Although entry of foreign banks was not unknown to emerging markets, the last decade witnessed a wave of foreign entry activity in these markets. There has also been an active trend toward cross-border mergers and acquisitions among financial institutions across world regions. There is still discussion on the benefits of such process but it has undeniably changed the competitive landscape (Berger, 2000). Latin America and the Caribbean is a region where this trend has been particularly intense. Many countries suffered financial crises after the initial period of liberalization since the beginning of the 1990s. The cycle of liberalization-crisis-re-liberalization has created three important phenomena: a reduction in the total number of banks, a decline in the participation of state-owned banks in total intermediation, and finally, perhaps most importantly, an increase in the market shares of foreign banks. Table 1 reports changes in the market share of foreign and state-owned banks. As seen in the Table 1, the market share of foreign banks increased tremendously over the period 1995–2001. Table 1. The market share of foreign and state-owned banks, 1995–2001 Country The market share of foreign banks (1995–2001) (%) The share of foreign banks from other Latin American countries in foreign banking 2001 (%) The share of Spanish and American banks in foreign banking 2001 (%) The market share of state-owned banks (1995–2001) (%) Argentina 21.7–48.5 3.4 57.1 36.0–32.4 Brazil 8.4–29.7 0.3 52.9 52.3–33.4 Chile 15.7–46.8 0.4 86.1 16.3–13.3 Mexico 6.2–75.5 0.0 90.9 −21.7 Venezuela 3.3–43.2 0.5 89.3 33.7–6.9 Source: All the figures are taken from Bankscope IBCA (own calculations), Latin Banking—Citibank Guide and Directory 2002, Miami August 2002, and the book “La banca en Latinoamérica: reformas recientes y perspectivas”, by J.M. Liso, M. Soler, M. Manero, M.P. Buil (Liso et al., 2002). The figures do not include local credit cooperatives for Brazil. Note: Mexico had a nationalized banking system during the period 1982–1991. The market share is calculated with respect to total assets. Table options Large international banks like Banco Bilbao Vizcaya Argentaria, Santander Central Hispano, Citibank, HSCB Bank, Fleet Boston, ABN AMRO Bank, Scotiabank and JP Morgan are very active players across all markets. The operations of these financial institutions, at a regional level, have increased dramatically compared to intraregional integration among native banks. As reported in Table 1, the market share of foreign banks from other Latin American countries is very small in large economies in Latin America. The share of Spanish and American banks in foreign banking, on the other hand, is large. Therefore, inter-country financial integration has been accomplished by international banking institutions. Domestic regulation has also been updated to international standards. Although the market share of state-owned banks has decreased over the years, they are still major players in the Latin American and Caribbean banking systems. The prevailing organizational forms in the region are commercial banks. Only 12% of all banks are specialized credit institutions. With variations from country to country, most of the commercial banks in the region can engage in security, insurance, real estate, and have limited degrees of ownership in non-financial firms.2 A publicly debated issue in most of these economies is that of the large spreads between lending and deposit rates. These large spreads have been associated with low levels of efficiency and blamed for low private saving and investment rates. Table 2 shows average performance ratios for Latin American and the Caribbean banks, and other world regions. Total costs to assets are the highest at a world level. Overhead costs are only the second to those of Eastern Europe. Among the explanatory factors of inefficiency that are being publicly discussed are low levels of competition, fragile legal and volatile macroeconomic environments, a small deposit base, disperse and impoverished populations that require extensive branch networks, a slow introduction of risk management techniques and deficient regulation. No systematic study on these issues has been done for the Latin American and the Caribbean banking systems to provide answers for why banking is so costly in the region. This paper tries to bring evidence on some these possible explanations. Table 2. Comparative financial highlights 1995–1999 Europe (excluded eastern Europe) Eastern Europe USA/Canada South and Central America Middle East Far East Oceania Africa Overheads/average assets (%) 3.09 7.39 5.14 6.61 2.47 2.72 3.75 5.38 Total costs/average assets (%) 7.41 14.44 9.03 16.55 7.11 8.19 9.08 11.35 Non performing loans/gross loans (%) 7.47 11.00 1.25 6.38 12.36 8.45 2.74 17.37 Equity/total assets (%) 9.16 18.31 10.21 16.03 14.95 12.21 9.67 13.73 ROAA (%) 0.77 1.79 1.23 0.98 1.53 0.27 1.07 2.03 ROAE (%) 7.77 11.08 13.57 7.58 14.21 2.38 18.51 19.25 Total assets (Th. US$) 5553132 664744 10080016 1539199 3,037632 16370360 921660 7241759 Number of banks 5.503 685 1068 1113 220 1083 116 532 Source: Bankscope IBCA. Table options Consistent with the globalization of regional economies and financial markets, econometric studies have examined cost and profit efficiency in the context of multi-country common cost and profit frontiers (Dietsch and Lozano-Vivas, 2000, Ruthenberg and Elias, 1996, Maudos et al., 2002, Pastor et al., 1997, Allen and Rai, 1996, Berger and Humphrey, 1997, Hughes et al., 1996, Berg et al., 1993 and Fecher and Pestieau, 1993). To the present, there are no studies on cost or profit efficiency of banking sectors in the Latin American and the Caribbean banking region. The main objective of this paper is to provide extensive information on the efficiency of Latin American and the Caribbean banking systems. We identify and estimate a common cost frontier with country-specific environmental variables for 16 countries and derive X-inefficiency, scale and scope economies for a large sample of banks in the region. We then use bank specific variables to explain the sources of inefficiency across banks. We follow the Dietsch and Lozano-Vivas (2000) approach to estimate cost inefficiency of 16 Latin American and the Caribbean banking systems. They compare cost efficiency of the French and Spanish banking sectors and emphasize the influence of environmental conditions on the cost efficiency. In their study, the common cost frontier incorporates country-specific environmental conditions. Inclusion of environmental variables into the analysis allows the researcher to verify the degree of similarity between banking technologies. Our approach assumes that globalization implies a rapid availability of banking technologies and financial liberalization implies a more fluid flow of financial resources and management across countries. To compare the cost efficiency of banking industries we include the appropriate environmental variables in the cost frontier estimations. Therefore, once a common frontier is derived for the banking sector at a regional level, differences in banking efficiency across countries can be explained by a global best practice bank. The analysis of efficiency in the Latin American and Caribbean banking systems is important for its policy implications. If low efficiency and competition are to be blamed for large spreads, which affect both total investment and saving rates, understanding efficiency determinants can shed light on appropriate policies with clear macroeconomic and regulatory implications. The rest of the paper is organized as follows. In Section 2, we discuss the methodology and the econometric specification used to estimate the common cost function. The data and empirical results of the estimation are reported in Section 3. Section 4 explores the sources of efficiency in the banking systems. The paper’s concluding remarks are provided in Section 5.

This study investigates the cost efficiency of the Latin American and Caribbean banking industries for the period 1995–1999 using a stochastic frontier model. We used IBCA information to form an unbalanced panel and estimated economies of scale, economies of scope and X-inefficiency scores for a sample of 481 banks. The results indicate a wide range of inefficiency levels across countries, from 0.093 for Honduras to 0.203 for Venezuela. Also, the largest economies in the region also tend to be the most inefficient. Very small and very large bank are significantly more inefficient than large banks. Average cost inefficiency is 0.171. This means that the average bank in the sample could have reduced cost by 17% had they all been operating at full efficiency. The findings of this paper further indicate that there are no diseconomies of scale in any of the countries, for any year, for any organizational type of banks and for any size class. While most of the empirical studies in this literature report that scale economies are exhausted at relatively low levels of output, our findings suggest that economies of scale exist at any production scale. This remarkable result indicates that Latin American and Caribbean banks operate at scale below technological possibility. This finding, along with the generalized level of X-inefficiency, presents a particular combination. The findings also indicate that there are economies of scope for some banks regardless of their size. This means that joint production of outputs is less costly than producing each output separately. Therefore, banks in the region not only operate above the optimum frontier but also fail to reach optimum scope and scale. This paper also investigates the sources of inefficiency. The results suggest that underperforming banks tend to be smaller in size, to be undercapitalized, to present poor profit performance, to be more dependent on non-interest income, to be more risky, to have a less stable deposit base and to intermediate less. Banks tend to be more efficient in countries that grow faster, have higher density of demand for banking services and have banking systems with less market power.