Economies of scale, economies of scope and structural change in the Mexican commercial banking system

The Mexican commercial banking system has been reformed continuously in the last 15 years with banking authorities playing an important role shaping the sector. They acted on the belief that banks were subject to both scale and scope economies. Thus they promoted the consolidation of the banking system into a smaller number of institutions.;This dissertation focuses on the validity of these two assumptions. It estimates a multiproduct cost function to assess returns to scale, economies of scope, and structural change of Mexican banking. The study was carried out with quarterly data from 1980 through 1982 for 29 commercial banks, accounting for more than 97 percent of the total funds mobilized in the commercial banking system.;Extensive testing of alternative forms of cost function were undertaken. Tests favored the use of a translogarithmic specification. Output aggregation was rejected, requiring a cost function with 6 banking services. Structural change through time and among institutions was identified by means of the method of recursive residuals. A uniform cost function for all banks and all periods was rejected. Test procedures allowed for an appropriate estimation of the parameters of scale, cost complementarities, marginal costs and elasticities of substitution.;The study concludes that the cost structure among banks varied due to product mix and scale of operation and to changes in the micro and macroeconomic environment.;Second, no evidence was found supporting the existence of scale economies. Cost structures for different bank groupings indicate that the production of banking services is characterized by constant returns to scale.;Third, cost complementarities exist between several kinds of deposits and loans, between deposits, and between various kinds of loans. However, these cost complementarities do not exist for all the combinations of outputs.;Fourth, augmented economies of scale, i.e. considering the branch network, do not exist. The augmented measure for economies of scale points to diseconomies.;Finally there was no evidence to support the authorities argument regarding economies of scale. The banks are not homogeneous intermediaries. Five classes of banks were identified that vary in product mix and scale of operation. Designing policies on the assumption that all banks have similar cost functions and behave similarly can lead to undesired results.