Economies of Scale of Financial Institutions

Knowledge about economies of scale are necessary for regulatory and management decisions. Because of its structure, the financial industry also provides us with an excellent source of data for measuring the cost function. The studies surveyed are evaluated with respect to these needs. The general form of the cost function is discussed first in Section II. The nature of the banking and savings and loan industries allows use of a reduced form single equation. This equation can be specified validly with the cost data available from fnancial institutions. The usual problems in defning cost (such as the difference between accounting and economic cost) are not very serious for these institutions, as is discussed in Section III. Definition of of fnancial institutions is a problem that has plagued most studies of the industry as well as studies of cost fSnctions in general. Some researchers have defined output as total gross revenue, others as total assets or total deposits, and others as numbers of deposit and loan accounts. After analysis, in Section IV I conclude that the number of deposit and loan accounts is the preferable measure of output. However the particular defnition of output used depends heavily on the method of measuring economies of scale. The various techniques used by researchers are discussed in Section V, a discussion which confirms the use of the number of deposit accounts and loans. The findings of the preferred studies by Benston and Bell and Murphy presented in Section VI covers six years of data for commercial banks and five years of data for savings and loan associations. In general they find economies of scale for both types of institutions. Consistent and significant economies of scale are found for *Support of The Security Trust Company of Rochester is gratefully acknowledged. GEORGE J. BENSTON is professor in the Graduate School of Management and Center for Research in Government Policy and Business, University of Rochester. This content downloaded from 207.46.13.176 on Mon, 20 Jun 2016 06:42:05 UTC All use subject to http://about.jstor.org/terms GEORGE J. BENSTON : 3 13 the demand deposits and real estate loans functions. Time deposit3 and instalment loans show significant elasticities of under 1.0 for some of the years but not for all. The elasticities for business loans also are generally but not consistently significantly under 1.0. Of the other direct cost analyses, safe deposit and trust show consistent economies of scale. Interestingly, administration expenses do not show consistent significant elasticities of under 1.0, a fnding which is contrary to the often expressed belief that administration is a source of considerable economies of scale. Occupancy and business development expenses do appear subject to economies of scale, but these findings are not reliable due to the nature of the data. For an average commercial bank, overall elasticity with respect to operating costs is .93. A 10 percent increase in the output at the average bank of each service while holding all other independent variables constant at their geometric mean values is accompanied by a 9.3 percent increase in operating costs. For savings and loan associations, consistent and significant economies of scale were found for all years. These average about .923. Interestingly this elasticity is almost exactly the same as those found for commercial banks for the mortgages and time deposit functions: it appears that commercial banks and savings and loans have just about the same cost structure with respect to size of operations. Branch banking was found to be more expensive than unit banking for both commercial banks and savings and loans. However, these costs increase at a decreasing rate as additional branches are added. Further, for commercial banks, the additional cost of branching appears to be offset by the economies of larger scale operation, since larger banks tend to be branch banks. This finding does not hold as strongly for savings and loan associations. Larger associations are not necessarily those with the greater number of branches nor are the economies of scale sufficient to offset the higher cost of branching. It should be noted that the cost of inconvenience to a customer of unit banking is not included in operating costs. The implications of the fndings are discussed in Section VII. Elasticities measured indicate that larger firms are preferable ceteris paribus. However, for specifc regulatory decisions the amount of the expected savings and operating costs should be compared to an estimate of the disadvantages, if any, of favoring larger banks. Nevertheless, one can generalize that the size of the economies of scale is not such that medium sized banks would be squeezed out by larger banks if growth and mergers were allowed. However, small banks do appear to have a considerable cost disadvantage, especially with respect to branches of larger banks. In choosing between chartering a small unit bank or allowing branching of a larger bank, the regulatory authorities should prefer the branching application unless considerations of competition are overriding. The data presented provide useful estimates of marginal costs for both commercial banks and savings and loan associations. These should be of value to managers of the institutions. Sources of economies of scale found were investigated. It appears that a major source is the use of lower skilled labor by larger firms. The ability of larger firms to use sophisticated machinery also was found to be a source of economies of scale in the demand deposits, instalment loans and business loans functions. The model specified and the data used fulfill virtually all of the theoretical and statistical requirement for estimating cost functions. Thus they provide better evidence than was previously available on the existence of economies of scale. This content downloaded from 207.46.13.176 on Mon, 20 Jun 2016 06:42:05 UTC All use subject to http://about.jstor.org/terms 3 14 : MONEY, CREDIT, AND BANKING I. THE PURPOSE OF ECONOMIES OF SCALE STUDIES The findings of any analysis are in large measure a function of the purpose for which the analysis is undertaken. We cannot measure or even conceive of all possibilities, so in deciding which events to record and study (and even how these events are recorded) we are guided by the questions that we hope our analysis will answer. Therefore, it is useful to begin this survey by considering the questions that a study of scale economies of financial institutions might seek to answer. This should provide a basis for evaluating the studies. The questions to which economies of scale are related may be grouped under three headings: (1) regulation-entry, merger, branching, and some prices in the finance industry are regulated (in part because of assumptions about scale economies), requiring regulators to estimate the cost consequences of their decisions; (2) importance to the finance industry-if firms understand their costs better, they may be more efficiently managed and make decisions more rationally, and (3) economic understanding-knowledge of the relationship between cost and output is basic to microeconomics, industrial organization, etc., and a study of the finance industry should provide empirical estimates that are relevant to theory. Each of these are discussed next in greater detail.