THE DETERMINANTS OF THE FLOW OF SAVINGS TO CREDIT UNIONS, 1965-76

The purpose of this investigation was to study the demand for credit union shares. More specifically, this was accomplished by doing an in-depth analysis of the flow of savings to credit unions. The theoretical model focuses on the individual member-saver. Each individual is viewed as a risk averse investor who maximizes total utility by holding different financial assets subject to a wealth constraint. Utility, in this context, is represented by net interest return and a set of nonprice incentives which include the credit union\u27s ability to meet member credit demands and the availability of a payroll deduction plan for automatic saving. From the credit union\u27s viewpoint, the net flow of savings depends on the factors affecting the individual\u27s decision to allocate his savings, total saving of each member, and the number of potential members. The interaction of these variables determines the credit union\u27s supply of funds schedule. Given the rate of return that a credit union can offer at a point in time, the net flow of savings to a credit union is found. In order to estimate parameters for the theoretical model, a random sample of 430 Federally chartered credit unions operating in each year over the period 1965-76 was chosen. The data source was a set of computer tapes supplied by the National Credit Union Administration. It was suspected that credit union behavior might differ with the size of each organization, hence, the sample was stratified into three subgroups according to size of total membership. The three subgroups consisted of the smallest 25% of the credit unions in the sample with respect to membership, the largest 25%, and the middle 50%. Parameters were then estimated for each subgroup for each year from 1966 through 1976. Parameters were also estimated for two larger periods (1965-70 and 1971-76) using averaged data. Estimation was done using ordinary least squares regression analysis. Examination of regression results as well as various descriptive statistics obtained from the sample suggest several conclusions. First, the use of the stratified sample design was vindicated since the results from the subgroup containing the largest credit unions were substantially different from the other subgroups with respect to value of coefficients, explanatory power, and in some cases, significance of variables. Second, the results support an increased interest rate sensitivity on the part of credit union members after 1971. Third, the past ability of a credit union to meet its members\u27 demand for credit is a highly significant factor in explaining the flow of savings to a credit union. This suggests that the typical credit union member behaves more like a borrower than a saver. Fourth, the aggregate flow of savings to credit unions has demonstrated more instability since 1971. This is probably due to the increased interest rate sensitivity of credit union members and is possibly related to the initiation of deposit insurance for credit unions in 1971