The impact of differing operating environments on US Credit Union Performance, 1993–2001

Abstract

Today US credit unions operate within a highly competitive financial market place. Set against this competitive operating environment, the present study employs stochastic frontier analysis to evaluate the performance of large credit unions (assets greater than $50 million) over the period 1993 to 2001. Although credit unions may share a common co-operative philosophy, differences between credit unions are also apparent across a range of operational, structural and locational characteristics (environmental conditions). The impact of these different environmental influences is modelled in two ways. One assumes that environmental factors affect the efficiency with which the production process is operated, while the second assumes that the environment affects the production process itself. Net and gross cost efficiency measures are obtained for both models, with the differences between these measures for a specific credit union being viewed as the impact that environmental variables have on the inefficiency of that credit union. In addition, if it is assumed that the main environmental factors are accounted for in the modelling, then a credit union's net efficiency measure may be interpreted as a measure of managerial performance when operating in equivalent environments. The analysis revealed that different environments (the age of the credit union; the potential for expansion within the existing common bond; whether the credit union has the option of expansion through the addition of select employee groups; whether the credit union is state or federally regulated; whether insurance is provided at state or federal level; as well as regional characteristics such as per capita income and the level of unemployment) account for much of the variability in cost efficiency between credit unions and once credit unions are placed in broadly equivalent operating environments only marginal differences are apparent in their managerial performance.

Acknowledgements

The authors are indebted to Callahan and Associates for supplying the data on which this paper is based.

Notes

¹ Ferguson and McKillop (1997) provide a typology outlining the characteristics one might expect from credit union movements at different stages of development.

² Kebede and Jolly (2001) provide an analysis of the effects of financial structure on low income credit unions.

³ Credit unions’ exemption from federal income tax dates back to the Revenue Act of 1916, which provided tax-exempt status to mutual thrift organizations and co-operatives. The US Attorney General ruled in 1917 that credit unions which, at that stage, were all state chartered were entitled to the exemption. The first federal credit unions were chartered in 1934 and granted tax-exempt status in 1935 under a ruling by the Internal Revenue Service.

⁴ Frame et al. (2003) examines whether credit unions use their tax advantage to benefit members or whether it is consumed by managers.

⁵ Fried and Lovell (1993, 1994) emphasized that not all US credit unions provide complete, consistent and believable information.

⁶ Investments include: trading securities; available for sale securities; held-to-maturity securities; deposits in commercial banks, S&L's and savings banks; membership capital and paid-in capital at corporate credit unions; and all other investments.

⁷ Other loans encompass: new vehicle loans; used vehicle loans; other lines of credit to members; and other loans to non members.

⁸ Off-balance sheet loans are included in reflection of the growth in importance of this loan class to credit unions over the period under investigation. Berger and Mester (1997) state that off-balance sheet items are often effective substitutes for directly issued loans, requiring similar information-gathering costs of origination and ongoing monitoring and control of the counterparties as well as similar revenues.

⁹ Legget and Strand (2002) provide an extremely useful analysis of the addition of multiple membership groups and agency control within US credit unions.

¹⁰ Even in states, which have not enacted legislation similar to that operating at federal level, opposition by other financial institutions to the expansion of federally chartered credit unions is particularly vocal. For example, the Utah Bankers Associations filed a law suit in mid-July 2003 challenging the NCUA's decision to grant increased membership flexibility to two Utah based federally chartered credit unions.

¹¹ Emmons and Schmid (1999) provide an overview of certain theoretical issues pertaining to the common bond and the manner in which it impacts on the operation of US credit unions.

¹² These geographical areas have become much wider during recent times. An example is America First which in 2003 was granted a charter by the NCUA permitting it to serve members in six Utah counties.

¹³ Clark (1988) states that for the multi-product financial firm overall economies of scale occur if total costs increase proportionately less than output when there is a simultaneous and equal percentage increase in each of the financial firm's products.

¹⁴ It should be noted that Fried et al. (1993) also explored the impact of resource slacks and found that when radial inefficiency and slack are combined a less definitive picture emerged due to large credit unions having more slack in resource use than small credit unions.

¹⁵ Hermalin and Wallace (1994) and Kaparakis et al. (1994) found a significant negative relationship; Berger et al. (1993) and Berger et al. (1997) found a positive relationship; and Aly et al. (1990), Berger and Hannan (1998), Cebenoyan et al. (1993), Mester (1993, 1996), and Pi and Timme (1993) found an insignificant relationship.

¹⁶ See also Caudill et al. (2001) for a study of the effect of charter status and ownership type on the efficiency of the thrift industry.

¹⁷ Similar conclusions to those of Reichert and Rubens (1994) emerge from the work of Srinivasan and King (1998) and Wolken and Navratil (1985). The latter examined the de novo chartering decisions of credit unions between 1978 and 1980 and found that the more liberal the state regulations the greater the probability that state chartering would be chosen. Specifically, intrastate branching ability, broader investment powers and the lack of state taxes were the key determinants.

¹⁸ Coelli et al. (1999) note that without guidance from the data the ex ante selection of one model type over the other is a difficult task. It could, however, be argued that the ‘type one’ model is to be preferred as the estimated frontier represents the lower boundary of the cost possibility set, irrespective of environmental issues.