Heterogeneous credit union production technologies with endogenous switching and correlated effects

Abstract

Credit unions differ in the types of financial services they offer to their members. This paper explicitly models this observed heterogeneity using a generalized model of endogenous ordered switching. Our approach captures the endogenous choice that credit unions make when adding new products to their financial services mix. The model that we consider also allows for the dependence between unobserved effects and regressors in both the selection and outcome equations and can accommodate the presence of predetermined covariates in the model. We use this model to estimate returns to scale for U.S. retail credit unions from 1996 to 2011. We document strong evidence of persistent technological heterogeneity among credit unions offering different financial service mixes, which, if ignored, can produce quite misleading results. Employing our model, we find that credit unions of all types exhibit substantial economies of scale.Keywords: Credit Unions, Correlated Effects, Ordered Choice, Panel Data, Production, Returns to Scale, Switching Regression JEL Classification: C33, C34, D24, G21 * Email : emalikov@stlawu.edu (Malikov), drestr16@eafit.edu.co (Restrepo), kkar@binghamton.edu (Kumbhakar).We would like to thank the editor, the associate editor and two anonymous referees for many insightful comments and suggestions that helped improve this article. We are also thankful to Alfonso Flores-Lagunes, Thierry Magnac, Dave Wheelock, Jeff Wooldridge and seminar participants at Syracuse University, University at Albany, Federal Reserve Bank of Dallas and the 2013 Midwest Econometrics Group Meeting at Indiana University Bloomington for many helpful comments and suggestions. Any remaining errors are our own. We also gratefully acknowledge the technical assistance from the Center for Scientific Computation APOLO at EAFIT University in setting up the cluster we used for the estimation. Restrepo-Tobón acknowledges financial support from the Colombian Administrative Department of Sciences, Technology and Innovation, Colombian Fulbright Commission and EAFIT University. The original version of this paper was circulated under the title "Are All U.S. Credit Unions Alike? A Generalized Model of Heterogeneous Technologies with Endogenous Switching and Correlated Effects".

Economies of diversification in the US credit union sector

“…Consistent with the findings of Wilson (2011) andMalikov et al (2015b), who, however, investigate credit union costs at the conditional mean only, we also find the overwhelming evidence of statistically significant increasing returns to scale across credit unions of all types and quantiles of the cost distribution. We find that, for a credit union in the middle of the cost distribution, the consolidation of the issuance of real estate loans with consumer loans and investments reduces credit union costs, on average, by an astounding 45%.…”

“…Malikov et al (2015b) document that the overwhelming majority of retail, or so-called natural-person, credit unions (more than 99%) offer one of the following three financial service menus to their members: (i) consumer loans and investments y 1 = (y 3 , y 4 ); (ii) real estate and consumer loans, as well as investments y 2 = (y 1 , y 3 , y 4 ); and (iii) all types of financial services-real estate, business and consumer loans, and investments y 3 = (y 1 , y 2 , y 3 , y 4 ). Malikov et al (2015b) document that the overwhelming majority of retail, or so-called natural-person, credit unions (more than 99%) offer one of the following three financial service menus to their members: (i) consumer loans and investments y 1 = (y 3 , y 4 ); (ii) real estate and consumer loans, as well as investments y 2 = (y 1 , y 3 , y 4 ); and (iii) all types of financial services-real estate, business and consumer loans, and investments y 3 = (y 1 , y 2 , y 3 , y 4 ).…”

“…23 The sample period we consider in this paper runs from 2001 to 2006. See the Appendix in Malikov et al (2015b), whom we follow closely, for details on construction of the variables. Further, given the large wave of mergers and acquisitions taking place in the US credit union sector over our sample period (see, e.g., Bauer et al, 2009), we limit ourselves to continuously operating retail credit unions only.…”

“…The latter has recently become a subject of particular scholarly interest (e.g., Emmons & Schmid, 1999;Malikov et al, 2015bMalikov et al, , 2016Wheelock & Wilson, 2011;Wilcox, 2005Wilcox, , 2006. We next proceed to measuring the degree of scale economies (if any) exhibited by credit unions.…”

“…The increasing availability of credit information via nation-wide credit-reporting bureaus, establishment of the federal deposit insurance fund for credit unions and the growth in credit card lending by larger commercial banks have dramatically eroded benefits of the so-called "soft" information that lies at the core of credit unions' business formula (Walter, 2006). Indeed, Wheelock and Wilson (2011) and Malikov, Restrepo-Tobón, and Kumbhakar (2015b) document pervasive increasing returns to scale among retail credit unions in the USA in the 1989-2006 and 1994-2011 periods, respectively. Indeed, Wheelock and Wilson (2011) and Malikov, Restrepo-Tobón, and Kumbhakar (2015b) document pervasive increasing returns to scale among retail credit unions in the USA in the 1989-2006 and 1994-2011 periods, respectively.…”