Review of Industrial Organization 20: 105–113, 2002.
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
Credit Unions: Fringe Suppliers or Cournot
ROBERT M. FEINBERG
Department of Economics, American University, 4400 Massachusetts Ave., N.W., Washington, DC
Abstract. Recent work has demonstrated the competitive relationship between credit unions and
banks in consumer ﬁnancial services. One issue underlying the nature of competition between the
two, however, concerns the most appropriate way to model their interactions. Two possible ap-
proaches are the dominant-ﬁrm price-leadership model and the generalized Cournot model. In the
former model, credit unions act as fringe suppliers who are price-takers in a homogeneous product
market. In the latter, they possess (limited) market power. One way to distinguish the two is by ex-
amining the impact of credit union market shares on their pricing, as the two models imply differing
effects. Our results are more consistent with the “credit unions as fringe suppliers” view. Using a
pooled cross-section time series of 77 small local consumer lending markets throughout the U.S.,
each with 10 observations over 5 years, the focus is on a loan product ex ante thought to be sold
in local markets, unsecured (non-credit card) loans. For this product, increasing credit union market
shares reduces credit union loan rates, consistent with a fringe supplier hypothesis.
Key words: Credit unions, local banking markets, consumer loans.
The continuing dramatic consolidation of the ﬁnancial services industry has nat-
urally raised concerns about competitive effects. Many economists have suggested
that, given the wide range of sources of ﬁnancing available to most consumers, an-
titrust authorities should be cautious about intervening in bank mergers. However,
in smaller and more-remote communities, especially for loan products not widely
available through a national market (as well as for deposit products), increasing
The author thanks Judy Karofsky and Matthew Shearer for research assistance and Bill Kelly
and an anonymous referee for helpful suggestions, and acknowledges research funding support from
the Filene Research Institute and the Center for Credit Union Research, School of Business, Uni-
versity of Wisconsin–Madison. Previous versions of this paper were presented at seminars at the
College of William and Mary, the 1998 Southern Economic Association annual meetings, Charles
River Associates, Inc. (Washington, DC ofﬁce), and the 1999 Western Economic Association Inter-
national Conference. This article was written while the author was on leave as Program Director for
Economics, National Science Foundation. All views expressed, of course, are those of the author