FRBSF Economic Letter
2006-19 | August 4, 2006
Performance Divergence of Large and Small Credit Unions
By various measures, larger credit unions have recently had stronger financial performance than smaller credit unions, indicating that these institutions face large and pervasive economies of scale (Wilcox 2005a). This Economic Letter uses data from the 1980-2004 period to show that this performance difference is a long-running state of affairs. Moreover, these data reveal increasing performance divergence over this period–that is, a widening in the gap in financial performance between large and small credit unions. Thus, it is not surprising that the number of smaller institutions has been shrinking, while the number of larger institutions has been rising. Specifically, between 1980 and 2004, the number of small credit unions (less than $100 million in assets in 2004 dollars) shrank from 17,132 to 7,859, while the number of large credit unions (over $1 billion) grew from 2 to 98. If performance divergence continues, it is likely to quicken the pace of consolidation in the credit union industry; nonetheless, thousands of small credit unions may well survive for decades.
One measure of the relative cost efficiency of a bank or a credit union is lower noninterest expense, which consists of wages, salaries and benefits, depreciation of equipment and buildings, costs of supplies, marketing, office operations, and travel, among others. As the string of negative readings in Figure 1 shows, noninterest expenses were consistently lower at large credit unions than they were at small credit unions. In particular, at large credit unions, the ratio of noninterest expenses to assets was more than 100 basis points (1 percentage point) lower on average than that for small credit unions. To put the size of this cost advantage in perspective, note that, on average, noninterest expenses at small credit unions are about one and a half times as large as those for large credit unions. Expressed differently, if large credit unions had noninterest costs in 2004 as high as those of small credit unions, all else equal, the net income of large credit unions as a group would be negative. Instead, large credit unions enjoyed a very profitable year.
Figure 1 also shows that the long-running cost advantage of large credit unions has increased from an average of fewer than 100 basis points during the 1980s to more than 120 basis points over the past dozen years. Diverging costs make it ever more difficult for small credit unions to offer financial services and interest rates on loans and deposits that effectively compete with those of larger credit unions.
This sizable increase in the already large cost advantage of large credit unions might have several sources. One might be the faster growth of large credit unions relative to smaller institutions, which would allow them to achieve relatively more scale-related cost savings. In addition, recent legislation may have imposed costs on banks and credit unions that weigh more heavily on small institutions. To the extent that the costs of complying with the Bank Secrecy Act and anti-money laundering and other regulations increased after our data end in 2004, the cost disadvantages of smaller institutions may have worsened even further. Such cost divergence can further stimulate consolidation.
Figure 2 plots the annual differences in the ratios of net income (or return on assets, ROA) and of interest expenses between large and small credit unions (measured as a ratio to assets). Credit unions are mutually owned by their members, not by stockholders. Thus, all else equal, both higher ROA and higher interest expenses (which are interest incomes to credit union members) signal that credit unions are providing greater benefits to their members. On average, larger credit unions both earned more net income and paid more interest to their members than small credit unions did. Because they have substantially lower noninterest expenses than small credit unions do, large credit unions have the wherewithal both to pay higher interest rates and to generate higher earnings. In addition, members of larger credit unions also benefit from lower loan interest rates and a wider range of financial services (Wilcox 2005b, CUNA 2006).
Figure 2 shows that the persistent reductions in noninterest expenses at large relative to small credit unions tend to be mirrored by persistent upticks in the earnings of large relative to small credit unions. While large and small credit unions had approximately equal average ROAs over the 1980s, the earnings of large and small credit unions steadily diverged since then, with the difference in their ROAs averaging over 40 basis points since 2000.
In contrast to noninterest expenses and earnings, interest expenses at large and small credit unions have shown no clear tendency to diverge. Large credit unions paid about 40 basis points more in interest to their members than small credit unions, both on average since 1980, and in 2004.
Since large credit unions have long been more efficient, earned more, and paid higher interest rates to their members than small credit unions, why have small credit unions historically had so many members and assets? Is the relative shifting of members and assets toward large credit unions likely to continue and, if so, at what pace and with what repercussions?
The history, the unwinding, and the recent additions of financial regulation each helps answer these questions. Just as banks faced stringent branching and ownership restrictions before the 1980s, restrictions on fields of membership (FOMs) usually confined individual credit unions to serving the employees of a single government agency or company, or even a single plant of a company. Thus, millions of people were eligible to join only one, often-small, credit union, or to join none at all. Thus, as for banks, the large numbers of small credit unions historically were partly an artifact of regulation.
Over the last 25 years, changes in regulation have greatly expanded FOMs; for example, sometimes individual credit unions can serve all of the employees of a nationwide company, all of the local or state government employees within a state, or all of the residents and employees within a local area (which might encompass several counties or millions of residents). Larger FOMs have both allowed individual credit unions to grow and thereby reap certain economies of scale and created more overlapping FOMs, which permit individuals to choose among credit unions. To the extent that credit union members migrate toward the (typically larger) credit unions that pay higher deposit rates, expanded FOMs contributed to the decline in the share of total credit union assets in small credit unions from 69% in 1980 to 46% in 1990, to 29% in 2000, and to 21% in 2004.
While Figure 1 shows the average amounts by which the costs of small credit unions exceed those of large ones, it does not reveal how much the cost disadvantages vary between small credit unions. In 2004, more than 4,000 small credit unions, which together held most of the assets of all small credit unions, had noninterest expenses that exceeded the average for large credit unions by more than 100 basis points. For a time, many of these relatively inefficient credit unions will likely survive, if not thrive. However, their earnings and interest rates are likely to preclude growth sufficient to reduce their average costs significantly. As a result, these credit unions likely will constitute shrinking shares of an otherwise sizable and sound industry. Further liberalization of FOMs or other changes that raise the average sizes of credit unions would likely also further stimulate consolidation among credit unions.
Nonetheless, the credit union industry is unlikely to be dominated by a few nationwide institutions any time soon. One reason is that legislation and regulation still restrict the expansion of the FOM for an individual credit union at most to the employees of a profession or business or to a local area. Another reason is that costs at many small credit unions are low enough to keep them competitive. In 2004, for example, more than 1,500 small credit unions (accounting for 15% of assets in all small credit unions) had noninterest expense ratios that were lower than the average ratio for large credit unions. While the total number of credit unions has fallen from a peak of 23,866 in 1969 to 9,483 in 2004, the rate of consolidation has slowed recently. Indeed, if the 1995-2004 rate of consolidation continued, for example, the U.S. would still have over 5,000 credit unions in 2025.
Similarly, for commercial banks, the loosening of geographic and other restrictions contributed greatly to the consolidation from a post-Great Depression peak of 14,482 banks in 1984 to 7,630 in 2004. Nonetheless, in recent years, small banks have prospered, numerous new banks have been chartered, and the pace of consolidation has slowed. (Perhaps further testimony to the prospects for small credit unions generally is the virtual absence of any new credit unions being formed over the past decade.) Indeed, if the 1995-2004 rate of bank consolidation continued, the U.S. would still have about 4,000 commercial banks in 2025. Together, their current healthy performance and condition and the pace of recent consolidation suggest that the U.S. may well have thousands of small credit unions and small commercial banks for decades to come.
The credit union industry’s responses to the deregulation, technological advances, and additional regulations of the recent past suggest that it will be consolidating for many years to come. The industry’s successes suggest that some small credit unions (like some small banks and some small nonfinancial businesses) are sufficiently cost-efficient and attuned to the needs and circumstances of their local deposit and loan customers that they will thrive.
Many other credit unions, however, will not thrive or even survive. Continuing performance divergence will make it increasingly difficult for smaller credit unions to serve their members as effectively as larger credit unions do and to meet standards for safety and soundness. In that event, less-efficient and less-profitable credit unions will increasingly feel pressures to liquidate, merge, or convert to other charters. The size and shape of the credit union industry will reflect which of these options are favored by credit union members and managements and their regulators.
James A. Wilcox
Visiting Scholar, FRBSF, and
Professor, Haas School of Business, UC Berkeley
[URLs accessed August 2006.]
CUNA (Credit Union National Association). 2006. Credit Union Report Year-end 2005. Madison, WI.
Wilcox, James A. 2005a. “Economies of Scale and the Continuing Consolidation of Credit Unions.” FRBSF Economic Letter 2005-29 (November 4).
Wilcox, James A. 2005b. Failures and Insurance Losses of Federally-Insured Credit Unions: 1971-2004. Madison, WI: Filene Research Institute.
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120