Along with the decision of when and where to open a new branch office, one of the key decisions made by banking administrators is when and where to *close* a branch office. We decided to take a look at recently closed branches in order to identify what common characteristics we might be able to identify.
BranchInfo by RPM Consulting
To conduct our analysis, we used the most current version of BranchInfo. RPM's BranchInfo competitive bank branch database tracks not only branches that are currently open, but also branches that have been closed since the last data update. The current version of BranchInfo, released in February 2011, contains bank branch locations and ownership as of February 15, 2011, as well as branches which were closed between July 1, 2010 and February 15, 2011. Also, new to BranchInfo beginning with the last update, are a series of key market potential and competitive metrics calculated for a one-mile radius surrounding each branch. In order to better understand the characteristics associated with branches that institutions have chosen to close and/or consolidated into other locations, we decided to conduct an analysis of the closed locations, comparing them to locations which have remained open.
To conduct our analysis, we began with the 422 branches reported as having been closed between July 1, 2010 and February 15, 2011. We then identified the complete branch networks for all institutions that had reported a closed branch during that time period so that we could compare closed branches to open branches for the same institutions that had closed one or more branches. This approach allowed us to compare the 422 closed branches to 43,439 open branches for the same institutions.
Once our database was assembled, we compared the following characteristics for the open and closed branches:
- Age of the branch
- MarketBank deposit potential
- MarketBank loan potential
- MarketBank investment potential
- Number of competing branches
- Average MarketBank deposit potential per branch (within a one-mile radius)
- Branch deposit market share
- Size of branch in deposits dollars
- Type of institution regulator
- Branch office versus main office
- Service type (full service brick & mortar, etc.)
- Each branch's share of total bank deposits
- Bank Asset Size
Each variable was tested for statistical significance using either chi square analysis or one-way analysis of variance, with .05 being set as the alpha value (for you quantitative types).
For our study, we broke down the variables we analyzed into several categories, focusing on the physical/administrative characteristics, competitive environment, and market potential.
First looking at the physical and administrative characteristics of closed branches, we found that the age of a branch was a statistically significant when comparing open and closed branches. Closed branches are, on average, much younger than branches that remain open. While the average age (calculated using the "Date Established" field) of an open branch is about 36 years, the average of a closed branch is just 26 years.
Another significant difference concerns the service type of the facility. Overall, we found that "Limited Service" facilities (such as drive-through branches, and seasonal/mobile offices) were much more likely to have been closed compared to full-service offices.
In terms of the type of institution most likely to close a branch, state savings assocations and state savings banks closed 7.9% and 5.2% of their branches, respectively, compared to state and federal commercial banks, which closed less than 1% of their branches.
Closed branches also have a tendency to represent somewhat larger contributions to banks' overall deposit levels. Our analysis found that although the average open branch represented approximately .5% of its institution's deposits, the average closed branch represented a considerably larger 3.2% of its institution's deposits (consistent with a smaller institution). In terms of deposits, closed branches were somewhat smaller than open branches, holding an average of $34.6 million in deposits, compared to $60 million for open branches (excluding outliers).
Finally, the size of the institutions also seems to be significantly different when comparing open and closed branches. Although the average institution held approximately $125 million in assets, those with closed branches were considerably smaller, holding an average of just under $40 million in assets.
Based on the percentage of total branches closed during the second half of 2010, states with the highest percentage of closures (2% or more of the institutions' branches being closed) included Idaho, North Dakota, and Montana to the north, stretching south to Arkansas and Mississippi.
In our analysis, we studied the nature of each branch's competition by assessing the number and size (based on deposit balances) of competitor branches located within a one-mile radius. Interestingly, the number of competitors does not prove to be significantly different between open and closed branches, however an analysis of market share does surface an interesting difference. Although the average share of deposits for open branches was approximately 27.1%, the market share of closed branches was somewhat higher, at 30.7%, suggesting that a closed branch does not necessarily equate to an under-performing branch.
One measure that we often use to determine whether an market is potentially over-banked, is the average estimated MarketBank potential deposit balances divided by the number of bank branches present. In areas in which tend to be saturated with bank branches, we generally find that the average deposit potential per branch is typically lower than areas with unmet potential. An assessment of the average deposit potential per branch found that closed branches tend to be in somewhat lower potential markets than open branches. The average potential per bank branch in open-branch market areas was about $18 million, which was higher than the $15 million for closed branch market areas.
We also used our MarketBank retail market potential database to quantify the consumer deposit, retail loan, and retail investment potential indices for a one-mile radius surrounding each branch (an index value of 100 equates to the national average for balance potential per resident household). Not surprisingly, retail deposit, loan and retail investment potential was substantially higher among still-open branches compared to closed branches. Open branches had average deposit, loan and investment potential indices of 104.5, 101.1, and 105.9, respectively. All are significantly higher than the deposit index of 100.5, loan index of 92.8, and investment index of 99.3 for closed branches.
Based on these observations, we can conclude that institutions that close branches tend to be somewhat smaller thrifts, rather than commercial banks, and the closed branches tend to be rather younger, and more likely limited service facilities, with almost half the deposits of branches that remain open. The average market share of closed branches tends to be close to or above the average for other branches in their local markets, and the average deposit potential per competing branch tends to be somewhat below average, suggesting that the areas in which branches are closed tend to be rather smaller, as well as competitive and/or saturated markets. Additionally, the closed-branch markets tend to be hold lower than average potential for lending and investment products, although relatively average potential for deposits.
Although closed branches tend to be smaller branches in lower potential markets, the decision to close a branch is not one that is made consistently from branch to branch or from institution to institution, suggesting that very often, the decision may not be based on market conditions as much as it may be based on logistical or contractual issues. For example, given that limited service facilities are more often closed, it is possible that branches that have been opened as temporary facilities to serve a particular need are often closed when the need no longer exists and the branche's usefulness comes to an end.
Although this analysis gives us a start in understanding some of the underlying characteristics, a better understanding of the nature of closed bank branches should also include additional local market and locational information, such as daytime population, business mix, employment, and demographics of those residing within the markets.