Are Branches Dying? Or Not?

Subscribe TodayIf Murphy can have his own law, then so can I. Here’s Shevlin’s Law: For every statistic that proves a point, there are two that refute it.

Nowhere is this more true than with the life/death of bank branches.

Alternative Facts

A recent Economist article points out the decline in the number of bank branches:

“Since the financial crisis, banks have closed over 10,000 branches, an average of three a day. In the first half of 2017 alone, a net 869 brick-and-mortar entities shut their doors.”

But the total numbers don’t get under the covers of at least one important trend: Large institutions have acquired other banks, and closed many of the superfluous branches that were right next door to their existing branches.

If you’re a fintech-loving branchophobe (like me), you conveniently ignore that explanation and attribute the decline to the impact of fintech.

To show how easy it is to construct an alternative story line to the branches are dying narrative, consider this trend: Over the past 40 years, the number of branches per bank has grown nearly eight-fold (from 2.3 in 1977 to 15.8 in 2016), and in fact, grew by 50% in the 10-year period of 2006 to 2016.

The Economist article went on to point out:

“By 2009, America had 35 branches for every 100,000 adults. Since then, however, ultra-low interest rates and thickets of new regulations have squeezed bank profits. They have responded by trimming branches from a peak of about 100,000 to roughly 90,000.”

What the article didn’t mention, however, is that the “trimming” reduced the number of bank branches per 100k adults to where it was pre-crisis, and in fact, that number increased between 2014 and 2015.

Read More:

Talkin’ ‘Bout My Generation

Old people want branches and young people don’t, right? Oh wait, no, young people do want branches because they like to talk to people before making their decisions.

I don’t know which it is — the data tells two different stories. On one hand, there’s data from Salesforce that shows that consumers of all ages prefer digital channels to the branch, and in fact, the percentage that prefers the branch is a minority, and doesn’t vary all that much for consumers over the age of 35 (the top line represents online, the red line is branch, and the dark gray bar is mobile).

I’m not sure how to reconcile the Salesforce data with findings from a recent Fiserv study, however, which found that 44% of consumers prefer to do their “standard daily transactions” at a traditional branch with tellers. I find this really hard to believe.

Read More:

Shrinky Dinks

Another common narrative is that the size of bank branches is shrinking. One CNBC article claimed in its title that “Bank branches are evolving…and shrinking,” and that’s just one of many articles making that claim.

Another article here on The Financial Brand cited a study that shows the opposite, however. According to JLL, the average size of new bank branch projects is 8% larger than the average size of existing branches. And in fact, in the West, new branches will be 50% larger than existing branches. They grow ’em big out there, don’t they?

Shevlin’s Law

The purpose of this post is not to refute the “branches are dying” narrative. I just wanted to point out that there’s data to support whatever argument you want to make about the life or death of branches. And I was hoping to build a little awareness of Shevlin’s Law.

Ron ShevlinRon Shevlin is Director of Research at Cornerstone Advisors. Check out more of his ideas and research on Cornerstone's Insight Vault. And don't forget to follow him on Twitter at @rshevlin.

This article was originally published on August 2, 2017. All content © 2018 by The Financial Brand and may not be reproduced by any means without permission.


  1. I’ve been trying to get stuff named after me with little success.

    Shevlin’s Law: Figures don’t lie but liars figure?

    Very good points.

  2. I think “the Marsico Doctrine” has a nice ring to it. Now that you have a title, you just need to come up with the content.

  3. Branch model has to evolve with new technology, having said that, physical branch relevance still reflects in the fact of the direct correlation between branch share and market share, greatly important for revenue generation. I like your approach, figures can tell different realities (all of them right).

    Great article (and thinking).

  4. Thanks, Luis. Appreciate the kind words.

  5. I think the most critical KPI for branches is one that is ignored (perhaps strategically) by most banks – or at least not published. That metric is:

    The number of visits to a branch per customer annually

    If you track that number over time you get a definite decline from 10-20 x per year in the 90s to 2-3x per year today. That number effectively means that if the decline continues, cross-sell, upsell and pretty much every other sales metric will continue to decline, and that means that branch economics at some point in the near future fail. Not for every branch, but for large swathes of branches.

    Ultimately the only way to staunch that trend is to convince people that human facing interactions top convenience – I don’t think that is possible en masse.

    I know what you’re saying about you choose the data you want – and I can see that. However, the underlying metrics of visits per individual customer per year, and what that means for branch economics is the big deal


  6. Brett, you are right. If you take all transactions done by a customer, less than 6% would be sales. Meaning service transactions are the most frequent, and for those, digital channels (on line – mobile) platforms are key. However, majority of sales are done in face to face channels (especially at branches) and those visits are declining at a rapid speed, making those windows for cross-selling more scarce. Good point.

  7. The growth or fall in the number of branches will depend on an individual bank. It won’t be uniform across the industry. Ergo Shevlin’s Law. I introduced a new metric called “Ideal Branch Count” in my blog post IBCO depends upon a number of factors internal and external to a bank. Overbranched banks will slash their branch network by 40% whereas underbranched banks will grow their branch network by 50%. I humbly submit that IBCO thus reconciles the differences surfaced by Shevlin’s Law.

Speak Your Mind


Show Comments