April 7, 2017
Authored by: Robert Klingler and Jonathan Hightower
(A print-ready version of this post is also available: Landscape of the U.S. Banking Industry.)
From 2006 through 2016, the number of insured depository institutions in the United States has fallen from 8,691 charters to 5,922, a decline of 2,769 charters or a 32% loss. This headline loss number is worth talking about, but is neither news nor new. The loss of charters is a frequent source of discussions around bank board rooms, stories from trade press, and chatter at banking conferences. The number of insured charters has also been in steady decline, with at least 33 years of declining numbers.
However, a deeper dive into the numbers reveals some unexpected trends below the headline 32% loss of charters.
Note: We’ve also recorded an accompanying podcast for The Bank Account on the Truth About Industry Consolidation. The podcast contains additional analysis to the numbers presented here, and is a useful addition, but not a substitute, to this content. In addition to listening to this episode, we encourage you to click to subscribe to the feed on iTunes, Android, Email or MyCast. It is also now available in the iTunes and Google Play searchable podcast directories.
Links to items mentioned in the podcast, or otherwise potentially of interest on the topic:
- The State of Banking in Georgia: 2015 vs. 2005
- The State of Banking in Atlanta: 2015 vs. 2005
- Economies of Scale Encourage Continued Consolidation
- American Banker story by Allison Prang on “Why some states are losing more banks than others” (subscription required)
- Wall Street Journal story on “Why You Probably Work for a Giant Company, in 20 Charts” (subscription required)
- Jamie Dimon’s Shareholder Letter.
State of Banking Landscape as of December 31, 2016
As of December 31, 2016, we had 5,922 institutions with $16.9 trillion in total assets.
The four largest depository institutions by asset size (JPMorgan, Wells Fargo, Bank of America and Citi) hold $6.84 trillion in assets, or 40.5% of the industry’s assets.
There are 111 additional banks that have assets greater than $10 billion, holding $6.98 trillion. That’s 1.9% of the total charters, holding 81.9% of the aggregate assets.
We then have 627 institutions with between $1 and $10 billion in assets. This group, which represents 10.6% of the charters, holds $1.79 trillion, or 10.6% of the industry’s assets.
87.5% of the bank charters in the United States are institutions with less than $1 billion in assets, with 663 institutions between $500 million and $1 billion and 4,517 institutions with less than $500 million in assets. The 663 institutions hold $470 billion in assets, while the smallest 4,517 banks hold an aggregate of 800 billion in assets, collectively representing less than 7.5% of the industry’s assets.
Comparing with 2006
At December 31, 2006, we had 8,691 insured institutions, with $11.9 trillion in assets. That’s a loss of approximately 2,800 charters in 10 years, or about 280 per year (and pretty close to one per work day). This is a 32% decline in the number of charters nationwide.
That matches almost exactly the 32% decline we previously identified as having left Georgia between 2005 and 2015 and the 30% decline in banks with branches in metropolitan Atlanta.
The overall allocations are similar, although slightly less concentrated. The four largest banks, with Wachovia preceding Wells Fargo, held $3.91 trillion, or 23.2% of the industry’s $11.9 trillion in assets. An additional 115 institutions with assets over $10 billion contributed another $5.07 trillion in assets. As such, institutions over $10 billion in assets held 53.2% of the industry’s assets (compared to 81.9% in 2016).
Smaller banks still dominated with the landscape, with 92.5% of the institutions holding less than $1 billion in assets, and 84.8% holding less than $500 million.
As a whole, the industry grew by almost $5 trillion in total assets, with banks over $10 billion representing $4.8 trillion of that increase.
A Loss of Smaller Banks
The overall 32% decline in number of charters was definitely not evenly distributed amongst all banks.
The number of banks larger than $1 billion increased slightly from 2006 though 2015, growing from 652 to 742, a 14% increase!
Similarly, the number of banks between $500 million and $1 billion remained virtually unchanged.
Accordingly, the entire 32% loss in overall charters, or nearly 2,800 charters, were lost in banks with less than $500 million in assets. Even amongst this group, the losses in charters were strongly correlated with asset size, with the smallest banks experiencing the sharpest declines.
Banks with between $250 million and $500 million in total assets declined by 85 charters, representing a 7% decline.
Banks with between $100 million and $250 million in total assets declined by 680 charters, representing a 27% decline.
Banks with less than $100 million declined by 2,091 charters, representing a 58% decline. A few further notes on this:
- There are still 1,542 banks in the United States with total assets of less than $100 million. That alone is a fairly amazing fact.
- Banks smaller than $100 million represented over 75% of the charters lost between 2006 and 2016.
- This group is probably the most affected by the dearth of de novo charters since 2010.
- Mathematically, the group cannot maintain this rate of loss; with 1,542 charters remaining, it can’t lose another 2,091 charters in the next ten year period.
However, asset size shouldn’t be the only factor in predicting whether an individual charter disappeared in the last ten years. The FDIC also separately published the number of charters that have elected Subchapter S tax treatment. Given the limitations on shareholders for Subchapter S institutions, these institutions are generally smaller community banks. Subchapter S institutions did decline, from 2,357 to 2,006, but that’s only a 15% decline.
One cause of the overall decline in number of charters that doesn’t get much attention, and shouldn’t necessarily be viewed as problematic, is the decline in organizations with multiple charters (or the consolidation of charters into a smaller number).
If one looks solely at the number of independent banking organizations (so counting a multi-bank holding company as one organization), the number of banking organizations fell from 7,528 organizations in 2006 to 5,522 in 2016. That represents a loss of 2,000 organizations, or a decline of 20%; still a large number, but not nearly as great as the 2,800 charters identified above.
In 2016, 632 charters were held by 241 multi-bank holding companies (representing 2.6 charters each).
In 2006, 1,670 charters were held by 518 multi-bank holding companies (representing 3.2 charters each).
That’s a 62% decline in the number of banking charters in a multi-bank holding company structure. That means internal reorganizations appear to have potentially accounted for the loss of almost 1,000 charters, potentially representing 37% of the decline over the last 10 years.
What Has Happened to our Small Banks?
Given the disproportionate losses to the numbers of the smallest banks over the last ten years, we immediately wondered the cause for all of the losses. Members of our group threw out differing opinions of the dominant factor: the most skeptical cited failures, while others cited merger activity and intra-company consolidations. The most optimistic opinion was that many of the companies were growing out of their size category.
Seeing the large numbers of banks that were lost among banks with less than $100 million in assets and those with between $100 million and $250 million, we realized that time would not allow us to research what happened to each institution that no longer exists. Therefore, we randomly selected samples (of at least 30, thanks to our statistics professors) of lost institutions from each of these two size categories.
Here’s what we found:
As set forth in the above chart, merger and consolidation activity certainly accounted for a large number of the losses in our sample groups, but it was far from the only contributor, accounting for 50%-60% of the reductions in each category. Growth was the next strongest factor, accounting for 42% of the number of banks leaving the under $100 million category. Given the financial crisis, failures accounted for 18% of the loss of banks in the $100-$250 million category and about 13% in the under $100 million category.
Both the merger/consolidation rate and the failure rate in each sample group were ahead of the merger/consolidation rate and the failure rate for the industry, as shown below.
|Sample Banks Under
|Sample Banks Between
$100 and $250 Million
|Ten-Year Merger Rates||47%||59%||29%|
|Ten-Year Failure Rates||13%||18%||7%|
How Good Was the Growth?
While the table above shows the pace of mergers and failures compared to the industry, what should we make of the growth rate of these small banks? FDIC-insured institutions grew at an average rate of 3.5% per year from the end of 2006 to the end of 2016. We asked ourselves, if these small banks grew at the average growth rate of the industry, how many should have grown out of their size range? To estimate that number, we applied this average industry growth rate to the entire population of banks in each size category as of December 31, 2016. We then calculated the percentage, based upon asset size of that date, that would have grown out of its size category with “normal” ten-year growth; i.e., the average growth of the industry. The chart below compares this “normal” growth rate to actual movements based on growth.
As shown above, the under $100 million group fared better growing independently than the $100 million to $250 million group did. A part of that can be attributed to the larger size range in the larger group ($150 million versus $100 million), and we acknowledge that many more banks in the smallest category start closer to the top end of the range given that no banks have $0 in assets. Finally, the larger merger rate in the $100 million to $250 million size category suggests that some of those banks chose to merge rather than continue to pursue independent growth. Even with that said, we found the growth of these microbanks to be impressive against the overall growth in the industry.
That conclusion, however, was not enough for our curious minds. Knowing that industry growth is disproportionately influenced by the largest banks in the country, we thought we would compare the growth of our microbanks against a sector that is considered to be a bit more flat in terms of size: the credit union industry.
During the measurement period, credit unions grew at an average rate of 6.2% per year (we’ll save the disparity between bank growth and credit union growth for another day). If we apply that rate of growth as the standard for our microbanks, about 44% of the banks in that category as of the end of 2006 should have grown out of the category, meaning that our sample group generally tracked the growth rate of credit unions.
Does it Matter Where You Are?
We next asked each other whether the location of the bank makes a difference in how it was likely to exit its size category over the last ten years. Our anecdotal understanding had been that banks headquartered in Metropolitan Statistical Areas (MSAs) are more likely to have growth opportunities available to them, while banks in rural areas appear to be drying up through mergers, consolidations and failures. The results were interesting.
First, in our sample, 90% of the banks between $100 million and $250 million in assets were headquartered in MSAs. With so few rural banks in the sample, further analysis would not have been helpful. However, in our sample of banks with assets below $100 million, 49% were headquartered in an MSA, a near perfect balance. These results support the general assumption that the smallest banks are more likely to be located in rural areas.
For our microbanks, MSA-headquartered banks were slightly under-represented in the merged/consolidated and growth categories, which again together constituted 87% or our sample. For the relatively small percentage of failures in the group, however, MSA-headquartered banks constituted 83% of the category, a large over-representation. This made some sense to us. During the crisis we saw a number of failures of very small banks in metro areas. These banks were not, in our view, true community banks but tended to be real estate-concentrated and brokered deposit funded small banks with no true community. In some cases, these were de novo banks that never had a chance to hit their stride, but the lesson remains: it is not a good idea to be a very small bank in a very large market unless you have a true niche to rely on as your community.
What Do We Make of This?
While many conclusions can be drawn from the data presented above, we believe most of them back conventional wisdom. The industry is “consolidating up,” meaning that the smallest of banks are shrinking as a portion of overall bank population. The smallest banks behave much like other microbusinesses throughout the U.S. economy – they are dynamic, constantly changing through growth, merger activity, and sometimes failure. We view these changes as a part of the natural life cycle of all small businesses, and none of these outcomes should be viewed as particularly problematic. Even in the case of failures, this undesired outcome can be viewed as the failure of experimental and entrepreneurial business models, a natural part of capitalism, and the losses to the deposit insurance fund from these failures are so minimal that they do not outweigh the benefit to the industry of allowing banks to try new approaches.
Dealing with banks of all sizes as we do, we see the unique benefits that various bank business models provide to the overall economy. Our largest bank clients are a source of sophisticated global financial solutions to some of the largest and most complicated businesses in the country while also providing widespread banking access to small businesses and consumers. Our smallest clients, on the other hand, tend to provide financial opportunities to small businesses and consumers that might not otherwise be able to access funding and financial advice from other providers. This benefit is particularly profound in rural communities that do not provide in-person access to a great number of bankers.
Given that we see a unique benefit to the overall economy from these small banks situated in rural communities, we should ask what our policymakers can do to encourage a reversal of the troubling trend toward their extinction. This effort, of course, would necessarily be focused on encouraging new market entrants; i.e., de novo banks, to be established in these communities.
The first step in that process is of course to understand the specific reasons behind the sharp decline in applications for de novo charters over in the recent past. We were encouraged by the House Financial Services Committee’s recent hearing to try to understand the root of that issue. As one might expect, opinions ranged from focusing on the impact of excessive regulation to the interest rate environment to the overall economy. We view all of those as being contributing factors and acknowledge that only some of them can be effectively influenced by regulators and policymakers. Many would agree, however, that this trend is a problem in need of a solution.
Depending on one’s political leanings, one might view the solution as decreasing regulation, particularly on new or small banks located in rural areas. Others might suggest the opposite approach of implementing strong and meaningful regulatory assistance for these institutions to ensure that they can afford the heavy burden of compliance and receive assistance rather than enforcement when they make a mistake. No matter the final approach, we hope to see an embrace of the value of a vibrant and diverse banking system that is able to serve the financial needs of all forms of businesses and consumers wherever located and with the delivery channels they desire.
What Do You Think?
As with any statistical analysis based on sampling, there can be flaws – thus the old saying that there are “lies, damn lies, and statistics.” We’re fond of this presentation, but it is based on the work of a couple of busy lawyers on nights and weekends. We strongly encourage an enterprising team of students or analysts to dig deeper.
More importantly, we would love to hear from you with regard to your thoughts on the takeaways. Feel free to find us on LinkedIn or twitter (Rob is @RobertKlingler and Jonathan is @HightowerBanks) and let us know your thoughts. These are important issues for the industry, and we think it is important for our community to think through them.