Tony Wang16 min readBanking Deserts, Inverted: Rural States Have 5x More Bank Branches Per Capita
We mapped 62,892 US bank branches and 14,620 credit unions. Rural states have far more per capita than cities, and the 9 biggest banks are only 31% of listings.
"Banking desert" is a real, well-documented policy term — it usually means low-income or rural communities losing their last local bank branch to closures and consolidation. That's a genuine problem, and it's not what this data is built to measure directly. But when you actually map where bank branches sit today, relative to population, the geography that shows up is the opposite of what the phrase implies: it's dense, wealthy metro areas that are thin on branches per resident, and thinly-populated rural states that are thick with them.
Rural states have far more branches per capita
| Rank | State | Listings | Per 100k residents |
|---|---|---|---|
| 1 | North Dakota | 376 | 47.20 |
| 2 | Nebraska | 762 | 38.00 |
| 3 | South Dakota | 346 | 37.42 |
| 4 | Kansas | 1,088 | 36.63 |
| 5 | Iowa | 1,164 | 35.91 |
| 6 | Arkansas | 1,088 | 35.23 |
| 7 | Montana | 339 | 30.21 |
| 8 | Wyoming | 175 | 29.78 |
| 9 | Missouri | 1,858 | 29.75 |
| 10 | Vermont | 178 | 27.45 |
| … | (30 states omitted) | ||
| 41 | Washington | 1,224 | 15.38 |
| 42 | Oregon | 650 | 15.21 |
| 43 | Maryland | 942 | 15.04 |
| 44 | Georgia | 1,690 | 14.92 |
| 45 | Florida | 3,388 | 14.50 |
| 46 | Utah | 452 | 12.90 |
| 47 | California | 4,934 | 12.51 |
| 48 | Nevada | 373 | 11.42 |
| 49 | Arizona | 778 | 10.26 |
| 50 | Hawaii | 134 | 9.34 |
Show the full 50-state ranking
| Rank | State | Per 100k residents |
|---|---|---|
| 1 | North Dakota | 47.20 |
| 2 | Nebraska | 38.00 |
| 3 | South Dakota | 37.42 |
| 4 | Kansas | 36.63 |
| 5 | Iowa | 35.91 |
| 6 | Arkansas | 35.23 |
| 7 | Montana | 30.21 |
| 8 | Wyoming | 29.78 |
| 9 | Missouri | 29.75 |
| 10 | Vermont | 27.45 |
| 11 | Kentucky | 27.26 |
| 12 | West Virginia | 26.95 |
| 13 | Mississippi | 25.99 |
| 14 | Wisconsin | 25.69 |
| 15 | Oklahoma | 25.13 |
| 16 | Louisiana | 24.49 |
| 17 | District of Columbia | 23.92 |
| 18 | Minnesota | 23.53 |
| 19 | Massachusetts | 22.72 |
| 20 | Illinois | 22.50 |
| 21 | Tennessee | 21.93 |
| 22 | New Hampshire | 21.65 |
| 23 | Alabama | 21.58 |
| 24 | Maine | 21.44 |
| 25 | Connecticut | 20.98 |
| 26 | Ohio | 20.56 |
| 27 | New Mexico | 20.00 |
| 28 | Delaware | 19.87 |
| 29 | Colorado | 19.45 |
| 30 | New Jersey | 19.26 |
| 31 | Virginia | 19.18 |
| 32 | Pennsylvania | 18.45 |
| 33 | Indiana | 18.02 |
| 34 | Rhode Island | 17.43 |
| 35 | Idaho | 17.34 |
| 36 | Texas | 17.01 |
| 37 | South Carolina | 16.65 |
| 38 | New York | 16.08 |
| 39 | Michigan | 15.49 |
| 40 | North Carolina | 15.43 |
| 41 | Washington | 15.38 |
| 42 | Oregon | 15.21 |
| 43 | Maryland | 15.04 |
| 44 | Georgia | 14.92 |
| 45 | Florida | 14.50 |
| 46 | Utah | 12.90 |
| 47 | California | 12.51 |
| 48 | Nevada | 11.42 |
| 49 | Arizona | 10.26 |
| 50 | Hawaii | 9.34 |
The shape makes sense once you think through why a branch exists at all: it has to sit within driving distance of the people who use it. A city can serve a huge population from a handful of downtown branches; a sparsely-populated state can't consolidate the same way without leaving whole towns with no branch at all, so it ends up with proportionally more locations per resident even though its absolute count is small. That's not evidence of generous banking access in North Dakota so much as evidence of geography forcing a different branch-to-population ratio — but it directly inverts the mental picture "banking desert" coverage usually paints, where rural areas are assumed to be the ones losing out.
Credit unions show the same shape — plus a Michigan-specific twist
| Rank | State | Listings | Per 100k residents |
|---|---|---|---|
| 1 | North Dakota | 89 | 11.17 |
| 2 | Montana | 104 | 9.27 |
| 3 | Michigan | 906 | 8.93 |
| 4 | Maine | 119 | 8.59 |
| 5 | Idaho | 166 | 8.29 |
| 6 | Wisconsin | 499 | 8.27 |
| 7 | Iowa | 250 | 7.71 |
| 8 | Vermont | 47 | 7.25 |
| 9 | Alaska | 53 | 7.23 |
| 10 | Alabama | 366 | 7.10 |
| … | (30 states omitted) | ||
| 41 | Mississippi | 100 | 3.40 |
| 42 | Georgia | 351 | 3.10 |
| 43 | Nevada | 97 | 2.97 |
| 44 | California | 1,143 | 2.90 |
| 45 | New York | 563 | 2.83 |
| 46 | Hawaii | 39 | 2.72 |
| 47 | Arizona | 202 | 2.66 |
| 48 | Arkansas | 73 | 2.36 |
| 49 | Maryland | 148 | 2.36 |
| 50 | New Jersey | 154 | 1.62 |
Show the full 50-state credit union ranking
| Rank | State | Per 100k residents |
|---|---|---|
| 1 | North Dakota | 11.17 |
| 2 | Montana | 9.27 |
| 3 | Michigan | 8.93 |
| 4 | Maine | 8.59 |
| 5 | Idaho | 8.29 |
| 6 | Wisconsin | 8.27 |
| 7 | Iowa | 7.71 |
| 8 | Vermont | 7.25 |
| 9 | Alaska | 7.23 |
| 10 | Alabama | 7.10 |
| 11 | Utah | 6.94 |
| 12 | Washington | 6.17 |
| 13 | Indiana | 6.07 |
| 14 | Kansas | 6.03 |
| 15 | New Mexico | 5.82 |
| 16 | Oregon | 5.59 |
| 17 | Louisiana | 5.39 |
| 18 | Rhode Island | 5.38 |
| 19 | District of Columbia | 5.27 |
| 20 | Minnesota | 4.99 |
| 21 | Tennessee | 4.93 |
| 22 | Colorado | 4.85 |
| 23 | Missouri | 4.79 |
| 24 | North Carolina | 4.69 |
| 25 | Nebraska | 4.64 |
| 26 | Oklahoma | 4.54 |
| 27 | South Carolina | 4.42 |
| 28 | South Dakota | 4.33 |
| 29 | Ohio | 4.28 |
| 30 | Delaware | 4.28 |
| 31 | Texas | 4.27 |
| 32 | New Hampshire | 4.26 |
| 33 | Virginia | 3.98 |
| 34 | Connecticut | 3.97 |
| 35 | Pennsylvania | 3.94 |
| 36 | Massachusetts | 3.84 |
| 37 | Illinois | 3.75 |
| 38 | Florida | 3.66 |
| 39 | West Virginia | 3.62 |
| 40 | Kentucky | 3.57 |
| 41 | Mississippi | 3.40 |
| 42 | Georgia | 3.10 |
| 43 | Nevada | 2.97 |
| 44 | California | 2.90 |
| 45 | New York | 2.83 |
| 46 | Hawaii | 2.72 |
| 47 | Arizona | 2.66 |
| 48 | Arkansas | 2.36 |
| 49 | Maryland | 2.36 |
| 50 | New Jersey | 1.62 |
North Dakota tops both rankings, which is a clean, independent confirmation of the rural-density thesis — two entirely different categories, same top state, same underlying reason. But Michigan breaks the pattern in an interesting way: it's an unremarkable 39th for bank branches (15.49 per 100k, below the national average) yet 3rd for credit unions (8.93 per 100k, more than double the national average of 4.30). That gap isn't about Michigan being rural or under-banked — it's a specific regional culture. Michigan has one of the strongest credit union traditions in the country, rooted in the auto industry: DFCU Financial, one of the state's largest credit unions and a top domain in our sample, began life as the Dearborn Federal Credit Union serving Ford employees. When an employer that size anchors a credit union for generations, it shows up in exactly this kind of data as a state-level outlier that has nothing to do with population density.
Branch banking barely resembles the rest of this series
Every other industry we've mapped in this series — funeral homes, veterinary clinics, dental practices, hair salons — has shown at least one dominant hidden or visible chain controlling a meaningful, sometimes majority, share of locations. Branch banking looks completely different.
| Chain | Share of a 500-record sample |
|---|---|
| Bank of America | 7.6% |
| US Bank | 6.2% |
| Chase | 5.0% |
| PNC | 2.6% |
| Huntington | 2.2% |
| Regions | 2.0% |
| BMO | 2.0% |
| Wells Fargo | 2.0% |
| Truist | 1.6% |
| Top 9 banks combined | 31.2% |
| Long tail (hundreds of community/regional banks) | 68.8% |
| Credit union | Share of a 500-record sample |
|---|---|
| Security Service FCU | 2.6% |
| Indiana Members CU | 2.0% |
| Navy Federal | 1.4% |
| CommunityAmerica CU | 1.2% |
| Everwise CU | 1.2% |
| DFCU Financial | 1.2% |
| Credit Union of Colorado | 1.2% |
| GreenState CU | 1.0% |
| Addition Financial CU | 0.8% |
| Top 9 credit unions combined | 12.6% |
| Long tail (hundreds of local/regional credit unions) | 87.4% |
Two things stand out. First, even the single biggest name in each category is small in absolute terms — Bank of America at 7.6% is the largest share anywhere in either table, nowhere close to Service Corporation International's 15.8% single-company share of funeral homes elsewhere in this series. Second, credit unions are noticeably more fragmented than banks even at the very top: Navy Federal, the largest US credit union by assets and membership, is only 1.4% of the sample by physical branch count. That's the clearest assets-vs-branches gap in the whole dataset — Navy Federal serves millions of members primarily through digital banking and a much smaller physical footprint relative to its size, so branch count badly understates how big it actually is. The reverse is also true industry-wide: bank asset concentration is a well-documented, separate story (a handful of megabanks hold a large share of US banking assets) that this branch-count data simply isn't built to measure — what we can say cleanly is that physical branch presence, specifically, remains remarkably unconsolidated.
The BB&T-to-Truist rebrand, finishing live in the data
That last point is also a genuinely new contamination case for this series' running list: q="BB&T" doesn't primarily fail because "Bank" is too generic (the trap that broke q="Bank of America" and q="PNC Bank" elsewhere in this piece) — it fails because the literal substring "&T" also appears inside completely unrelated company names. It's a small, specific reminder that full-text search on a name field can be defeated by punctuation patterns just as easily as by common words.
How we did this, and the caveats
Every number above comes from a live search/facets call against Crawlora's Google Maps business dataset, 2026-08-27 snapshot, cross-referenced against US Census Bureau Vintage 2025 (July 1, 2025) state population estimates.
| What | Detail |
|---|---|
| Per-capita rankings | category=bank (62,892 listings) and category=credit_union (14,620 listings), each summed across 50 of 51 US jurisdictions |
| Chain-penetration | Two separate 500-record national samples (has_website=true, sort=updated_at_desc, deduplicated), one per category, each classified by website domain |
| Contamination check | q="Bank of America", q="U.S. Bank", and q="PNC Bank" all returned totals suspiciously close to each other and to unrelated brands, with unrelated bank names in their samples — confirmed unusable, not used for any number above |
| Merger check | q="Truist" (1,543, clean), q="SunTrust" (1 stale listing), q="BB&T" (2 genuine of 20 sampled, rest contamination) |
Caveats worth stating plainly: the chain-penetration percentages are sample-based estimates, not an exhaustive census of 62,892 bank and 14,620 credit union listings — a different 500-record sample would move individual small-bank shares by a point or two, though the overall "far less consolidated than everything else in this series" shape is stable across everything we checked. This measures physical branch locations, not assets, deposits, or membership — Navy Federal's low branch share is the clearest example of why that distinction matters; a bank or credit union can be enormous by any financial measure while remaining small by branch count. Alaska is missing from the bank facet and Wyoming from the credit union facet — both jurisdictions likely just have too few listings to clear whatever threshold the facet aggregation uses, not a data quality issue specific to this piece. And as with every study on this dataset: it's live-growing, so a re-pull months from now will show different totals.
If you want to run your own consolidation or density checks on this data, that's exactly what Crawlora's Google Maps business API is for — the same structured-data approach behind our funeral home, veterinary, and dental studies.
Query 132M+ Google Maps business listings directly
Crawlora's Google Maps dataset turns business listings into structured, queryable JSON — filter and facet by category, country, state, county, city, rating, and website status. 2,000 free credits a month, no card.
Frequently asked questions
Which state has the most bank branches per capita?
North Dakota, at 47.20 bank branches per 100,000 residents — 5x Hawaii's 9.34, the lowest in the country. North Dakota also leads the credit-union ranking (11.17 per 100k), making it the only state to top both categories.
Is 'banking desert' the right way to describe rural America based on this data?
Not by branch density. Rural, sparsely-populated states (North Dakota, Nebraska, South Dakota, Kansas, Iowa, Arkansas) have the most bank branches per capita in the country; dense Sun Belt and coastal states (Hawaii, Arizona, Nevada, California) have the fewest. That's the opposite of what the 'banking desert' framing implies, though it doesn't contradict documented local branch closures in specific low-income communities — this data measures nationwide density, not access trends within any one place.
How consolidated is bank branch ownership compared to other industries?
Far less than everywhere else in this series. A 500-record sample classified by website domain found the 9 biggest banks (Bank of America, US Bank, Chase, PNC, Huntington, Regions, BMO, Wells Fargo, Truist) combined for just 31.2% of branches — versus, for example, Service Corporation International's 15.8% share alone in funeral homes. Credit unions are even less consolidated: the top 9 combined are only 12.6%.
Why is Navy Federal Credit Union's branch share so low if it's the largest US credit union?
Because branch count and financial size measure different things. Navy Federal is the largest US credit union by assets and membership, but only 1.4% of our credit-union sample by physical branch count — it serves most members digitally rather than through a proportionally large branch network. It's the clearest example in this data of why branch presence understates an institution's true scale.
Why couldn't you just search 'Bank of America' or 'PNC Bank' directly to measure their branch counts?
Because the word 'Bank' is common enough within category=bank that pairing it with almost any second word barely narrows the search — q='Bank of America', q='U.S. Bank', and q='PNC Bank' all returned totals suspiciously close to each other and to unrelated brands, with unrelated bank names showing up in their samples. We used a large sample classified by actual website domain instead, the same method proven throughout this series.