This page explains how every figure in the Market Intelligence resource is built: which sources are used, how geographies are defined, how each metric is calculated, and where the data should be read with caution. It is written for anyone who wants to check the work before relying on it.
Why 293 markets
A Combined Statistical Area is built from the Metropolitan Statistical Areas within it. Ranking both would count the same counties twice: San Jose and San Francisco would appear on their own and again inside the combined Bay Area market. To avoid that, the national ranking uses a non-overlapping universe:
- 181 Combined Statistical Areas. Each CSA is treated as a single market.
- 112 Metropolitan Statistical Areas that are not part of any CSA. These are included on their own because no combined market contains them.
Together these are the 293 markets in the ranking. Every county in the country is searchable separately, and each county is shown with the market it belongs to, or noted as being outside any ranked market.
One consequence is worth stating plainly: the Geography filter's standalone metro option returns only those 112 MSAs, not all U.S. metropolitan areas. An MSA that sits inside a CSA is represented through its parent market. This keeps the ranking honest but means the resource is not a complete list of every MSA.
Sources and vintages
| Data | Source | Vintage |
|---|---|---|
| Gross domestic product | U.S. Bureau of Economic Analysis, CAGDP1 county real GDP | 2024, chained 2017 dollars |
| Population | U.S. Census Bureau population estimates | Vintage 2025 |
| Deposits, institution counts, concentration | FDIC Summary of Deposits | June 30, 2025 |
| Market and county boundaries | U.S. Census Bureau CBSA/CSA delineation | July 2023 |
Because GDP is a 2024 measure and population is a 2025 estimate, GDP per capita pairs two different years. It is a useful comparative ratio between markets, not a precise figure for either year on its own.
Data dictionary
| Metric | Definition and calculation | Read with care |
|---|---|---|
| 2024 GDP | Real gross domestic product in billions of chained 2017 dollars. Market GDP is the sum of its component counties. | Chained dollars are inflation adjusted, so values are comparable across markets but are not current-dollar output. |
| 2025 Population | Census Bureau resident population estimate. Market population is the sum of its component counties. | Estimates are revised annually and are not a census count. |
| GDP per capita | Market GDP divided by market population. | Pairs 2024 output with 2025 population. Very high values often indicate production concentrated among few residents, such as energy or industrial markets. |
| Deposits | Total deposits reported by all FDIC-insured branches located in the geography, in billions. | Deposits are attributed to the branch where they are booked. Large custodial, brokered, credit-card, or centrally booked balances can make a market appear far larger than its local economy. |
| Institutions | Count of distinct FDIC-insured institutions with at least one branch reporting deposits in the geography. | Counts institutions with a physical branch presence. Lenders serving the market without a branch are not included. |
| Top-3 deposit share | Combined share of geography deposits held by its three largest institutions by reported deposits. | A measure of deposit concentration, not of banking competition. It excludes credit unions, nonbank lenders, loan production offices, and digital competitors. |
| Deposit trend | Total market deposits for each June 30 reporting date from 2020 through 2025, aggregated from branch-level records. | Reflects where deposits are booked in each year. A market can gain or lose balances through a merger or internal reorganization without any change in local activity. |
| Leading institutions | The ten institutions holding the most deposits in the market as of June 30, 2025, with branch counts and the change in that institution's deposits since 2020. | An acquirer absorbs the balances of banks it bought, so five-year growth can reflect acquisition rather than organic gain. Institutions with no 2020 presence are marked as new. |
| Deposits per branch | Market deposits divided by branch count, in millions. | Elevated where a market holds large centrally booked balances, so it is not a measure of branch productivity. |
| HHI | Herfindahl-Hirschman Index computed from every institution's deposit share in the market, on the standard 0 to 10,000 scale. | Computed on deposits only. It is not the product-market HHI used in regulatory merger review and should not be read as a supervisory measure. |
| Share of market GDP / deposits | A county's GDP or deposits as a percentage of its parent market total. | Small counties are rounded to one decimal, so shares within a market may not sum to exactly 100 percent. |
Deposit concentration bands
Markets are grouped into three descriptive bands by top-three deposit share:
| Band | Top-3 share |
|---|---|
| Fragmented | Under 45 percent |
| Moderate | 45 to 60 percent |
| Concentrated | Over 60 percent |
These bands are categories created for this resource. They are not regulatory classifications, they do not correspond to any supervisory threshold, and they do not measure every form of banking competition. They are a starting point for comparison, not a conclusion.
Known limitations
- Deposits reflect booking location. This is the single most important caveat. A market with a large credit-card or custodial bank will show deposits far out of proportion to local commercial activity. Sioux Falls is the clearest example in the data.
- Trends can reflect booking, not activity. This is the most important caveat on the six-year series. When a bank is acquired, its balances move to the acquirer's reporting, which can move a market's total sharply without any change in local lending or deposit gathering. Markets showing a large single-year move, or whose change is driven by one institution, carry a note in their profile identifying it. Two distinct patterns are separated: changes accompanied by a sharp shift in concentration, which indicate one holder entering or leaving, and broad market-wide moves with concentration roughly unchanged, such as the deposit inflows seen across many markets in 2020 and 2021.
- Five-year institution change is not organic growth. An acquirer absorbs the deposits of banks it purchased, so a large increase often reflects acquisition rather than competitive gain.
- Geography definitions change. Census delineations are revised periodically. A market's counties, and therefore its totals, can change between vintages.
- Counties outside ranked markets. Rural and micropolitan counties are searchable and carry their own economic profile, but are not part of the 293-market ranking.
Quality control
Before publication, the dataset is checked for internal consistency: that per-capita figures reconcile to their component GDP and population, that county totals sum to their parent market within rounding, that every county maps to a valid market, that deposit concentration values fall between zero and one hundred percent, that rank ordering matches the underlying metric, and that no field is null, zero, or negative where a value is required. Figures shown on the page are verified against the build output to the decimal.
Deposit and concentration figures are taken as published by the FDIC and have not been independently recomputed from branch-level records.
Update schedule
The FDIC publishes the Summary of Deposits each autumn for the preceding June 30. The resource is rebuilt after that release, with BEA and Census figures refreshed to their current vintages at the same time. The vintage line above the tool always states which releases are in use.
Corrections
If you find a figure that appears wrong, please say so. Accuracy matters more here than completeness, and a correction is genuinely welcome. Reach us by clicking here.
← Back to Market Intelligence