Technical writing
The gun dealer map: what ATF's Federal Firearms Licensee data shows and what it hides
ATF publishes a complete list of every licensed firearms dealer, manufacturer, importer, and collector in the United States—roughly 70,000 to 80,000 active licensees—as a free CSV download updated every month. No API key required. It is one of the more useful federal regulatory datasets that almost nobody outside the gun-violence research and investigative journalism community knows exists.
The dataset tells you who holds a federal firearms license and where they operate. It does not tell you which dealers' guns get traced to crimes. That gap is not accidental. It is the result of a two-decade-old appropriations rider called the Tiahrt Amendment.
The FFL system
Under the Gun Control Act of 1968, anyone in the business of dealing, manufacturing, or importing firearms must hold a federal license from the Bureau of Alcohol, Tobacco, Firearms and Explosives. The licensing requirement creates a regulatory surface that ATF can inspect, audit, and revoke. There are seven license types:
- Type 01 — Dealer in firearms and gunsmith. The most common type; covers retail gun stores, gun shows, and online dealers.
- Type 02 — Pawnbroker in firearms. Licensed to buy and resell used firearms accepted as collateral.
- Type 03 — Collector of curios and relics. Allows individuals to acquire C&R-classified antique and collectible firearms across state lines without going through a dealer.
- Type 06 — Manufacturer of ammunition for sale or distribution.
- Type 07 — Manufacturer of firearms for sale or distribution. Includes hobbyist-scale manufacturers; this category has grown substantially since 2010.
- Type 08 — Importer of firearms and ammunition.
- Type 11 — Importer of destructive devices, which includes large-bore weapons and certain explosive devices regulated under the National Firearms Act.
An FFL is required for commercial transactions. Private sales between individuals in the same state do not require a license or a background check under federal law, though some states impose additional requirements.
The data
ATF publishes the complete FFL list monthly at atf.gov/firearms/listing-federal-firearms-licensees/complete. The file is a pipe-delimited CSV, no authentication required. Fields included:
License_Type— one of the seven type codes aboveLicense_Name— licensee legal name (individual or business)Business_Name— trade name, if different from license namePremise_Street,Premise_City,Premise_State,Premise_Zip,Premise_County— physical location of the licensed premisesLicense_Regn— ATF field division region code (I through X)Expir_Date— license expiration date; active licenses expire on a three-year cycle
The Trace, a nonprofit newsroom covering gun violence, maintains historical monthly CSVs from January 2018 onward—useful for tracking licensee churn, new entrants, and revocations over time.
What it shows
At any given time the active FFL list contains roughly 70,000 to 80,000 licensees. Several patterns stand out:
Geographic concentration. Texas, Florida, California, Pennsylvania, and Ohio consistently rank as the top five states by raw FFL count. FFL counts also track political and population geography closely: rural counties have high FFLs per capita relative to urban counties, reflecting both demand patterns and the prevalence of home-based and small commercial dealers in rural areas.
Type 01 dominates. Dealer/gunsmith licensees (Type 01) account for roughly 50,000 of the active FFL universe at any point—about two-thirds of all licensees. Type 03 collector licenses represent a large share of the remainder.
Type 07 manufacturer growth. The number of Type 07 manufacturer licenses grew significantly after 2010. Some of this reflects a rise in small-scale commercial manufacturers; some reflects individuals acquiring Type 07 licenses to manufacture NFA-regulated items (suppressors, short-barreled rifles) for personal use, which the license permits.
What it hides
The FFL list tells you that a dealer exists and where it is. It does not tell you which dealers' guns are most frequently traced to crimes. That information is blocked by the Tiahrt Amendment.
ATF maintains the eTrace system: law enforcement agencies submit recovered crime guns for tracing back to the original point of sale. Every trace leads to a licensed dealer (or the last licensed dealer in the chain). This trace data is operationally useful for identifying dealers with unusually high rates of crime-gun recovery—a signal for straw purchasing, inadequate record-keeping, or theft.
Under the Tiahrt Amendment, ATF cannot release individual dealer-level trace data to the public. Law enforcement agencies get eTrace access for active investigations. Journalists and researchers get only state-by-state aggregate summaries, published annually in PDF format as ATF's Firearms Trace Data report. The PDFs show how many guns were traced in each state, the most common manufacturers and calibers recovered, and time-to-crime statistics. Individual dealer records: not available.
The Tiahrt Amendment backstory
Congress first attached the Tiahrt Amendment to a Justice Department appropriations bill in 2003. The immediate political context was a series of lawsuits filed by the mayors of Chicago, New York, and other cities against gun manufacturers and dealers, using ATF trace data to argue that specific dealers supplied a disproportionate share of crime guns. The amendment was designed specifically to block that evidentiary pathway.
Representative Virgil Goode (R-VA) sponsored the original rider. Representative Todd Tiahrt (R-KS) became its public champion, and the provision took his name. The NRA lobbied actively for the amendment. It has been renewed in every Justice Department appropriations cycle since 2003.
The practical effect: you can build a map of every licensed gun dealer in America down to the street address. You cannot build a map of which dealers have the highest crime-gun trace rates.
The theft and loss angle
ATF publishes an annual FFL Theft/Loss Report separately from the trace data. FFLs are required to report firearms stolen from or lost at their licensed premises. The annual report aggregates these at the license-type and ATF-region level—not at the individual dealer level—which preserves some privacy while still surfacing geographic and type-based patterns.
The 2024 report showed 3,842 FFLs reported theft or loss events totaling 12,711 firearms. Crossing that regional/type breakdown with the full FFL list by state and ATF region allows rough identification of geographic concentrations of theft and loss without individual dealer attribution. It is a partial substitute for the trace data the Tiahrt Amendment withholds.
Cross-reference opportunities
The FFL list supports several useful cross-joins with other federal datasets:
FFL counts + NICS check volume. The FBI's National Instant Criminal Background Check System publishes monthly check counts by state. States with higher FFLs per capita tend to show higher NICS check rates—the relationship is not causal but the correlation is consistent and the data supports state-level analysis. NICS data is available at fbi.gov/how-we-can-help-you/more-fbi-services-and-information/nics.
FFL counts + FBI UCR crime data. The academic literature on whether more FFLs per capita correlates with higher crime rates is mixed—no robust causal relationship has been established after controlling for confounders. The datasets support exploration, though, and the question is common enough in policy research that having the FFL geography available in a queryable form saves significant setup time.
FFL ZIP codes + Census demographic data. FFL density by county or ZIP, joined to Census population and income data, enables per-capita normalization and socioeconomic stratification of the dealer landscape.
Accessing the data
The raw monthly CSVs are available directly from ATF with no authentication. The Federal Regulatory Data Hub ingests the current month's file and exposes it as a structured API endpoint:
# All active FFLs (paginated) curl https://api.ai-analytics.org/datasets/atf-ffl # Filter by state curl "https://api.ai-analytics.org/datasets/atf-ffl?state=TX" # Filter by license type curl "https://api.ai-analytics.org/datasets/atf-ffl?license_type=01" # Filter by state and type curl "https://api.ai-analytics.org/datasets/atf-ffl?state=FL&license_type=07"
The endpoint returns structured JSON with the same fields as the ATF CSV, plus a normalized county FIPS code appended for Census joins.
For FEMA's National Flood Insurance Program claims data—another federal dataset with significant geographic concentration and policy implications: FEMA NFIP flood claims: mapping repetitive loss properties and the federal subsidy behind them →
For NLRB union election petition and outcome data by employer, industry, and region: NLRB union election data: petition filings, employer objections, and what the outcomes show →