FEMA Firefighter Grants: The Federal Record of Funding the Fire Service

Most of what FEMA records is about a disaster that has already happened—the declaration, the debris, the rebuilt firehouse. Firefighter grants are the part of FEMA that pays for the ordinary day: the self-contained breathing apparatus a crew straps on before walking into a structure fire, the turnout gear that keeps the heat off, the new pumper that replaces a thirty-year-old engine, the training that turns a volunteer into a certified firefighter, and—through a separate program—the salaries that keep enough firefighters on the truck to do the job safely. OpenFEMA publishes that everyday-readiness record as roughly 74,000 grant awards: the federal map of which fire and EMS departments the country chooses to equip and staff.

This article covers what the firefighter-grants dataset is and how it differs from the rest of FEMA's disaster-centric data; the history of the Assistance to Firefighters Grants program and why Congress created a federal stream that funds local departments directly; the wider program family—AFG itself, SAFER for staffing, Fire Prevention and Safety for prevention and research, and EMPG for emergency management; the competitive, peer-reviewed award process and the cost share that scales with community size; the kinds of things the money buys, from breathing apparatus to apparatus to firefighter health and wellness; the structural divide between funding equipment and funding people; how the dollars reach volunteer and rural departments that have no other way to buy a fire truck; how the data joins to FEMA's disaster declarations, Public Assistance, and hazard-mitigation records; a Python workflow that pulls awards from OpenFEMA and sums the federal share by program and by state; and the caveats—award-level grain, program-family mixing, and the gap between an award and the outcome it buys—that every analyst must internalize before drawing conclusions.

What the dataset is

FEMA's grant programs come in two broad kinds. The disaster grants—Public Assistance, Individual Assistance, Hazard Mitigation Assistance—flow only after a presidentially declared disaster, and they dominate the public picture of what FEMA does. The non-disaster, or preparedness, grants flow every year regardless of whether a disaster strikes; they exist to build capability before the emergency, not to reimburse the cost of one after the fact. The firefighter grants are the best-known of the non-disaster grants, and they share a defining feature that sets them apart from most federal money: rather than passing through a state agency, the Assistance to Firefighters Grants (AFG) program awards money directly to the local fire department or EMS organization that applied for it.

In our database this record is stored as the table fema_afg_grants, with the grain of one row per award: a department that wins an equipment grant one year and a vehicle grant three years later contributes two rows. The roughly 74,000 awards span the firefighter-grant programs and their sibling non-disaster preparedness grants. The columns—drawn directly from the OpenFEMA Non-Disaster and Assistance to Firefighter Grants dataset—identify the recipient, where it is, which program funded it, and how much federal money the award carried:

award_number          -- the FEMA grant award identifier (e.g. EMW-YYYY-FG-NNNNN)
fiscal_year           -- the federal fiscal year of the award
program_name          -- Assistance to Firefighters Grants / SAFER /
                      --   Fire Prevention and Safety (and sibling
                      --   non-disaster preparedness programs)
program_abbreviation  -- short program code, where populated
vendor_name           -- the recipient: the fire department or EMS organization
vendor_state          -- recipient state (two-letter postal code)
region                -- the FEMA region (1-10) the recipient sits in
award_amount          -- the federal share obligated for the award
id                    -- OpenFEMA record UUID

The vendor_name and award_amount are the load-bearing columns: together they answer the program's central question—which department received how much federal money. The program_name is what lets an analyst separate the firefighter programs from the rest of the non-disaster grant family that shares this dataset, and it is the field on which the equipment-versus-staffing distinction turns, because AFG and SAFER are different program_name values. The vendor_state and region place the award geographically, enabling the per-state and per-region rollups that show how the federal investment is distributed across the country. The award_number is the persistent identifier that, in the fuller FEMA grant records, ties the headline award to the activities and equipment it funded and to the cost-share terms. It is worth stressing what the dataset is and is not: it is a record of awards made, not of equipment delivered, lives saved, or response times improved—the gap between the dollar and the outcome is the central interpretive caution of the whole dataset.

What it is and the AFG program history

For most of American history the fire service was a purely local responsibility, paid for out of municipal budgets and, for the volunteer departments that protect most of the country's land area, out of pancake breakfasts, donations, and bond issues. That model left a structural gap: the smallest and poorest communities, which often face the same fires and rescues as wealthy suburbs, frequently could not afford modern protective gear, safe apparatus, or adequate training. The Assistance to Firefighters Grants program, first funded in 2001, was Congress's answer—a federal grant stream aimed squarely at improving the safety and readiness of first responders by helping local departments buy the things their own budgets could not stretch to cover.

What distinguished AFG from the start was its directness. Most federal aid to localities passes through state government, which adds a layer of allocation and politics between the money and the department. AFG awards money straight to the fire department or EMS organization that wrote the application, evaluated on the merits of that application against national program priorities. The program was conceived as a way to put resources where the need and the risk are greatest rather than where the state formula happens to send them, and from its origins it placed deliberate emphasis on the volunteer and combination departments that protect the large majority of the nation's communities and that have the least independent capacity to fund themselves. After the 2001 attacks and the subsequent reorganization of federal emergency functions, AFG was folded into the Department of Homeland Security and ultimately administered by FEMA, where it sits today within the agency's preparedness grant portfolio—the everyday-readiness counterpart to FEMA's disaster-response machinery.

The program family: AFG, SAFER, FP&S, and EMPG

“Firefighter grants” is best understood as a family rather than a single program, because Congress addressed different parts of the readiness problem with different grants, and the analytically important distinctions live in the program_name field. Treating the family as one undifferentiated pool of money obscures the most interesting structural fact about it—that it funds two fundamentally different things, equipment and people, under separate authorities and on separate terms.

Assistance to Firefighters Grants (AFG) is the equipment-and-readiness core. It funds firefighting and rescue equipment, personal protective equipment—turnout gear and self-contained breathing apparatus chief among them—apparatus and vehicles such as pumpers and tankers, training and certification, and a growing emphasis on firefighter health and wellness, including the cancer-screening and behavioral-health programs that have become a priority as the occupational hazards of the job have come into sharper focus. Staffing for Adequate Fire and Emergency Response (SAFER) funds something AFG deliberately does not: the hiring and retention of firefighters themselves. SAFER grants pay the salaries and benefits needed to put more firefighters on the truck— to meet minimum safe-staffing standards and to keep volunteer departments from collapsing as their ranks age out—and so represent the people half of readiness, as opposed to AFG's equipment half.

Two further programs round out the federal stream. The Fire Prevention and Safety (FP&S) grants fund the other end of the problem—stopping fires before they start, and making the firefighting itself safer. FP&S supports community fire-prevention programs, smoke-alarm and public-education campaigns, and, importantly, firefighter-safety research and development, the studies that have reshaped how the service thinks about cancer risk, cardiac events, and the design of protective equipment. Separately, the Emergency Management Performance Grants (EMPG) support state and local emergency management—the planning, coordination, and personnel that hold the broader preparedness system together—rather than the fire service specifically. Together AFG, SAFER, FP&S, and EMPG constitute a major, recurring federal investment in the everyday capacity of the fire and emergency services, distinct in purpose and rhythm from the disaster grants that flow only after catastrophe.

The competitive award and the community-scaled cost share

Firefighter grants are competitive, not formula-based. A department does not receive an annual entitlement; it writes an application describing the need it wants to address—a specific piece of equipment, a class of protective gear, a number of staff—and that application competes against thousands of others for a finite pool of money. The applications are scored through a peer-review process in which working fire-service professionals evaluate requests against the year's published program priorities, which weight factors such as the applicant's call volume, the risk it faces, the gap the grant would fill, and the cost-effectiveness of the request. The peer-review design is deliberate: it puts the funding decisions in the hands of people who understand the operational reality of the fire service rather than leaving them to a mechanical formula, and it is why the dataset reflects assessed need and application quality, not just population or geography.

Every award also carries a non-federal cost share: the recipient must contribute a matching portion of the project's total cost from its own resources. The defining feature of the firefighter-grant cost share is that it scales with community size. The smallest communities face the lowest match requirement—a recognition that a rural volunteer department serving a few hundred people simply cannot raise the same proportion of a fire truck's cost as a city utility can—while larger jurisdictions, with broader tax bases, are asked to contribute a higher share. The graduated match is one of the program's most consequential equity mechanisms, because it is what makes a grant actually usable by the departments the program most wants to reach. For the analyst, the cost-share design is a reminder that the award_amount in the dataset is the federal share, not the full project cost: the true value of the funded project is larger, by the matching contribution, and the match itself varies systematically with the size of the community the recipient serves.

What the money buys: equipment, apparatus, training, and wellness

The activities a firefighter grant funds map closely to the hazards of the job, and the program's priorities have shifted over time as the understanding of those hazards has deepened. The most frequently funded category is personal protective equipment: the turnout coats and pants, helmets, gloves, and—above all—the self-contained breathing apparatus (SCBA) that lets a firefighter breathe in a smoke-filled structure. SCBA is expensive, has a finite service life, and is non-negotiable for interior firefighting, which makes it a perennial grant request for departments that cannot afford to replace an entire complement of air packs on a municipal budget. Closely related are the firefighting and rescue equipment grants—hoses, nozzles, thermal-imaging cameras, extrication tools—and the apparatus and vehicle grants that fund the single largest purchases a department makes, the pumpers, tankers, and rescue vehicles whose cost can dwarf an entire small department's annual budget.

Beyond hardware, the grants fund training and certification—the courses that turn a willing volunteer into a competent, safe firefighter or a certified EMS provider—and an increasingly prominent category of firefighter health and wellness. The recognition that firefighters face elevated rates of occupational cancer, cardiac disease, and behavioral-health crises has pushed FEMA to fund medical screening, fitness and wellness programs, and the equipment and procedures—such as gross decontamination of contaminated gear—that reduce chronic exposure to the carcinogens released in a fire. This evolution is one of the more revealing things visible in the long arc of the data: the same program that began by buying air packs and pumpers now also invests in keeping firefighters healthy across a career, reflecting a broader shift in how the fire service understands what it means to keep its people safe.

Equipment versus staffing, and the volunteer and rural fire service

The single most analytically interesting cut of the firefighter-grant data is the split between funding equipment and funding people. AFG buys things; SAFER pays salaries. The two answer different failure modes of the fire service—a department can have firefighters but no working air packs, or working air packs but too few firefighters to crew the truck safely—and the balance between them in any state or period is a meaningful signal about where the system's constraint sits. Equipment dollars are one-time capital that leaves a department better equipped indefinitely; staffing dollars are recurring operating support that a department must eventually replace from its own budget when the grant ends, which makes SAFER both more valuable and more fraught for the recipient. Tracking the equipment-versus-staffing share over time, and across states, is one of the cleanest questions the dataset supports, and it speaks directly to a long-running policy debate about whether federal money should buy gear or hire people.

The other defining theme is reach into the volunteer and rural fire service. The United States is protected overwhelmingly by volunteer and combination departments, especially across rural America, and those departments are precisely the ones with the least independent capacity to buy a fire truck or replace an aging set of breathing apparatus. The graduated cost share, the program priorities that weight need and risk, and the direct-to-department design all point the money toward them. Whether the dollars actually reach the small, volunteer, rural departments—rather than concentrating in the larger career departments that have the grant-writing capacity to win competitive awards—is one of the most important equity questions the data can be made to answer, and it is a question that requires care, because the very smallest departments are also the ones least likely to have someone on staff who can write a competitive application in the first place. The federal record of who is funded is, read closely, also a record of who has the capacity to ask.

Joining to FEMA's disaster, recovery, and mitigation data

The firefighter-grants table is most valuable not in isolation but as one facet of the integrated FEMA grant and disaster record, and it joins to the rest of the family along two axes: geography (state and FEMA region) and the recipient identity. The firefighter grants are the everyday-readiness layer; the disaster grants are the catastrophe layer; reading them together is what turns a list of awards into a picture of a community's relationship with the federal emergency system.

The most natural companion is the disaster declarations record, which catalogs where and when the federal government has declared emergencies and major disasters. Joining firefighter-grant awards to declarations by state and county lets an analyst ask whether preparedness investment tracks disaster exposure—whether the departments in the most frequently declared corners of the country are also the ones being equipped and staffed, or whether the everyday-readiness money flows on a logic of its own. The Public Assistance record, which documents how disaster recovery money is actually spent after a declaration, is the downstream complement: a fire station damaged in a flood may be rebuilt with Public Assistance dollars, while the same department is equipped, year over year, with AFG awards—two FEMA streams meeting at the same firehouse. And the Hazard Mitigation record, which funds work to prevent the next disaster, sits alongside the firefighter grants as the other preventive program: mitigation reduces the disaster that arrives, while firefighter grants improve the response to the disasters that arrive anyway. Together the four datasets— declarations, Public Assistance, hazard mitigation, and firefighter grants—span the full FEMA lifecycle, from preparing for an emergency, through declaring it, to recovering from it and guarding against the next one.

Analytical uses

A national, recipient-resolved, dollar-stamped record of federal investment in the fire service supports a distinctive set of analyses, most of which turn on the program and geography fields.

Federal investment by state and region is the most immediate use: summing the federal share by vendor_state and region shows where the firefighter-grant money lands, and—normalized by population, by number of departments, or by disaster exposure—whether the distribution matches need. The equipment-versus-staffing splitexploits the program_name field to measure the balance between AFG capital and SAFER staffing in any state or period, the cleanest read the data offers on where the system's binding constraint sits.

Reach into volunteer and rural departments brings the recipient identity to bear: classifying recipients by the kind of department they are and the community they serve—and weighting by the federal share they receive—tests whether the program's equity design is working in practice. Finally, the long-run shift in program priorities uses the fiscal-year field to trace how the federal idea of readiness has changed—from a program that began by buying air packs and pumpers to one that now also funds firefighter health, wellness, and safety research—a story the data tells decade over decade in the changing mix of what the money buys.

Python workflow: firefighter grants from the OpenFEMA API

The script below pulls non-disaster firefighter-grant awards from FEMA's OpenFEMA REST API, filters to the three firefighter programs (AFG, SAFER, and Fire Prevention and Safety), and computes the headline metrics: total federal share, the breakdown by program, the top states by dollars, and the equipment-versus-staffing split between AFG and SAFER. No API key is required for public data. The dataset is the Non-Disaster and Assistance to Firefighter Grants entity, which also carries the other non-disaster preparedness grants—Port Security, Homeland Security, EMPG, and the rest—so the filter on program_name is what isolates the firefighter stream; any production use should be validated against the current OpenFEMA data dictionary and should page through the full result set.

import requests, pandas as pd

# OpenFEMA REST API -- no API key required for public data.
# The Non-Disaster and Assistance to Firefighter Grants dataset holds one
# row per award, keyed by an award number, and carrying the program name
# (Assistance to Firefighters Grants / SAFER / Fire Prevention and Safety,
# alongside the other non-disaster preparedness grants), the recipient
# (vendorName), the recipient state (vendorState, a 2-letter postal code),
# the FEMA region, the fiscal year, and the federal award amount.
BASE = "https://www.fema.gov/api/open"
AFG = f"{BASE}/v1/NonDisasterAssistanceFirefighterGrants"

# The three firefighter-grant programs in the dataset's programName field.
FIRE_PROGRAMS = [
    "Assistance to Firefighters Grants",
    "Staffing for Adequate Fire and Emergency Response (SAFER)",
    "Fire Prevention and Safety",
]


def fetch_all(url, params=None, page=10000):
    # OpenFEMA pages with $skip / $top; loop until a short page comes back.
    params = dict(params or {})
    params["$top"] = page
    out, skip = [], 0
    while True:
        params["$skip"] = skip
        r = requests.get(url, params=params, timeout=120)
        r.raise_for_status()
        body = r.json()
        # The records live under a key named for the entity; grab the list.
        key = next(k for k, v in body.items() if isinstance(v, list))
        rows = body[key]
        if not rows:
            break
        out.extend(rows)
        if len(rows) < page:
            break
        skip += page
    return out


def analyze():
    df = pd.DataFrame(fetch_all(AFG))
    if df.empty:
        print("No firefighter-grant awards returned.")
        return

    # Keep only the three firefighter programs; the dataset also carries
    # Port Security, Homeland Security, EMPG, and other non-disaster grants.
    fire = df[df["programName"].isin(FIRE_PROGRAMS)].copy()
    fire["awardAmount"] = pd.to_numeric(fire["awardAmount"], errors="coerce").fillna(0)

    print(f"${fire['awardAmount'].sum():,.0f} in federal share across "
          f"{len(fire):,} firefighter-grant awards")

    # --- Federal share by program (AFG / SAFER / FP&S) ------------------
    by_prog = fire.groupby("programName")["awardAmount"].sum().sort_values(ascending=False)
    print("  Federal share by program:")
    for p, amt in by_prog.items():
        print(f"    {str(p)[:46]:<46} ${amt:>15,.0f}")

    # --- Top states by firefighter-grant dollars -----------------------
    by_state = fire.groupby("vendorState")["awardAmount"].sum().sort_values(ascending=False)
    print("  Top 10 states by federal share:")
    for s, amt in by_state.head(10).items():
        print(f"    {str(s):<6} ${amt:>15,.0f}")

    # --- Equipment / readiness (AFG) vs. staffing (SAFER) --------------
    afg = fire.loc[fire["programName"] == FIRE_PROGRAMS[0], "awardAmount"].sum()
    safer = fire.loc[fire["programName"] == FIRE_PROGRAMS[1], "awardAmount"].sum()
    total = afg + safer
    if total:
        print(f"  Equipment/readiness (AFG): {afg / total:.1%}  |  "
              f"Staffing (SAFER): {safer / total:.1%}")
    return fire


analyze()

Two practical notes apply. First, the award_amount the script sums is the federal share, not the full project cost; because every award carries a non-federal cost share that scales with community size, the total project value is larger than the federal total, and the gap is itself uneven across recipients—so a state-by-state comparison of federal dollars understates, by a varying amount, how much fire-service investment the program actually catalyzed. Second, the equipment-versus-staffing ratio in the script is a coarse first pass that compares the AFG and SAFER program totals; a finer analysis would reach into the underlying FEMA grant records, which carry the activity and equipment detail beneath each award, to separate—within AFG—protective equipment from apparatus from training from wellness. For national-scale work, FEMA's OpenFEMA bulk data downloads (the full CSV and Parquet files) are far more efficient than thousands of paginated API calls and ship with the authoritative, version-stamped column definitions for the release.

Limitations and analytical caveats

The firefighter-grants dataset is the most comprehensive public record of federal investment in the fire service, but it carries structural limitations that an analyst must internalize before drawing conclusions from it.

The grain is the award, and the dataset mixes programs.One row is one award, not one department, one fire truck, or one outcome; a department that wins several grants appears several times, and naive counting will conflate awards with recipients. Equally important, the underlying OpenFEMA dataset is the broader Non-Disaster and Assistance to Firefighter Grants record, which carries Port Security, Homeland Security, EMPG, and other preparedness grants alongside the firefighter programs. Any firefighter-specific analysis must filter on program_name first; failing to do so silently folds unrelated grant streams into the totals.

The award amount is the federal share, not the project cost. Because the cost share scales with community size, the same federal dollar represents a different fraction of total project cost depending on who received it—a larger fraction for a small rural department, a smaller fraction for a big-city utility. Treating the award_amount as the full value of the funded work, or comparing federal totals across recipients of very different sizes without accounting for the match, misreads what the number means. The federal share measures FEMA's contribution, not the scale of the improvement on the ground.

An award is an input, not an outcome. The dataset records that a department was given money to buy gear, hire staff, or run a training program. It does not record whether the equipment was delivered and used, whether the staffing was retained after the grant ended, or—most importantly—whether responder safety or response performance actually improved. The competitive, application-driven design also means the data reflects who applied and applied well as much as who needed the money most: a small volunteer department with no grant-writer may be invisible in the record not because its need is low but because it never filed. Reading the absence of awards as evidence of low need—or the presence of awards as evidence of effectiveness—over-reads what the dataset can bear.

Held with these caveats in mind, the fema_afg_grants table is a uniquely valuable resource: a recipient-resolved, dollar-stamped, program-coded record of how the federal government invests in the everyday readiness of the fire and emergency services—the turnout gear, the breathing apparatus, the apparatus, the training, and the firefighters themselves—the preparedness half of a FEMA whose disaster half only takes over once the prevention and the readiness have run out.