Skip to main content

House Fire Frequency in Japan

Japan's house fire frequency runs near 100 fires per day nationwide, and homes produce a share of fire deaths far larger than their share of fire cases.1 The pattern matters more than the total: cooking stoves lead building fires, tobacco leads residential deaths, and older residents face most of the fatal risk.12

Confirm current details with official sources

Procedures, fees, and requirements can change. Confirm current details at the Fire and Disaster Management Agency (FDMA, 消防庁) fire-statistics pages.

Overview​

Japan recorded 37,141 fires in calendar 2024, down 4.0 percent from the prior year (as of 2025-11-25; confirm current figures with FDMA).1 Deaths totaled 1,451, and residential-fire deaths excluding arson-suicide victims stood at 1,030 (as of 2025-11-25; confirm current figures with FDMA).1 The daily rate sits near 101 fires per day, roughly one fire every 14 minutes.1

Residential cases form only about 3 in 10 of all fires, yet residential deaths form about 7 in 10 of all fire deaths.3 That asymmetry is the core of this article. Frequency describes the country; the death share describes where protection pays off.

How Often Fires Happen​

The national totals​

The 10-year series moves between roughly 34,000 and 39,000 fires per year with no sustained collapse, so year-to-year drops deserve caution rather than celebration.1 In 2021 the total was 35,222 fires, of which 76.0 percent were accidental fires (失火, shikka, "accidental fire").4

Building fires numbered 20,908 in the 2024 preliminary cut (as of 2025-05-30; confirm current figures with FDMA).1 Homes dominate that category: residences were the most common fire-origin building use at 11,411 cases, 56.6 percent of building fires.5

Homes carry the deaths​

In 2024, residential fires numbered 11,839 against 37,141 total cases, while residential deaths numbered 1,109 against 1,451 total deaths (as of 2025-11-25; confirm current figures with FDMA).3 The preliminary cut tells the same story inside building fires: residential deaths were 87.5 percent of building-fire deaths while residential cases were only 53.7 percent of building-fire cases (as of 2025-05-30; confirm current figures with FDMA).1

The 2021 white-paper cut found building-fire deaths at 1,165, with 90.8 percent occurring in residences.6 Readers comparing editions should note that each year's exclusion rules for arson-suicide victims match within that edition.

What Starts Residential Fires​

Cooking and stovetop fires​

The こんろ (konro, "cooking stove or burner") tops building-fire causes at 2,637 cases, 12.6 percent, in the 2024 preliminary cut (as of 2025-05-30; confirm current figures with FDMA).1 Nationally across all fire types, konro caused 2,718 fires in 2024, ranking third behind tobacco and open burning (as of 2025-11-25; confirm current figures with FDMA).1

The mechanism is mundane. In 2021, gas konro caused 2,247 of 2,678 konro fires, and leaving the burner on (消し忘れ, keshi-wasure, "leaving the burner on") caused 1,154, the largest single sub-pattern.4 Unattended cooking, not equipment failure, drives the category.

Tobacco, heating, and electrical fires​

Tobacco (たばこ, tabako, "tobacco") led all national causes in 2024 at 3,058 fires, 8.2 percent (as of 2025-11-25; confirm current figures with FDMA).1 In 2021 the tobacco total was 3,042, with discarding burning material in unsuitable places causing 1,921 of those.4 Among 2024 residential-fire deaths, tobacco led identified causes at 147 deaths and heating stoves followed at 111 (as of 2025-11-25; confirm current figures with FDMA).1

Electrical devices caused 2,577 fires nationally in 2024 (as of 2025-11-25; confirm current figures with FDMA).1 Wiring and wiring-device deaths in residences totaled 115 in the same year, a reminder that the wall socket deserves the same respect as the stove (as of 2025-11-25; confirm current figures with FDMA).1

Unidentified causes dominate the death table

Unknown or under-investigation causes covered 44.9 percent of 2024 residential deaths, so ranked causes describe the identified subset only.1 Treat any single-cause share as a floor, not a ceiling.

Arson and the declining trend​

Arson (放火, hōka, "arson") caused 2,377 fires in 2024, with suspected arson adding 1,527 for a combined 3,904, or 10.5 percent (as of 2025-11-25; confirm current figures with FDMA).1 The long series shows arson declining across recent editions, down 4.0 percent year on year on the combined measure in 2021.4

Tokyo data links the broader improvement to prevention stock: alarm-inclusive installation rates hold near 90 percent since 2015 while per-capita residential fire counts eased into the low teens (as of the Tokyo report covering through 2023; confirm current figures with the Tokyo Fire Department).7 Alarms, code updates, and campaigns share the credit; no single measure explains the curve alone.

Who Is Most at Risk​

Older residents and late-night fires​

Residents aged 65 and older suffered 779 of the 1,030 residential-fire deaths in 2024, a 75.6 percent share (as of 2025-11-25; confirm current figures with FDMA).1 The 2021 share was nearly identical at 74.1 percent.6 Fire death risk rises steeply with age, with the oldest brackets running several multiples of the all-age average.2

Deaths concentrate in the December-to-March heating months and in the midnight-to-6am sleeping band.62 Delayed escape (逃げ遅れ, nige-okure, "delayed escape") caused 46.0 percent of deaths in 2021, with late discovery and smoke-blocked exits the largest sub-patterns.6 A working alarm addresses exactly that failure mode.

Good to know​

The statistics describe frequency, not your building's risk​

National totals average fireproof mansions together with dense wooden stock in older neighborhoods. Your building's age, structure, and alarm status move personal risk far from the mean, so read the frequency as background and the cause ranking as your checklist.3

Deaths cluster where alarms are missing or expired​

FDMA analysis of 2020-to-2023 accidental residential fires found alarms roughly halved deaths and damage versus unalarmed cases (analysis window 2020-2023; confirm current figures with FDMA).3 The death table's nighttime weight is consistent with that finding: sleeping residents need a sensor awake for them.2

Compare causes across years, not across countries​

FDMA cause categories and foreign datasets use different definitions and collection methods, so cross-country rankings mislead more than they inform. Track one dataset's series over time instead; the domestic trend is the actionable signal.14

See also​

References​

Footnotes​

  1. 消防庁. 「令和6年（1～12月）における火災の状況（確定値）」. FDMA press release, 2025-11-25. https://www.fdma.go.jp/pressrelease/houdou/items/20251125boujyo.pdf ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7 ↩8 ↩9 ↩10 ↩11 ↩12 ↩13 ↩14 ↩15 ↩16 ↩17 ↩18

  2. 消防庁. 「2．火災による死者の状況」令和6年版消防白書. https://www.fdma.go.jp/publication/hakusho/r6/chapter1/section1/para1/67986.html ↩ ↩2 ↩3 ↩4

  3. 消防庁予防課. 「住宅防火関係」. https://www.fdma.go.jp/relocation/html/life/juukei.html ↩ ↩2 ↩3 ↩4

  4. 消防庁. 「4．出火原因」令和4年版消防白書. https://www.fdma.go.jp/publication/hakusho/r4/chapter1/section1/para1/10105.html ↩ ↩2 ↩3 ↩4 ↩5

  5. 消防庁. 「5．火災種別ごとの状況」令和5年版消防白書. https://www.fdma.go.jp/publication/hakusho/r5/chapter1/section1/para1/66824.html ↩

  6. 消防庁. 「2．火災による死者の状況」令和4年版消防白書. https://www.fdma.go.jp/publication/hakusho/r4/chapter1/section1/para1/10103.html ↩ ↩2 ↩3 ↩4

  7. 東京消防庁. 「住宅火災・放火火災の実態」. https://www.tfd.metro.tokyo.lg.jp/content/000044665.pdf ↩