Skip to content

Weekday Japan business intelligence for finance professionals.

Join the list
Tokyo Brief東 京 ブ リ ー フJapan's day, wrapped and delivered by morning.

Policy Watch

Draft assessment finds buprofezin unlikely to harm honeybee colonies, and Japan wants comment by 13 November

Japan's farm ministry has opened a voluntary comment period on a draft finding that the insecticide buprofezin, used on rice, fruit and tea, does not threaten honeybee colony survival when applied as registered.

By Tokyo Brief DeskOct 16, 20263 min read
A honeybee on a citrus blossom beside a rice paddy, with a spray nozzle and a ledger of dose ratios in the foreground.

Japan's agriculture ministry has put a draft honeybee assessment of the insecticide buprofezin out for public comment, and the draft reaches a clean result: used as applied for, the chemical is not expected to impair the survival of honeybee colonies. The Ministry of Agriculture, Forestry and Fisheries (MAFF) is accepting comments from 15 October to 13 November 2026, in a voluntary consultation rather than one required under the Administrative Procedure Act.

What the draft concludes

The draft assessment was approved on 10 September 2026 at the 22nd meeting of the honeybee impact assessment subcommittee of the Agricultural Materials Council's pesticide panel. Under Article 39(1) of the Agricultural Chemicals Regulation Law, the farm minister must hear the council before registering or changing a pesticide registration.

The assessment covers 14 formulations. It sets four toxicity reference values: 200 µg of active ingredient per bee for adult contact, 160 µg per bee for a single oral dose, 170 µg per bee per day for repeated oral dosing, and 100 µg per bee for larvae. Each comes from a laboratory result reported only as a lower bound (a median lethal dose reported as "greater than" the highest dose tested).

Honeybee toxicity reference values in the draft
Each value is drawn from a laboratory result reported as a lower bound. Units are micrograms of active ingredient (ai) per bee.
Life stageTestReference value
AdultSingle contact (48 h LD50)200 µg ai/bee
AdultSingle oral (48 h LD50)160 µg ai/bee
AdultRepeated oral (10 d LDD50)170 µg ai/bee/day
LarvaOral (72 h LD50)100 µg ai/bee

How the screening worked

The first-stage test compares estimated bee exposure with the half-lethal dose. If that ratio does not exceed 0.4, the draft treats bee mortality as not exceeding 10%, the level it counts as natural mortality. Every use assessed for foliar spray and soil treatment came in at or below 0.4, so the draft says a second-stage assessment is unnecessary. Ratios in the annex tables run as high as 0.21.

Some uses were not estimated at all because bees are presumed not to be exposed. These include crops harvested before flowering or that do not flower, such as spring onion, tobacco and tea. The exposed uses are foliar sprays on rice, fruit trees and other crops, and soil treatment on rice and sudachi citrus. The draft's conclusion holds only for the registered methods of use.

Evidence base

The tests are mostly unpublished laboratory studies of the technical-grade active ingredient, dated between 1996 and 2019 in the draft's source list. A literature search of 626 papers found none using the European honeybee that could inform the toxicity values. The call for published literature (20 March to 18 April 2026) drew no submissions. The draft says buprofezin was registered in the US, Australia, the EU and South Korea as of March 2025.

How to comment

Comments must be in Japanese and go through the e-Gov form or by post to MAFF's pesticide office. They must arrive by 13 November, and the online window closes at 23:59 that day. MAFF will not take telephone comments or reply individually, and it may publish comments with personal information removed. The notice sets no date for a final decision; MAFF says it will decide after considering the comments.