A G20 country enacts a 32-hour working week

Where things stand (16 Oct 2026)

None of the 19 G20 countries has a statutory standard week of 32 hours or less, and none has a bill that both sets one and has a realistic chance of passing. The lowest existing standard is France's 35 hours (Code du travail L3121-27). Recent and pending cuts all stop well above 32:

• Mexico: a constitutional amendment (DOF 3 Mar 2026) and a Federal Labour Law change (DOF 1 May 2026) take the week from 48 to 40 hours by 2030. Sources: El Economista, 3 Mar 2026; Deloitte Flash Legal 09/2026.
• Brazil: the Senate passed PEC 221/2019 75–1 on 14 Oct 2026, cutting the week from 44 to 42 and then 40 hours over 14 months. It is awaiting promulgation, and 40 is the floor. Agência Brasil: https://agenciabrasil.ebc.com.br/politica/noticia/2026-10/fim-da-6x1-saiba-quando-entra-em-vigor-nova-jornada-de-trabalho
• South Korea: the most ambitious bill (Democratic Party, Kang Hoon-sik) cuts the week from 40 to 36 hours (Maeil Labor News, 4 Mar 2026, labortoday.co.kr idxno=233016). The tripartite roadmap put any cut to statutory hours on the "future review" list (worklaw.co.kr, 30 Dec 2025). In September 2026 the labour minister said the government is subsidising cuts in hours actually worked rather than cutting statutory hours (mk.co.kr/news/politics/12161385). The party talks of a four-day week as a long-term goal, but its legislative step is 36.
• United States: Sanders and Takano reintroduced the Thirty-Two Hour Workweek Act (overtime threshold falling from 40 to 32 over four years) in September 2026. Republicans control both chambers (USA Today, 11 Sep 2026).
• Italy: the Chamber rejected the opposition's 32-hour bill on 3 Mar 2026. It only offered incentives in any case (Il Sole 24 Ore).
• Australia: the main union body asked the employment standards review to cut 38 hours to 35 (Guardian, 10 Mar 2026).
• Moving the other way or standing still: Japan is relaxing its overtime caps (Seoul Economic Daily, 20 Aug 2026). Russia's Duma says it has no plans to shorten the week (Amic.ru, Oct 2026). Indonesia's new labour law keeps 40 hours (Oct 2026).
• France: Le Pen leads first-round polls at 35–37% (Toluna Harris for RTL, 12 Oct 2026). La France Insoumise's programme restores an effective 35 hours, sets 32 only for arduous or night work, and leaves wider adoption to collective bargaining (Poligraph, 27 Aug 2026). The Greens likewise limit 32 hours to arduous jobs (parlons-politique.fr, 13 Jul 2026). Both would be sector-specific, which does not qualify.

Reference class

I counted G20 cuts to statutory standard hours since 1980 from known legal history: France 39→35, Japan 48→40, Korea 48→44→40, China 48→40, Brazil 48→44→40, Mexico 48→40, Italy 48→40, and others. That is about 15 cuts, none below 35. No country anywhere has a general statutory 32-hour week. Even modest cuts fail: Spain's 37.5-hour bill (2025) and Italy's 2026 bill both went down. Proposed "next steps" (36, 35, 40) all sit above the threshold.

Paths to YES (rough contributions by end-2031)

• France, about 0.35%: a left president and majority in 2027 that goes beyond its own programme.
• United States, about 0.35%: a Democratic sweep in 2028, removal of the filibuster, and politics driven by an AI jobs shock.
• South Korea, about 0.3%: a law with a binding phase-in to 32 hours. Such a law would count even if it took effect after 2031.
• Others, about 0.7% combined: Australia, Brazil, the UK and the remaining G20 members.
• Unknown political shifts, about 0.3%.

Total: about 2%.

Shape of the forecast

I built quarterly hazards in Python and took the cumulative product:

• 0.01% for Q4 2026 (nothing is pending);
• 0.03–0.04% per quarter in 2027, when a new French government is possible;
• 0.06–0.08% in 2028;
• about 0.1% in 2029, with a new US Congress and a new Korean National Assembly;
• 0.12–0.14% in 2030;
• 0.15–0.17% in 2031.

That gives about 0.15% by end-2027, 0.43% by end-2028, 0.85% by end-2029, 1.37% by end-2030 and 2.0% by end-2031.

Strongest case against my estimate

An AI-driven jobs shock could be faster and more politically explosive than I assume. Sanders already frames the 32-hour week as the AI dividend. A binding phase-in qualifies on enactment, and 19 countries are in play. Against that, every live proposal stops at 35–40 hours, and law-making cycles are long. That keeps my estimate low, but not at zero.

Reading of the criteria

Laws setting 35 or 36 hours, sector-specific rules, public-sector rules, and Chinese State Council regulations do not count. Only an enacted national statute published in an official gazette does.

Question. By the last day of each quarter, has any G20 country enacted a statutory standard full-time working week of 32 hours or fewer?

Definitions.

• G20 country: one of the 19 sovereign members listed by the G20 at opening: Argentina, Australia, Brazil, Canada, China, France, Germany, India, Indonesia, Italy, Japan, Republic of Korea, Mexico, Russia, Saudi Arabia, South Africa, Türkiye, the United Kingdom, or the United States. Membership of the European Union or African Union alone does not qualify.
• Statutory standard full-time working week: a national law setting the generally applicable ordinary weekly working hours for adult private-sector employees at no more than 32 hours, or setting the general weekly threshold above which overtime compensation or statutory compensatory rest is required at no more than 32 hours. Ordinary exemptions for particular occupations, industries, or employee classifications do not disqualify a general standard. A rule confined to a named industry, occupation, public-sector workforce, or subnational jurisdiction does not qualify.
• Hours: ordinary working hours, excluding overtime. A statutory averaging arrangement qualifies if it explicitly sets average ordinary weekly hours at 32 or fewer over its reference period; annual leave, holidays, sickness, or other absences cannot be deducted to turn a higher weekly standard into a qualifying one. A four-day week qualifies only if its ordinary weekly hours meet the numerical threshold.
• Enacted: all constitutionally required steps for the law to become law have been completed, including assent or promulgation where required. A binding, automatic phase-in to 32 hours or fewer qualifies upon enactment, even if its effective date falls after the forecast horizon; a target requiring further discretionary legislation or approval does not.

Resolves YES if a qualifying official publication establishes that a G20 country has enacted such a national law. A maximum weekly limit of 32 hours or fewer also qualifies if it applies generally to adult private-sector employment. The law need not prohibit overtime, mandate four working days, preserve weekly pay, or mention AI or automation.

Resolves NO / Exclusions. No qualifying publication by the horizon; proposals, bills passed by only part of the legislature, manifestos, recommendations, nonbinding targets, voluntary or opt-in trials, collective agreements, employer policies, and public-sector-only or sector-specific reductions do not qualify. Neither a statistical average of hours actually worked nor a definition of full-time status solely for benefit or insurance eligibility qualifies. Regulations or executive orders without a qualifying enacted statutory standard do not count. Merely allowing individuals to request shorter hours does not establish a general standard.

Resolving source. The enacting country's official gazette or official legislative publication of the enacted law, together with official enactment-status records where necessary, is controlling. For example, France publishes the Code du travail through Légifrance (the existing 35-hour standard is Article L3121-27). The ILO's TRAVAIL Database of Conditions of Work and Employment Laws, working-time category, and ILO NATLEX may locate relevant legislation but cannot substitute for the official enactment record. This is a narrower source rule than Public confirmation (A2). Apply Resolution date (A3) and the Resolution window: use the first qualifying official publication on or after opening establishing completed enactment, not the law's future commencement date. A republication of a law enacted before opening does not count.