Skip to main content
Back to blog
Remote work by country: what the statistics agencies actually report
remote work data international

Remote work by country: what the statistics agencies actually report

There is a set of remote-work figures that circulates widely: the UK at 40%, Australia at 37%, Canada at 30%, Brazil at 17%, India at 20–25%. They appear in round-up articles, in decks, and increasingly in AI-generated answers, each time attributed to a national statistics agency.

We checked all of them against the agencies themselves. Most do not survive.

This matters beyond pedantry. If you are deciding where to hire, or arguing for a policy, or simply trying to understand whether remote work is bigger in Britain than in Brazil, you are being handed numbers that are wrong by a factor of two — and, more fundamentally, numbers that were never comparable in the first place.

What circulates, and what the agencies publish

Country Widely quoted What the agency actually reports
UK 40% work from home at least one day a week 28% of working adults were hybrid working (ONS, Jan–Mar 2025)
Australia 37% work from home part of the week 36% usually worked from home (ABS, Aug 2025) — including those doing so full-time
Canada 30% hold remote-eligible roles 17.4% mostly worked from home (StatCan, May 2025). Separately, 38.9% hold jobs that could be done from home
Germany 24% regularly work from home 24.3% worked at least partly from home (ifo, Feb 2026) — no frequency threshold at all
France 26% telework occasionally 26% telework at any frequency. Occasional alone is 8% (DARES via INSEE, 2023)
Brazil 17% hold remote roles 7.9% worked at their place of residence (IBGE, 2024)
India 20–25% of IT employees work remotely No such figure exists. NASSCOM is a trade association and publishes no measured share

Two are roughly right. One is right by accident. Four are wrong, and two of those are wrong by more than double.

The Brazil case shows how this happens. IBGE's real 2024 figure is 7.9%. The highest remote-work share IBGE has ever published for Brazil is 13.3%, from an emergency pandemic survey in May 2020 that used a different instrument entirely. There is no year and no series in which 17% is an IBGE number. Meanwhile Brazilian research institute IPEA has published estimates of potential telework in the 20–25% range — a different concept, from a different organisation — which is a plausible route by which a number in that neighbourhood acquired an IBGE label on the way round the internet.

The France case is subtler and more instructive. The 26% is real and it is DARES's. But it is the share teleworking at any frequency, from three days a week down to a few half-days a month. Relabelling it "occasionally" quietly swaps it for a category that is actually 8%. The accompanying "35% in Paris" belongs to no DARES publication we could find; the nearest official figure is INSEE's, for regular telework, in Île-de-France — a different agency, a different definition, and a region that extends well beyond Paris.

Why a league table is impossible anyway

Suppose every figure above were quoted correctly. You still could not rank these countries, because the agencies are not measuring the same thing. There are three distinct concepts in play, and they routinely get collapsed into one.

Capability — could this job be done from home? Statistics Canada puts 38.9% of Canadian jobs in this category, based on task content: work requiring public interaction, machinery, protective equipment or physical activity is excluded. Nobody's permission is involved and nobody has to actually do it.

Entitlement — is the arrangement available? Singapore's Ministry of Manpower reports that 89.7% of the covered workforce is at an establishment offering at least one form of flexible work arrangement. That is not 89.7% of Singaporeans working from home; it means their employer offers something flexible to someone.

Practice — did they actually work from home? This is what most people think they are reading, and it produces the smallest numbers: 36% in Australia, 28% in the UK, 17.4% in Canada, 7.9% in Brazil.

For a single country, those three answers can differ by a factor of four or five. Canada makes it plain: 38.9% could, 17.4% do. Quote the first and call it remote work and you have overstated reality by more than double without saying anything false.

Even within "practice", the thresholds differ

Narrow it to actual behaviour and the definitions still refuse to line up:

Agency What counts
ONS (UK) Worked at home at least one day and travelled to work at least one day, in the past seven days
ABS (Australia) "Usually" works from home in main job — includes fully remote
StatCan (Canada) Works most of their hours at home in the reference week
ifo (Germany) Works "at least partly" from home — one day a month counts the same as five days a week
DARES (France) Teleworks at least a few half-days per month
IBGE (Brazil) Performs work at their place of residence

Canada's "most of their hours" and Germany's "at least partly" sit at opposite ends of the same spectrum. Someone working one day a week from home is counted in Germany and not in Canada. That single difference could account for most of the apparent gap between them.

The population bases diverge too. ONS covers working adults aged 16+ in Great Britain, not the UK, and includes the self-employed. ABS covers employed people 15+. Canada's practice figure covers ages 15–69 in the ten provinces. France's covers employees aged 20–62 in metropolitan France, excluding the self-employed entirely. Germany's is not a household survey at all — it asks around 9,000 firms what share of their staff work from home, and its panel barely covers agriculture, public services, finance or healthcare.

The one comparison that is fair

If you want a cross-country picture, use a single source that applies one definition everywhere. The Indeed–OECD project on remote work does this for job postings, finding that the share advertising home working rose from about 2.5% to around 11% across the countries it studies between January 2020 and January 2023.

That number is lower than any of the practice figures above, and the gap is real rather than definitional: far fewer jobs are advertised as remote than are worked remotely. We went through that gap in detail here — it is the single most useful thing to understand about the current market.

What each country's data is actually good for

  • UK. ONS's hybrid measure is the cleanest read on split-week working anywhere. Note it comes from the Opinions and Lifestyle Survey, not the Labour Force Survey, so the well-publicised LFS response-rate problems do not affect it.
  • Australia. The occupational split is the useful part: 59% of managers and professionals usually work from home against 21% of all other occupations. Worth knowing that the survey behind it is being redesigned, so the series as constituted is ending.
  • Canada. The only country here publishing capability and practice side by side, which makes it the best single illustration of the gap between them.
  • Germany. Because it is firm-reported, ifo also publishes intensity — employees spend about 17% of working time at home — which most agencies do not collect.
  • Singapore. MOM measures provision rather than practice, and its 2024 reference date of 30 June predates the Tripartite Guidelines on Flexible Work Arrangement Requests, which took effect that December. The first data reflecting those guidelines is not due until around November 2026.
  • Brazil and India. Brazil has good household data showing a genuinely low rate. India has no representative official figure at all — neither NASSCOM nor the Periodic Labour Force Survey publishes one.

Common questions

Which country has the most remote work?

The question cannot be answered from official statistics, because each agency measures a different thing over a different population. Australia's 36% and Brazil's 7.9% are not on the same scale and should not be ranked against each other.

Why is the UK figure lower than I have seen elsewhere?

Because 40% is not an ONS figure. ONS reports 28% hybrid working for January to March 2025. An older ONS statement covering late 2022 put home and hybrid working combined at 44%, which may be the origin of the higher numbers.

What does "remote-eligible" mean?

It means the job could plausibly be done from home, not that anyone does it or is allowed to. Statistics Canada puts 38.9% of Canadian jobs in that category while only 17.4% of workers actually work mostly from home.

Browse what is actually live: remote jobs in the UK · Canada · Singapore · all remote and hybrid jobs

Sources

Every figure below was checked at the publishing agency, not at a secondary source.

Back to blog