On this page8 sections
Housing is the biggest thing households buy in every country that publishes a proper spending survey. In the United States it takes 33.4% of what the average "consumer unit" spends. In Canada it takes 32.1% of household consumption. In the United Kingdom the headline figure is 18% — and that gap is almost entirely an accounting artifact, not a fact about British houses.
That is the finding of this report, and it is a slightly awkward one for a page that promises a four-country comparison. The four national statistics agencies are not measuring the same thing. They count different units, over different periods, with different rules about where mortgage interest, property taxes and pension contributions belong. Stack their headline percentages in a bar chart and you produce a league table that means nothing.
So this report does two things. First it shows each country's own published breakdown, in full, exactly as its statistics agency reports it — the US 14-category table, the UK 13-category table, and Canada's headline shares. Then it shows the one measure that was actually built for cross-country comparison, and points out that it puts the four countries in a different order.
Four countries, four incompatible definitions
What each source actually measures. Read this before reading any number below it.
- Spending unit"Consumer unit", not household
- Period2024 calendar year
- Breakdown14 major components
- IncludesPersonal insurance and pensions, $9,797
- Spending unitHousehold
- PeriodApril 2024 to March 2025
- Breakdown12 groups plus "other expenditure items"
- Excludes from housingMortgage interest, Council Tax
- Spending unitHousehold
- Period2023 calendar year, biennial survey
- BreakdownConsumption of goods and services
- ExcludesIncome taxes, pension contributions, gifts
- Spending unitn/a
- Periodn/a
- BreakdownEnters via price and national-accounts measures
- What existsMonthly spending indicator, CPI, living cost indexes
Source: US Bureau of Labor Statistics, Consumer Expenditures — 2024, USDL-25-1586; Office for National Statistics, Family spending in the UK, April 2024 to March 2025; Statistics Canada, Survey of Household Spending, 2023
Methodology, and what is not comparable
Three national expenditure surveys carry the weight of the country sections. Australia has no equivalent, and two national-accounts sources carry the comparison at the end.
The United States figures come from the Bureau of Labor Statistics Consumer Expenditure Survey, release USDL-25-1586, published 19 December 2025 and covering calendar year 2024. Its unit is the "consumer unit", which is close to a household but not identical, and there were 135,760 thousand of them in 2024.
The United Kingdom figures come from the ONS Family spending bulletin for April 2024 to March 2025, published 11 June 2026, and from Table A1 of the workbook behind it. The unit is the household, the period is a financial year, and the sample is 5,000 responding households — small enough that lower-level categories carry real uncertainty.
The Canadian figures come from Statistics Canada's Survey of Household Spending, 2023, published 21 May 2025. It runs every two years, so the most recent reference year available in August 2026 is 2023 — a full year behind the US calendar-year figures, and roughly fifteen months behind the midpoint of the UK's April 2024 to March 2025 window.
Australia is the honest gap. Our verified source library holds no current category-level Australian household expenditure survey. What exists for Australia is the ABS Monthly Household Spending Indicator, the CPI, the Selected Living Cost Indexes and the national accounts — all useful, none of them a "where the money goes" table. So Australia appears in this report through measures that genuinely do cover it, and is absent from the ones that do not.
The cross-country section does not use any of those surveys. It uses the OECD's National Accounts at a Glance, read from the OECD's SDMX API in August 2026, and Eurostat's Statistics Explained article on household consumption by purpose, extracted November 2025. Both are national-accounts measures rather than surveys, which is what makes them comparable — and also what makes them a different thing from the tables above. Two limitations are worth stating up front: the OECD housing-share measure is populated for only 15 countries in 2023, so it is not a world ranking, and the OECD figures here are an API extraction rather than a dated publication, so they carry the August 2026 vintage rather than a release date.
Four things stop the national surveys from being stacked:
- Different units. A US "consumer unit" is not a UK or Canadian household.
- Different periods. 2024, the year to March 2025, and 2023 respectively.
- Different denominators. The US total of $78,535 includes $9,797 of personal insurance and pensions, of which $6,684 is compulsory Social Security deductions. The Canadian C$76,750 explicitly excludes income taxes, pension contributions and gifts. The UK £676.60 has no equivalent line at all.
- Different homes for housing costs. The ONS puts mortgage interest and Council Tax in "other expenditure items", not housing. Statistics Canada's shelter category covers mortgage or rent, property taxes, insurance, utilities and maintenance together.
Point four alone is worth nine percentage points, as the UK section below shows.
There is a fifth problem that is easy to forget: these are sample surveys, not censuses. The UK bulletin rests on 5,000 responding households, which is enough for the headline but thin once you cut it by region or by income decile — individual decile figures come from samples of roughly 440 to 540 households. The BLS marks which of its year-on-year changes are statistically significant and which are not, and most of the 2024 movements were not. Treat any single small category, in any of these tables, as indicative rather than precise.
The United States: a third of the budget is housing
The average US consumer unit spent $78,535 in 2024, up 1.8% on 2023. That is about $6,545 a month. Average income before taxes was $104,207, so the average consumer unit spends roughly 75 cents of every pre-tax dollar it earns — though that $104,207 is a mean, sitting well above the Census Bureau's median household income of $83,730.
How the average US household budget divides, 2024
Percent of total annual expenditures of $78,535, all 14 major components
Source: US Bureau of Labor Statistics, Consumer Expenditure Survey, Table B, 2024
In dollars: housing $26,266, transportation $13,318, food $10,169. Housing was the only major component with a statistically significant increase in 2024, at 3.3%. Transportation rose 1.1%, which was not significant — although one line inside it moved hard. Vehicle insurance rose 12.3% to $1,993 after an 11.5% rise the year before, a 25% increase across two years, and one of the steepest rises anywhere in the table.
The food line splits in a way that is worth holding onto for later. Of the $10,169 US households spent on food in 2024, $6,224 went on groceries and $3,945 on eating out — about 39% of the food budget consumed outside the home. Groceries rose 2.8% over the year while food away from home was essentially flat at 0.3%, so the eating-out share is drifting down slightly rather than up.
The $9,797 personal insurance and pensions line is the one that trips people up when they try to compare countries. It looks like saving. Most of it is not: $6,684 is compulsory Social Security deductions, and only $1,991 is contributions to retirement plans. If you strip that line out entirely, the remaining $68,738 of spending puts housing at 38.2% rather than 33.4%. Same household, same survey, five percentage points of difference from one classification decision.
The average also hides an enormous spread. Spending runs from $35,046 a year for the lowest income fifth to $150,342 for the highest.
The average US household does not exist
Average annual expenditures by income quintile, 2024
Source: US Bureau of Labor Statistics, Consumer Expenditure Survey, Table C, 2024
The United Kingdom: the 18% that is really 27%
UK households spent £676.60 a week in the year to March 2025, a nominal rise of £53.30 (9%) on the year before. That is roughly £2,932 a month. Once inflation is taken out the ONS puts the real-terms rise at £35.10, or 5%.
Average UK weekly household spending, FYE 2025
All 13 published categories, summing exactly to £676.60
Source: Office for National Statistics, Family spending workbook 1, Table A1, FYE 2025
Housing, fuel and power is the largest single item at £118.40 a week, 18% of the total. Inside it: net rent £57.70, electricity, gas and other fuels £35.90, water and miscellaneous dwelling services £12.90, maintenance and repair £11.90.
That is the survey's view of bills that are also published directly, one price series at a time. Adding up the six that are set for you — energy, water, council tax, broadband, mobile and the TV licence — comes to roughly £455 a month before a penny of rent or mortgage, which is the floor a UK budget starts from.
Read that 18% next to the American 33.4% and you would conclude British housing is half as expensive relative to income. It is not. The ONS classifies mortgage interest payments and Council Tax under "other expenditure items", and those come to £65.30 a week — £30.40 of mortgage interest and £32.30 of Council Tax or domestic rates. Move them where a Canadian or American reader would expect to find them and UK housing becomes £183.70 a week, or:
£183.70 ÷ £676.60 = 27.1%
That is about £796 a month. The 18% headline and the 27.1% recalculation describe the same households in the same year; only the classification changed. Note too that capital mortgage repayments — £37.50 a week — are treated as saving rather than expenditure and are excluded from the total altogether, so even 27.1% is not the full cash outflow.
Two more caveats before anyone quotes 27.1% as a UK housing burden. It is averaged across all households including outright owners who pay neither rent nor mortgage, so it understates the position of any individual mortgagor or renter. And spending is not evenly spread: the lowest-income tenth of households spent £291.30 a week against £1,345.60 for the highest tenth, with the richest fifth now spending 2.7 times as much as the poorest fifth. Food shows the pattern most clearly — the top tenth spends £104.20 a week against £41.70 for the bottom tenth, only 2.5 times as much despite far higher income, because food is the classic cost you cannot scale down. If you are building your own category list, that asymmetry is the reason a percentage-based budget behaves differently at different incomes.
Geography does the same job. On a three-year average covering FYE 2023 to FYE 2025 — a different basis from the single-year headline, so the two should not be mixed — households in the South East spent £724.20 a week against £505.50 in the North East. London is the instructive case: its housing, fuel and power spend of £182.20 a week is £69.70 above the £112.50 UK average on that basis, while its total of £695.90 is £73.30 above the £622.60 UK average. Almost the entire London premium is housing; on everything else combined Londoners spend roughly what the country does, with lower transport and recreation spending offsetting the rest. The budget does not get bigger by much; it gets rearranged.
Canada: shelter takes a third, and renters are catching up
Canadian households spent an average of C$76,750 on goods and services in 2023, 14.3% more than in 2021 — remembering that the survey is biennial, so that is a two-year change, not an annual one. Roughly C$6,396 a month.
Canadian household spending on goods and services, 2023
C$76,750 per household
- ShelterC$24,67132%
- TransportationC$12,09016%
- FoodC$12,04616%
- All other goods and servicesC$27,94336%
Show the data
| Category | Value | Share |
|---|---|---|
| Shelter | C$24,671 | 32% |
| Transportation | C$12,090 | 16% |
| Food | C$12,046 | 16% |
| All other goods and services | C$27,943 | 36% |
Source: Statistics Canada, Survey of Household Spending, 2023
Shelter at C$24,671 is 32.1% of consumption, about C$2,056 a month. Transportation is 15.8% and food 15.7% — the two are within C$44 of each other, which is unusual: in the US, transportation outspends food by a third.
The Canadian survey also splits shelter by tenure, and the split matters more than the average. Homeowners spent C$27,831 on shelter in 2023, up 17.4% in two years. Renters spent C$18,333, up 20.2%. Renters pay less in absolute terms and their costs are rising faster. Recreation, at C$5,231, rose 23.9% over the same two years — but because the survey is biennial the comparison base is 2021, so most of that is a post-pandemic rebound in recreational services rather than an ongoing annual trend.
Canadian households are also carrying the debt that goes with those shelter costs: 14.75% of disposable income went on required debt principal and interest in the first quarter of 2026, with credit market debt at 179.6% of disposable income. On the OECD's comparable measure the same pattern shows up as Canadian households owing 181.1% of net disposable income against 98.9% in the US — a ranking driven by mortgage markets rather than by card habits. If that is your situation, the debt-to-income calculator will tell you where you sit against it.
Australia: what we can say, and what we cannot
We cannot show you an Australian version of the three charts above, because our verified source library does not contain a current category-level Australian household expenditure survey. Rather than borrow a number from somewhere less reliable, here is what the Australian Bureau of Statistics does publish.
Total household spending was 6.0% higher in June 2026 than a year earlier on a current-price basis, and rose 0.8% over the month — but that is a growth rate, not a breakdown, and it includes inflation as well as extra volume. The saving ratio was 6.2% in the March quarter of 2026, down from 7.0%.
What the ABS does show clearly is which costs are moving.
What is pushing Australian household costs up
Annual change in the CPI and selected components, 12 months to June 2026
Source: Australian Bureau of Statistics, Consumer Price Index, Australia, June 2026
Housing was the fastest-rising major group at 6.8% and the largest single contributor to annual inflation. Wages grew 3.3% in the year to the March 2026 quarter, below headline inflation at the time. Living costs rose between 3.7% for employee households and 4.7% for age pensioner households — a spread worth noting, because unlike the CPI the Living Cost Indexes include mortgage interest charges.
What can actually be compared
There is one measure in this report built for cross-country use: the OECD's national accounts figure for what households spend on housing, water, electricity, gas and other fuels as a share of net adjusted disposable income. Same definition, same year, same denominator.
Housing as a share of net adjusted disposable income, 2023
The one measure in this report that is built for comparison — among the 15 countries the OECD reports
Source: OECD, National Accounts at a Glance, measure P31S1M_CP04, 2023
This is the interesting part. On the national surveys, US housing (33.4%) looks heavier than UK housing (18%). On the OECD's comparable measure the order reverses: the UK is on 21.5% and the US on 15.7%, with Canada at 19.6% and Australia at 18.0% in between. Both orderings are correct within their own definitions. Neither is "the answer".
The denominators explain most of it. The national surveys divide housing by total spending; the OECD divides it by net adjusted disposable income, which adds the value of state-funded health and education. That change alone moves the US, where households carry $6,197 a year of healthcare inside their own budget, in a different direction from the UK.
A third framing is worth putting next to those two, if only to show how much the denominator does. Eurostat measures housing, water, electricity, gas and other fuels as a share of total household final consumption expenditure across the EU, and gets 23.6% for 2024 — with food at 13.2% and transport at 12.7%. That is a third definition again: consumption rather than income, national accounts rather than survey. It lands between the survey shares and the OECD income shares, which is roughly what you would expect and is the clearest evidence that the spread between "18%" and "33.4%" is measurement, not housing markets.
Income is comparable too, and worth having alongside.
How much is coming in
Cash income, PPP-converted US dollars per person — comparable across countries
Household gross disposable income per person, 2024
- USUnited States$66,155per person, 2024OECD National Accounts at a Glance
- AUAustralia$44,200per person, 2024OECD National Accounts at a Glance
- UKUnited Kingdom$39,412per person, 2024OECD National Accounts at a Glance
- CACanada$36,487per person, 2024OECD National Accounts at a Glance
And saving, on the OECD's net measure, runs 5.7% in the United States, 5.1% in Canada, 4.7% in the United Kingdom and 6.1% in Australia — all far below Sweden's 16.3%. That is a net measure after depreciation and is not the same thing as the gross saving ratios national agencies publish, so do not compare it with the UK's 8.9% or Australia's 6.2% headline ratios.
What this means for your budget
Four things follow from all of this, and none of them require you to pick a winning country.
Assume housing plus transport plus food is roughly 60% of your spending, then check. In the US those three are 63.3% of the budget. In Canada 63.6%. In the UK they come to 42.6% on the published categories, or 52.3% once mortgage interest and Council Tax are moved back into housing. Each of those is a share of that country's own total, which is defined differently in each case — so treat them as a rule of thumb about your own budget, not as a ranking. The rest of your budget is the part you can actually move, and it is smaller than most budgeting advice implies. That is the arithmetic behind why the 50/30/20 rule starts failing once housing alone passes a third of take-home pay.
Check where your own budget puts mortgage interest and property tax. This report's biggest single finding is that a nine-point difference in the UK housing share came from nothing but a classification choice. The same thing happens in your own spreadsheet. If Council Tax sits under "bills" and the mortgage under "housing", your housing number is wrong and you will make decisions on it. The step-by-step budget guide walks through building the categories once, properly.
Compare yourself to your own income band, not the national average. US spending runs from $35,046 to $150,342 across income quintiles; UK weekly spending runs from £291.30 to £1,345.60 across deciles. A national average is the midpoint of an enormous range, and for most households it is the wrong benchmark. Put your own numbers into the budget calculator and compare against the band you are actually in.
Watch the lines that moved, not the lines that are big. Housing is the biggest number in every table here and there is usually not much you can do about it this month. The movers are more actionable: US vehicle insurance up 25% in two years, Canadian renters' shelter costs up 20.2% in two years, Australian insurance premiums up 4.9% in a year. Those are renewal-letter problems, which means they respond to an afternoon of phone calls in a way that rent does not. The same logic applies to energy bills and to the subscriptions you stopped noticing.
One last number for perspective on why any of this matters. In the UK, the FCA's Financial Lives survey found in May 2024 that 42% of adults could not cover their living costs for three months if they lost their main source of household income. In the US, the Federal Reserve's 2025 survey found 55% of adults had three months of expenses set aside in a rainy day fund, and separately that 30% could not cover three months by any means at all, including borrowing or selling assets.
Do not read those as a league table either. They come from different years, different bodies and different questions — "could you cover three months if income stopped" is not the same question as "have you set money aside" — so 42% and 30% are the nearer pair, and neither is a clean UK-versus-US number. The full set of resilience questions, and why a $400 threshold and a three-month threshold produce answers that cannot be stacked, is worth reading before quoting any of them. What they agree on is the size of the exposed group. Whichever country's spending table you read, the gap between what goes out and what is held back is the thing that decides how a bad month goes, which is what the emergency fund guide is for.
Sources and reuse
Every figure on this page is drawn from an official statistical release, linked at the point of use. If you want to cite this report, cite the underlying agency for the raw numbers and this page for the comparison and the recalculations. The derived figures are ours, not the agencies': the UK 18% to 27.1% housing adjustment, the 38.2% US housing share after stripping out personal insurance and pensions, the 42.6% and 52.3% UK three-category shares, the C$27,943 "all other" residual in the Canadian donut, the London-versus-UK differences, and every monthly conversion of a weekly or annual figure. All of them are plain arithmetic on the published numbers, and all of them are shown so you can check them.
If a number here matters to a decision you are making, follow the link. All four agencies revise, and the Canadian survey in particular is two years behind the others.
If you want to see your own version of the tables above rather than the national average, iBudget sorts your spending into categories and shows you the shares. It is free to start, and you can pick the categories yourself — including deciding where the mortgage interest goes.
About iBudget
iBudget helps couples and families take control of their finances with simple, collaborative budgeting tools. Track spending, set goals, and build wealth together.
Start Your Budget
