When the Office for National Statistics (ONS) reports the UK inflation rate, it is describing an average across the whole country. But households do not experience that average equally: a renter, a mortgage holder and someone who owns their home outright face very different cost pressures. The Household Costs Index (HCI) exists to show those differences. In the year to June 2026, overall UK household costs as measured by the HCI rose by 2.8%1.
The HCI is a family of measures, not a single number. The ONS publishes separate inflation rates for households at different income levels, for renters and owners, for retired and non-retired households, and for households with and without children1. It also includes costs that the headline Consumer Prices Index (CPI) leaves out, above all mortgage interest payments and Council Tax, which is why the two measures can give noticeably different answers1.
What the Household Costs Index measures
The HCI measures how the cost of living changes for households as they actually experience it, meaning the payments they make. The ONS states that the HCIs "also include changes in mortgage interest rates, stamp duty and other costs related to the purchase of a dwelling", costs that are left out of CPI altogether1. The methodology adds that the HCIs take a payments approach, incorporating owner occupiers' housing costs, interest payments, student loan repayments and the full cost of insurance premiums2.
That payments approach is the heart of the idea. CPI, by contrast, measures the average change in the price consumers pay for a basket of goods and services3, and it excludes housing costs such as mortgage interest4. The result is two different answers to the question "how fast are prices rising?": one based on what things cost, and one based on what households pay out each month, including the cost of borrowing to buy a home.
The HCIs are published for household groups rather than only in aggregate. The groups covered are income deciles, tenure types, retirement status, and households with and without children2. So a reader can see, in a single release, how costs moved for a low-income household, a private renter, a retired household or a family with children, and compare that with the all-households figure.
The HCI is one of several inflation measures published in the UK. The main headline measures, CPI and CPIH, are explained in detail elsewhere on this site, as is what inflation is and how it affects your money.
Mortgage interest, Council Tax and equal weights: how the HCI differs from CPI
Three differences do most of the work in separating the HCI from CPI: what counts as a housing cost, how Council Tax is treated, and how the weights are calculated.
First, housing costs. The HCI includes mortgage interest payments, which CPI excludes. Mortgage interest in the HCIs is calculated using the Retail Prices Index (RPI) model, which estimates the interest due on a representative stock of mortgages for an average household2. The HCI also includes stamp duty and other costs related to buying a dwelling1.
Second, Council Tax. The ONS notes that "Council Tax is not included in CPI but contributed 0.21 percentage points to the HCI annual rate" in June 20261. In the previous quarter's release the same difference was measured at 0.22 percentage points5. Owner occupiers' housing costs added a further 0.09 percentage points to the gap between the two measures in June 20261.
Third, the weights. For the HCIs, the weight of each component in a household group's "fixed basket" is based on the average household's share of expenditure, an approach known as "democratic" weights5. CPI instead uses weights that reflect total spending across the economy, which gives more influence to households that spend more. Democratic weighting can work in either direction: in the year to March 2024 it partially offset the housing effect, lowering the all-household HCI relative to CPI by 0.37 percentage points6.
The combined effect of these differences varies with the economic moment. In June 2026 the gap between the HCI and CPI annual rates was 0.2 percentage points1. In the year to March 2024, when mortgage rates were rising steeply, housing and household services contributed 1.30 percentage points more to the all-household HCI rate than to CPI6, and in June 2024 the same division contributed 1.38 percentage points more7. In December 2025 the contribution was 0.43 percentage points8.
| Difference | HCI | CPI |
|---|---|---|
| Mortgage interest | Included, via the RPI model2 | Excluded |
| Council Tax | Included, 0.21pp contribution in June 20261 | Excluded |
| Stamp duty and purchase costs | Included1 | Excluded |
| Weights | Democratic, based on the average household's share of spending5 | Based on total spending |
| Student loan repayments, insurance premiums | Included on a payments basis2 | Treated differently |
A fuller comparison of the two measures, including CPIH, which does include some housing costs, is on the site's CPIH or the Household Costs Index page.
The latest HCI rate: 2.8% for all households
Overall UK household costs, as measured by the HCI, rose by 2.8% in the year to June 20261. That is the latest figure in a clear downward path from the peak of the cost of living squeeze. In the year to March 2024 the all-households rate was 4.4%6; by June 2024 it had fallen to 2.5%7; in March 2025 it stood at 2.6%9; and in December 2025 it was 3.6%8, before easing again to 2.8% by June 20261.
The path has not been perfectly smooth, and the December 2025 figure of 3.6%8 sits above both the March 2025 and June 2026 readings. The ONS also corrected the all-households annual growth rate for May and September 2025, which had been overstated by 0.1 percentage points8, so small revisions to recent months are possible.
The HCI rate and the CPI rate move broadly together, but the HCI has tended to sit above CPI because of the housing costs it includes. In June 2026 the gap was 0.2 percentage points, unchanged from the previous comparison1. When mortgage interest was the dominant force, the gap was much wider: housing and household services alone contributed 1.38 percentage points more to the HCI than to CPI in June 20247.
For context on the headline measures and their publication schedule, see latest UK inflation figures and when they are published.
Low-income and high-income households: why their rates differ
The HCI publishes separate rates for low-income and high-income households, using the second and ninth income deciles to represent them. The ONS explains that "the second and ninth deciles are used to represent low- and high-income households rather than the first- and tenth-income deciles" because the composition of the very lowest and highest groups can be unusual and may be influenced by unrepresentative expenditures1. The same exclusion is repeated across releases8.
The gap between the two groups has swung over the past few years, and the direction has changed. In the year to October 2022, at the height of the energy price shock, low-income households faced the higher HCI rate by a margin of 2.0 percentage points6. By the year to March 2024 the position had reversed: high-income households (decile 9) faced 5.0% while low-income households (decile 2) faced 3.9%, a gap of 1.1 percentage points, itself narrowed from 1.4 percentage points in October and November 20236. In March 2025 the rates were 2.5% for low-income and 2.7% for high-income households9, and in December 2025 both groups saw costs rise by 3.7%8.
What drives the gap is the mix of spending. In the year to March 2024, mortgage interest payments increased the annual rate for high-income households by 0.95 percentage points more than for low-income households, and restaurant and hotel prices contributed 0.26 percentage points more to high-income households' rate6. Working the other way, lower electricity, gas and other fuel prices reduced the rate more for low-income households, adding 0.60 percentage points to the gap, and alcohol and tobacco contributed 0.12 percentage points more to low-income households' inflation6. In the year to June 2024, falling gas, electricity and other fuel prices explained a 1.01 percentage point difference between the two groups, while mortgage interest contributed 0.83 percentage points more for high-income households7. Low-income households were also more affected by social rent payments, which contributed 0.57 percentage points to their inflation in March 20259.
Over longer periods the differences largely wash out. Cumulative inflation over the five years to June 2026 was 32.8% for high-income households and 32.7% for low-income households1; over the five years to December 2025 the figures were 33.2% and 33.1%8; and over the five years to March 2024 costs rose 28.6% for low-income households against 28.2% for high-income households6. June 2026 was the first month since March 2025 that low-income households had a lower annual rate than high-income households1.
Retired households, families with children, renters and mortgage holders
The HCI's tenure and life-stage groups show some of the sharpest differences, because housing costs hit each group differently.
A retired household is defined as one where the combined income of retired members amounts to at least half the total gross income of the household1. A child is defined as any person aged under 16, and unmarried people aged under 18 are also classed as children1. Within its expenditure statistics, the ONS further splits retired households into state pension and other retired households10.
In the year to March 2024, retired households faced an HCI rate of 3.4% while non-retired households faced 4.8%, and households with children also faced 4.8%6. By tenure, private renter households faced 4.6%, social and other renter households 4.3%, and mortgagor and other owner-occupier households 6.6% in January 20246. In March 2025 retired households saw 2.1% against 2.8% for non-retired households9, and in the year to September 2025, mortgagor and other owners saw 4.2% while social and other renters saw 4.1%11.
The drivers are visible in the contributions. Mortgage interest payments contributed 2.0 percentage points to the annual rate for mortgagor households in the year to June 2024, and private rental costs contributed 2.2 percentage points for private renter households7. Private renters' HCI rate was higher than social and other renters that year, at 3.2%, narrowing the overall gap between the two to 0.5 percentage points7. In June 2026, mortgage interest contributed 0.14 percentage points more to inflation for non-retired households than for retired households, and non-retired households saw a greater contribution from private rentals of 0.18 percentage points1. The difference between retired and non-retired rates increased by 0.4 percentage points between March and June 20261.
Rising rents are a live issue for renter households. ONS figures reported by Which? showed 50.6% of privately rented properties in England experienced a price increase between February 2022 and February 2023, with an average increase of 9.6% among those that rose12. More recently, UK private rents rose 3.7% annually to £1,393 in July 2026, with the North East and North West recording the steepest regional increases at 5.8%13. The site's guide to private rents explains how the ONS tracks what tenants pay.
Housing costs also weigh on older households who might appear insulated. Research by the Resolution Foundation found retired households and working-age households in which nobody works accounted for just over a third of those it described as "housing pinched", around 110,000 households14. For how Bank Rate changes feed into mortgage payments, see how a Bank Rate change affects your mortgage payments.
What pushed household costs up or down: housing, food, transport and energy
The composition of HCI inflation shifted noticeably between March and June 2026. Contributions to the annual rate from costs in housing and household services fell from 1.22 to 0.45 percentage points over that period1. Decreases were also seen in contributions from food and non-alcoholic beverages, which fell from 0.45 to 0.20 percentage points1.
Working in the other direction, contributions from transport rose from 0.58 to 0.79 percentage points between March and June 2026, helped by higher contributions from fuels and lubricants, which increased from 0.14 to 0.56 percentage points1.
How the make-up of household cost inflation changed between March and June 2026.
Energy has been the biggest single swing factor across the whole period. Contributions to the annual HCI rate from electricity, gas and other fuels decreased from 0.39 to 0.13 percentage points over the course of 20258, and in the earlier period from March to June 2024 the contributions from these categories dropped by 0.67 percentage points7. For low-income households the effect of energy prices is larger than for high-income households: the difference in contributions between the two groups fell from 0.13 to negative 0.19 percentage points between March and June 20261. How energy prices are regulated is explained in the guide to the energy price cap.
The Scottish Government's Economic Bulletin reported household costs in the UK rose by 4.1% in the year to September 2025 for the average household, with mortgagor and other owners and social and other renters seeing annual rates of 4.2% and 4.1% respectively11. The wider story of the 2021 onwards price surge is covered in the cost of living crisis.
How the ONS builds the index: prices, baskets and scanner data
The HCI is built from the same price collection machinery as consumer price inflation, but with different weights and extra housing components.
Prices come from several routes. The ONS consumer price statistics are compiled using a representative sample of approximately 760 goods and services, reviewed annually, and their prices are collected from approximately 20,000 outlets within the UK15. Around 160 items are collected centrally, and the Living Costs and Food Survey, a continuous survey of the expenditure patterns of private households based on a sample of around 6,000 households per annum, provides the spending data15. In Northern Ireland the survey is run by NISRA, and its primary publication is the Family Spending report, produced annually by the ONS16. Independent trackers note that each month around 180,000 prices are collected to calculate inflation17, and banks such as Halifax describe the ONS as monitoring the cost of over 700 goods and services that consumers commonly buy18.
A significant change arrived in 2026. The ONS has introduced scanner data covering approximately 50% of the grocery market from the February 2026 index1. Instead of collecting 25,000 prices per month directly from shops by price collectors for that share of the market, it now uses approximately 300 million price points derived from sales of over a billion units of products per month1. The same change was signalled in the December 2025 release8.
Spending patterns set the weights. For the HCIs, the weight of each component in a household group's "fixed basket" is based on the average household's share of expenditure, the democratic weighting described earlier2. House price data used for stamp duty and related costs come from the properties in the UK House Price Index (UKHPI) basket from Quarter 4 of the previous year, valued at January prices2. The UK HPI itself is calculated by the ONS and Land & Property Services/Northern Ireland Statistics and Research Agency19, and is explained in the guide to the UK House Price Index. The inflation basket page covers which prices are collected and how.
The stages from collected prices to published HCI rates for each household group.
Limits of the HCI: official statistics in development
The HCIs are classified as official statistics in development, a status the ONS states plainly in every release: "These are official statistics in development"1. The methodology guidance confirms the classification2, and the status is also noted in the ONS consumer price inflation releases20 and in Welsh Government statistics using a similar designation22. The practical meaning is that the ONS advises caution when using these data, as estimates may be revised6.
The classification matters because accredited official statistics have been assessed as fully compliant with the Code of Practice for Statistics, a standard the ONS's household income statistics, for example, have met23. The HCIs have not yet been through that assessment.
Corrections have happened, and readers comparing releases should know about them. In 2024 the ONS corrected an error regarding comparisons with CPI that came from using an outdated CPI weights file, affecting Figure 2 and Tables 26 and 27 in the datasets7. The ONS stated that all CPI publications and HCI data were otherwise unaffected, as were the main points in its publications6. It estimated the correction would increase the measured contribution of democratic weighting to the difference between HCI and CPI by up to 0.51 percentage points in January 2023, when HCI annual inflation was 12.1%6. Later, the annual growth rate for all households in May and September 2025 was found to have been overstated by 0.1 percentage points, and the April 2025 figure for income decile 10 was overstated by 0.2 percentage points8. A dataset revision also removed the phrase "over 12 months" from the title of Table 2, with no impact on the indices themselves24.
Release dates and where the figures are published
The HCIs are published every quarter, presented as monthly data, with each publication including three consecutive months2. The April to June 2026 edition of the reference tables was released on 28 August 2026, and the next release is scheduled for 27 November 202624. Earlier releases followed the same quarterly rhythm: the October to December 2025 edition gave a next release date of 28 May 20268, which was met with the January to March 2026 estimates5.
The weights used also lag the figures. The most recent estimates have been compiled using the weights for February to December 2025, because the 2026 weights could not be updated in line with the standard methodology1. The same pattern appeared in earlier years, with weights for February to December 2024 used for the 2025 estimates8, and weights for February to December 2023 used before that7.
The underlying data is published as reference tables covering inflation rates, indices, weights and contributions for income deciles, tenure types, retirement status and households with and without children, as monthly data24. The ONS's Consumer Price Inflation team handles enquiries24. Related statistics follow their own schedules: the UK House Price Index is published on the second or third Wednesday of each month, with Northern Ireland figures updated quarterly25.
Using the HCI to understand your own household's costs
The HCI is most useful as a way to see which group your household sits in and how its costs have moved, rather than as a precise personal measure. The published groups are broad: income deciles, tenure types, retirement status and households with or without children2. A household that is, say, a private renter with children on a low income will find its circumstances spread across several of the published series rather than in one.
Some costs are not yet captured. Payday loans are not currently in scope of the HCIs' weight, as no suitable data source has been identified2. The methodology also continues to evolve: improved private rental and second-hand cars indices were introduced in February 20247, and grocery scanner data covering around half the grocery market came in with the February 2026 index1.
Because the HCIs are official statistics in development, the ONS advises caution, and figures may be revised6. For decisions that depend on an official inflation measure, such as the uprating of benefits, the State Pension and tax thresholds, the linked measures are consumer price inflation measures, not the HCI; the site's guide to how inflation sets increases to benefits, the State Pension and tax thresholds explains that process. For understanding how your own group has fared, whether renters, mortgage holders or pensioners, the HCI is the measure designed to answer the question.
Sources25 cited
- Household Costs Indices for UK household groups: April to June 2026 Office for National Statistics, 28 August 2026
- Calculating the Household Costs Indices: methodology Office for National Statistics, 28 May 2026
- Useful terms and acronyms Pension Protection Fund
- How could CPI indexation affect pension income? Pensions Policy Institute, 25 January 2011
- Household Costs Indices for UK household groups: January to March 2026 Office for National Statistics, 28 May 2026
- Household Costs Indices for UK household groups: January to March 2024 Office for National Statistics, 30 May 2024
- Household Costs Indices for UK household groups: April to June 2024 Office for National Statistics, 28 August 2024
- Household Costs Indices for UK household groups: October to December 2025 Office for National Statistics, 26 February 2026
- Household Costs Indices for UK household groups: January to March 2025 Office for National Statistics, 29 May 2025
- Detailed household expenditure split by household composition, UK, financial year ending 2023 Office for National Statistics, 25 October 2024
- Scottish Economic Bulletin December 2025 Scottish Government, December 2025
- Rents hiked on the majority of homes in the last year: what are your rights? Which?, 2023
- Landlords still planning rent rises despite reform Mortgage Professional America, 2026
- Housing pinched: understanding households' spend on housing costs Resolution Foundation, 14 August 2014
- Consumer price inflation, UK: methodology Office for National Statistics, 25 March 2026
- Northern Ireland Living Costs and Food Survey Northern Ireland Statistics and Research Agency, 2026
- The impact of inflation Trust for London, 2026
- Cost of living FAQs Halifax, 2026
- UK House Price Index for July 2026 HM Land Registry, 16 September 2026
- Consumer price inflation, UK: August 2026 Office for National Statistics, 2026
- Consumer price inflation, UK: July 2026 Office for National Statistics, 2026
- Gross Disposable Household Income dataset Welsh Government, January 2026
- Household disposable income and inequality, UK: financial year ending 2023 Office for National Statistics, 24 September 2024
- Household Costs Indices for UK household groups: reference tables, April to June 2026 edition Office for National Statistics, 28 August 2026
- UK House Price Index for April 2026 HM Land Registry, 17 June 2026







MoneyHelperFree, impartial money and pensions guidance, set up by government
Citizens AdviceFree advice on money, consumer and legal problems in England and Wales
GOV.UKOfficial information on tax, benefits and government services
StepChangeFree debt advice and solutions from a charity
FSCSProtects your money if a bank, insurer or investment firm fails