Compliance review of Household Cost Indices

Published:

Overview

At the time of this report, the Household Cost Indices produced by the Office for National Statistics (ONS) are published as official statistics in development.

The Household Costs Indices (HCIs) aim to reflect inflation as experienced by different types of households in the UK (in terms of income, home ownership and so on), based on the costs they directly face.

This review makes recommendations to inform the development of these statistics and to strengthen their compliance with the Code of Practice for Statistics (the Code).

Why we did this review

Alongside its transformation of consumer price statistics, ONS has developed the Household Cost Indices (HCIs). Since December 2023, each quarter, ONS has published monthly estimates of the HCIs. ONS labels these estimates as official statistics in development, reflecting its intention to make further methodological improvements.

We decided to conduct this review given the significant user interest in the HCIs. The review aims to inform further developments and provide direction on the steps needed to achieve stronger compliance with the Code.

Findings




coins-pounds-sterling

Many users want a price index that better reflects households’ direct experience of inflation than the headline consumer prices index (CPI). The CPI is based on recognised economic principles and is the Bank of England’s headline inflation measure. The development of the Household Cost Indices, which by comparison capture more of the direct costs households face and are broken down by household type, represents an important step in meeting the need for an inflation measure which better reflects diverse household experiences.



calculator-coin-documents

The calculation for the HCIs was developed diligently and with due consideration of quality. ONS drew on external expertise, for example using the seminal paper on the topic (Astin & Leyland, 2015) and hosting a workshop with the Economic Statistics Centre of Excellence (ESCoE) to inform development, including in decisions about weighting and interest repayments on mortgage and non-mortgage loans.

data-charts-paper

The 2025 statistical releases were affected by errors, including in the calculation of education and stamp duty. However, the team at ONS has recognised this, and is turning its focus to improving the quality of these statistics. There have been no further errors so far in 2026.




calendar-clock

Publishing these estimates on a regular, quarterly basis from December 2023 was an important milestone. The publication of these statistics is lagged compared to other price statistics, like the Consumer Prices Index (CPI). CPI statistics are available 2–3 weeks after the reference period, while the HCI estimates are available around 8 weeks after the reference period. Changing this lag would be difficult because of the data sources involved, but it does limit the usefulness of the HCIs for some users.




thumbs-up-speech-bubble

Overall, users are happy with ONS’s engagement with them. They feel they receive regular updates and that the producers are open and responsive to feedback. For example, policymakers developing metrics for the Child Poverty Strategy praised ONS’s openness and responsiveness when exploring the use of the HCIs as a metric. This has led to its use as an effective metric for the policymakers, providing a solid basis with potential for further improvements.

Our judgement

ONS has developed the Household Cost Indices with careful consideration of methods and regular, positive user engagement. However, the errors observed in 2025 indicate a need for further quality improvements, and the publication lag limits their value for some users.  

The pace of development of the HCIs has varied over time, reflecting that the producer has had to juggle competing priorities across the breadth of price statistics. ONS will need to ensure that, in this important period for the statistics, it has the capacity to progress this development, including taking forwards the recommendations of this review.

We have identified three recommendations to strengthen the Trustworthiness, Quality and Value of the HCIs in line with the Code:

  • In order to ensure that the Household Cost Indices more accurately capture the costs faced by households, ONS should continue to pursue improvements to:
    • the calculation, including ongoing and proportionate developments to interest repayments on mortgage and non-mortgage loans, other non-capital repayments (for example, student loans), and reducing the dependence on the Wealth & Assets Survey in capturing loan and overdraft data.
    • the production process, including quality assurance, to minimise the risk of the errors observed in 2025 and facilitate the publication of robust, high-quality statistics on a regular cycle.
  • To better meet user needs for timeliness, ONS should explore the feasibility of aligning the publication schedule of these statistics with other inflation and price statistics, and seek to mitigate the impact of the time lag for affected users.
  • To improve transparency and accessibility, ONS should broaden its outputs and communications so that non-experts can better understand key information about the HCIs – including what they are (and what they are not), their purpose, and calculation basics including uncertainties. It should also improve navigability between different technical resources for more-expert users. This improvement could be incorporated into ONS’s ongoing development of the technical guidance for the Consumer Prices Indices.

In addition, given the evolving nature of household price statistics, ONS should maintain an open and active focus on user needs to ensure that decisions around these statistics are continuously developing to meet those needs.

Next steps

We expect ONS to provide OSR with regular updates on progress against recommendations during the next two years. ONS must also consider, as this work progresses, the stage at which it should remove the ‘in development’ status of these statistics. This would then be followed by a full assessment of these statistics to determine whether they can be labelled as accredited official statistics.