Dear Jonathan

Compliance Check of Valuation Office Agency (VOA) Council Tax statistics

We have recently completed a compliance check of your council tax statistics publications, Stock of Properties, 2023 and Challenges and Changes in England and Wales, March 2023 against the Code of Practice for Statistics. I hope that our reflections are helpful as you continue to develop these important official statistics. Some of the findings may also be more widely applicable to your non-domestic rating statistics, and we encourage your team to consider that broader applicability. If you would find it helpful for us to specifically review those, please do get in touch.

Your council tax publications are valuable to users who are trying to understand how council tax affects them. Breakdowns by house characteristics and number of bedrooms in the Stock of Properties release help users compare their banding with others. The Challenges and Changes publication indicates to users how likely a banding review is to increase, decrease or not change a property’s banding.

Our review found a range of positive features that demonstrate the trustworthiness, quality, and value of the statistics:

  • It is commendable that VOA has made its data available to other government departments. The use of council tax property attribute data by the Office for National Statistics (ONS) to supplement census data, and the Department for Levelling Up (DLUHC) use of the valuation list for its council taxbase statistics highlight the impact of VOA’s positive approach to data sharing and the additional statistical insights and increased efficiencies that this can lead to. For example, sharing VOA property attribute data with ONS has meant that a previous question on the number of rooms in a dwelling could be removed from the 2021 Census. VOA property attribute data is also a key source used to calculate the UK House Price Index.
  • The statistics provide granular and clear insights. They include helpful contextual information which allows users to correctly interpret the statistics and their structure is easy to follow. There is a good level of commentary and background notes to help users understand how council tax works, which will aid users in using the statistics appropriately. For example, the helpful flowchart showing the stages of the banding appeals process in the Challenges and Changes background notes outlines each stage from proposal to final decision.
  • The explanations of the methodology for the statistics should be commended, for example it is explained that the statistics are produced using ONS’s statistical geographies and National Statistics Postcode Look-up, which will help users understand the comparability of the statistics.
  • You have demonstrated a commitment to trustworthiness and transparency by making users aware of data limitations. This includes noting a key limitation, that the attributes of properties are updated only when they are sold in the background notes. The inclusion of a revisions policy and quality policy and details around VOA’s statutory duty in the background information, also show a commitment to transparency. Your team told us about your plans to publish more information on the extent of revisions made in the statistics. This should reduce the likelihood of misinterpretation and misuse, especially in the advent of legislative changes such as the planned revaluation of council tax banding in Wales, which you told us may lead to improved data quality and potential revisions to the statistics.
  • The implementation of Reproducible Analytical Pipeline (RAP) principles into the production of the statistics to improve the efficiency and reliability of the process is commendable and demonstrates a commitment to continuous improvement.
  • Your plans to further develop the statistical outputs, for example by introducing a flows measure between domestic and non-domestic tax, will give users a better understanding of the stock of taxable buildings. The potential introduction of dashboards and maps is also exciting and will be valuable to users through new ways to interact with your statistics.

To further enhance the trustworthiness, quality and value of these statistics, we have identified ways the statistics and their presentation could be improved:

  • Within the background documentation for each release you welcome feedback from users via email. However, there is little published information about your overall approach to user engagement. Having a clear user engagement strategy, especially in the context of the future planned development of these statistics, will allow users to benefit from knowing what changes are coming and facilitate an ongoing dialogue with your users.
  • The background documents for each publication are clear and well-structured but some of the information is quite general. Including specific details about approaches to data collection and assuring data quality for each set of statistics would reassure users. Publishing information, for example, around potential uncertainty in the local authority and house attribute data captured as part of a property valuation, will facilitate the appropriate interpretation and use of the data by users. This is important given the wider use of this data by other government departments. More published information, in the form of a data journey for example, could give an indication as to where uncertainty in the data may arise.
  • It was good to hear that you have previously used Quality Assurance of Administrative Data (QAAD) principles when exploring new data sources, however, we would encourage you to communicate this work to users as part of your documentation about quality. This will help users understand the quality of the source administrative data and how it can be used appropriately. A thorough appraisal of data quality will be especially important in the context of changes to data sources such as the planned changes to VOA administrative systems and the revaluation of Welsh council tax bandings. Given the extensive use of the VOA council tax and property attribute data across government, we encourage you to work with these stakeholders while you undertake future reviews of administrative data quality. This will allow you to understand data quality in relation to different uses, share resource and tailor your efforts accordingly.
  • Due to the difference in council tax policy between England, Wales, and Scotland it is important that the coherence and comparability of council tax statistics is made clear to avoid misleading comparisons. Stating which statistics can be compared and including links to relevant statistics is a good step to reducing the likelihood of misuse. The appropriate comparison of statistics is especially important with an approaching general election.
  • We welcome your commitment to reduce the number of individuals on your pre-release access (PRA) lists for both publications. This shows your commitment to trustworthiness and will reduce the likelihood that the statistics are shared more widely than is essential ahead of publication. However, your PRA lists for each publication are long, and to support ensuring proportionate access we recommend that the PRA lists are reviewed each time the statistics are published.
  • The cover sheets and notes provided in the Microsoft Excel data spreadsheets provide a good amount of information that will help users use the statistics appropriately. We recommend that similar notes are included with the CSV files as just providing raw data could lead to misinterpretation and/or misrepresentation.

I would like to thank your team for their positive engagement during this review. Please get in touch if you would like to discuss any aspects of this letter or if we can be of further assistance.

I am copying this letter to Tetyana Mykhaylyk, Director of Information, Data and Analysis, VOA; Colin Yeend, Head of Research & Analysis, Information, Data and Analysis Directorate, VOA; Sarah Windass and Anna McReady, the responsible analysts; and Sean Whellams, Head of Profession for Statistics, HM Revenue and Customs.

Yours sincerely

Mark Pont