© Crown copyright 2018
This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3 or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU, or email: email@example.com.
Where we have identified any third party copyright information you will need to obtain permission from the copyright holders concerned.
This publication is available at https://www.gov.uk/government/publications/quality-assurance-of-administrative-data-in-the-uk-house-price-index/valuation-office-agency-council-tax-valuation-lists
This is the Quality Assurance of Administrative data (QAAD) of Valuation Office Agency (VOA) Council Tax Valuation Lists used in the UK House Price Index.
The UK House Price Index (UK HPI) measures the change in the price paid to purchase residential property in the United Kingdom. A number of different administrative datasets are used in the production of the monthly HPI using a technique known as hedonic regression.In simple terms, hedonic regression is a technique which accounts for the changing quality of property transacted each period to isolate only pure price change, so that the change in price is not distorted by differences in the composition of property sold (for example, you cannot directly compare the price of a one bedroom property sold in one period with a three bedroom property sold in another).
The hedonic regression approach requires detailed information on the characteristics of property sold, both regarding the physical attributes of the property (such as size, floor space) and the location of the property (such as what type of neighbourhood, where in the country). For the production of the UK HPI this data is obtained from a variety of administrative data sources that cover the price paid for transacted property (such as the Price Paid Dataset collected by HM Land Registry for England and Wales), the attributes of a property (such as the Council Tax Valuation List maintained by the VOA and characteristics related to the location of the property (such as the type of neighbourhood where the property is situated, defined by the Acorn classification from CACI Ltd).
This document will focus on the Council Tax Valuation list which is maintained by the VOA. The VOA is an executive agency of Her Majesty’s Revenue and Customs (HMRC). It has been responsible for banding properties for Council Tax since the tax was first introduced in the early 1990s, and prior to the introduction of the Community Charge VOA was responsible for the earlier system of domestic rates.
It is the duty of VOA to act fairly and impartially and to make sure that each domestic property is correctly assessed and placed in the right band. Dwellings are allocated to their correct band based on their open market value as at 1 April 1991 for England (feeding into the 1993 list) and 2003 for Wales (feeding into the 2005 list). In order to make this assessment, VOA collect and maintain a set of property attributes which provide pertinent details of the dwelling and its location. There are currently 25 million properties registered on the Council Tax Valuation Lists.
2. Summary of process
Transaction data are provided to the Office for National Statistics (ONS) on the following information captured by VOA for the purpose of maintaining the Council Tax Valuation Lists:
- property address
- floor area
- number of rooms
- number of bedrooms
- property type
- effective date of entry into the list
Each month, these data are matched to the latest set of property transactions (by property address) with the resultant data then used in the production of the latest month’s house price index.
3. Assessment of the Council Tax valuation list data using the Administrative Data Quality Assurance Toolkit
The production and publication of house price data can be considered as medium profile, in that there is wider user and media interest in the results that are published, with moderate economic or political sensitivity.
Considerable reliance is placed upon Local Authorities to notify VOA of any changes to the valuation list. As there are many different data collection bodies (local authorities) and the data provided by VOA are among the most important variables in describing the price of a property, the level of risk of quality concern is classified as high. This means a comprehensive level of assurance is required for this data source.
3.1 Practice area 1: operational context and administrative data collection
Under the Local Government Finance Act 1992, it is the statutory requirement of VOA to maintain accurate valuation lists for Council Tax. VOA makes every effort to ensure the accuracy of the data used, however they only collect data that is needed to place an accurate band on the property. Property attribute data is only updated where information comes to VOA’s attention that a valuation list entry might be inaccurate and VOA does hold contemporary data for all dwellings.
There is also a Data Protection Act requirement for VOA to update the information that is held on the taxpayer when contact is made with them.
VOA is responsible for maintaining the Valuation Lists and this generally occurs where they come across information that indicates a List entry may be incorrect. Considerable reliance is placed upon the local authorities to notify VOA of any changes (including new builds, demolitions or alterations). There can be some variability in this process and lists are only updated when new information has been brought to VOA’s attention. Therefore there are some records in the list that have not been updated as regularly as others. Changes made to properties such as improvements are also difficult to capture in the Valuation Lists. A property that is subject to improvements cannot have its band reviewed and will not be updated on the VOA systems until the property has been sold.
VOA works hard to standardise its approach to data collection but given the large variety in dwellings there have being some different local interpretations over the years. This has resulted in some minor local differences in data collection. For example, in some local authorities gaining the information by inspection may be more prevalent than in other areas. Quality reviews take place and analysis has been conducted to compare the quality of the data collected through an inspection against that collected through other third party sources. There was found to be no measurable decrease in the quality of the data when an inspection did not take place.
There are some legislative requirements which place demands on the banding of the property, two important ones are: dis-aggregation where properties which are adapted for separate use can have their own list entry despite being within a single hereditament; and houses in multiple occupation, where requirements are met, can have individual occupations aggregated into one list entry. An additional requirement is made for Care Homes so that a single band can be placed on the entire property.
Transaction data are provided to ONS on the following information captured by VOA for the purpose of maintaining Council Tax Valuation Lists: address; property details such as area of the property, number of rooms, number of bedrooms, property type; Council Tax band and effective data of entry in the List.
VOA treat this information as personal property data and access is restricted to those who need to use it. As an executive agency of HMRC, the agency is governed under the Commissioners for Revenue and Customs Act (CRCA) 2005 which restricts access to the data. A legal gateway has been put in place to enable VOA to provide this information to ONS. This was done under the Statistics and Registration Service Act 2007.
The valuation list entries and attribute details are updated on an ongoing basis to reflect changes in taxable dwellings, and the changes to entries and attributes are maintained in the VOA database. The 3 main sources for updates are:
New entries – for example a new house is built and the dwelling attribute data is collected and entered in the database, and a new entry appears in the appropriate valuation list.
Changes to the Council Tax Valuation Lists – these changes are based on new information and include: insertions and deletions; amendments to dwellings as a result of physical changes, changes in the use of a property, dwellings split in two or more units, mergers to create one or more units; or decrease and increases to the band. The majority of cases in this category are those where a property is subject to improvements and then sold. The band can only be reviewed against the improvements where the property is subject to a ‘relevant transaction’.
Challenges against Council Tax Lists – a taxpayer can request a review of their list entry (challenge), in some circumstances they have the right to put in a proposal and subsequently appeal and if they are not satisfied with the outcome the case can be heard by the Valuation Tribunal. The majority of these challenges are to review the band but outcomes of a challenge can be band decreases and increases, or less often any of the other changes as above.
Historically the main method to collect data was through an inspection of the property with retention in paper records. This property attribute data was transferred from these paper records to electronic storage and forms the bulk of the VOA present database.
For new data and amendments to data, VOA use more modern methods to collect information including the following sources:
Verification by the taxpayer of the data already held. Agents acting for a taxpayer are asked to complete a questionnaire, which is then compared to internal records.
Third party sources such as: information is available to the public, for example local authority planning records; information available from local authority’s Finance Department building control records; and subscription to third party property data sources who are able to satisfy data accuracy standards.
External inspection of the property where the relevant details visible from the road can be confirmed.
Pre-arranged internal inspections of the property where attributes can only be confirmed internally or a re-measurement is required.
Within the VOA dataset, 2 variables are of interest and used in the UK HPI:
- area of accommodation- the area of the dwelling is shown in terms of EFA (Effective Floor Area) for flats and maisonettes or RCA (Reduced Covered Area for houses and bungalows
- number of habitable rooms – excludes bathrooms, conservatories, kitchen and utility rooms.
Find out further information on the VOA data and available variables in the VOA manual.
3.2 Practice area 2: communication with data supply partners
Introductory meetings took place between VOA and ONS to discuss the purpose and methodology used for the data, to make and agree data sharing agreements and to ensure the disclosure of the Council Tax Valuation Lists data was both reasonable and proportionate, as required under the Commissioners for Revenue and Customs Act. These meetings also included theassurance of efficient and accurate data collection, and finding solutions to the restrictions that legislation places on the sharing of Council Tax data.
A Service Level Agreements (SLA) exists between VOA and ONS involving the provision of the data by pre-set dates each month. Under this agreement, ONS require three month notice prior to any changes in processing code, data collection or format of data delivery to ensure sufficient time for amendments and testing. These agreements are reviewed on an annual basis to ensure the aims, benefits and terms of the SLA are mutually acceptable or do not require adjustments to be made.
VOA Council Tax Valuation List data are supplied to ONS via encrypted email data files on a monthly basis. Passwords are only supplied to those who have access to the data.
Quarterly face to face meetings are held between the VOA and Prices Division within ONS. Other ad-hoc meetings can be scheduled if necessary.
A number of other areas within ONS are currently investigating using the VOA Council Tax data for their own purposes. One example of this is its potential for replacing the questions on number of rooms and bedrooms on the 2021 Census questionnaire in England and Wales. As part of this wider use, and improved future engagement between ONS and the VOA for this dataset, ONS are looking to set up a single point of contact and single data feed between ONS and VOA. Further information on this approach and any quality assurance implications will be provided when the arrangement is in place.
3.3 Practice area 3: QA principles, standards and checks applied to data supplies
Following collection of data by case workers, the data attributes are input manually into the VOA central operational database. This input is subject to quality checking by the individuals inputting the data and undergoes a number of validations which are built into the software recording system to ensure the properties’ attributes are consistent with other comparable properties and as complete as possible.
Specific checks are completed on data completeness and missing and incompatible code. For example: The number of rooms should always be greater than the number of bedrooms (except studio or bedsit). Caseworkers are required to add at least eight of the 16 codes required and at least five of properties code within each local authority is monitored on a periodical basis. However, whenever VOA come into contact with a taxpayer, inspect a property or receive or research information about a property, all property attributes currently held by VOA should be checked and any missing attributes should be completed, if possible.
The minimum standard is that on clearance of any report, proposal or appeal, the ‘live’ assessment will have the 8 primary property attributes;
- group (broadly a signal of architectural style)
- type (detached, semi-detached)
- area (meaning size)
- number of room
- number of bedrooms
- number of bathrooms
- number of floors
Further information on these property details can be found in VOA’s Property Details Guide.
Searches have been run on missing and/or incompatible code. In the latter case, exceptions lists were produced to pick up incompatible information i.e. bungalows with multiple floors and flats that are coded as semi-detached.
Managers in VOA also undertake weekly spot checks on new and changed entries to ensure quality. Outliers and errors are flagged and signed off by managers before any amendments are made to the valuation list. There are six separate checks that cover data and there is a requirement for managers to quality assure at least one case for every staff member per week.
Once processed, the information is passed back to the local authorities to check the new valuation list entry is consistent with the information they hold.
VOA attempt to verify property attribute data with property owners or occupiers during any contact with them. Any changes due to challenges or new information will be amended in the valuation lists as soon as known and checked via the same input methodology.
The production team within ONS who use the VOA data have also shadowed VOA colleagues to ensure understanding of any data assurance practices. Quality checks performed by VOA on the data are predominantly at record level: either when the data is initially inputted or with property owners during any contact with them. Given ONS have access to record level VOA data, additional quality checks have been performed by ONS on the VOA data by matching against census and other data sources. This is described in the next section.
3.3 Practice area 4: Producers QA investigation and documentation
The VOA dataset is one of a set of characteristics (about a property) that is used an input in the production of modelled house prices, which in turn are used in the production of the monthly UK House Price Index (HPI). Read about the UK HPI production methodology used.
Evaluating the data source
From the VOA dataset the two variables used in the UK HPI are ‘Number of rooms’ and ‘Total Floor Area- measured in m squared ’. Quality assurance checks have taken place on the full stock of VOA data. This includes:
- checks for completeness - 98.5% of records have values for ‘number of rooms’ and a similar proportion have values for ‘Total floor area’
- duplicates- minimal duplicate records overall were found (<0.5%)
Further coverage checks are conducted on an annual basis during the construction of the weights each year.
Checks on number of rooms and number of bedrooms have also been made against the 2011 Census. This is included in an assessment of the Valuation Office Agency dataset published by Census. It is concluded in this report that the VOA rooms data is of sufficient statistical quality for use in the 2021 Census subject to some further development.
Comparisons were made at address level by linking 2011 Census and VOA records as at 2011. Estimates for the coverage of the VOA data are not currently available. However, a high match rate (94%) achieved between the 2011 Census and VOA data indicates the coverage of the VOA data are good for statistical purposes.
There are definitional, coverage, timing and data collection differences between the 2011 Census and Valuation Office data. The 2011 Census question asks respondents to count the number of rooms available for use by their household. The definition for number of rooms in VOA data is the total number of “living rooms” as reported by the “data collector”. The difference between the 2 datasets for rooms that are included and excluded in the definitions are summarised in Table 1.
Table 1: Room types included and excluded in the 2011 Census and VOA
|Halls and landing*||Excluded||Excluded|
|Separate dining rooms||Included||Included|
|Dining areas within other rooms*||Excluded||Excluded|
|Source: Office for National Statistics|
|Note 1: Room types marked with an asterisk * are not included by both the census and the VOA.|
|Note 2: In the VOA dataset, conservatories are included as a separate variable.|
The impact of this definitional difference is that for the majority of addresses, the number of rooms in the VOA will be at least one room less than that shown in the 2011 Census. This is because of the definitional difference regarding kitchens.
It is also important to note that the 2011 Census Quality Survey (CQS)(which was used to assess the accuracy of responses to the 2011 Census) found quality issues with the number of rooms question. The agreement rate between the CQS and the 2011 Census for the number of rooms question was relatively low at 67% (all other household questions had an agreement rate of over 88%).
On this basis, the “agreement rate” should be considered as the aggregation of the categories where the number of rooms recorded on the VOA dataset is the same as that recorded on the census, or where the number of rooms recorded on the VOA dataset is one or two less than on the census. After adjusting for definitional differences and the quality of the census, data an agreement rate of 85% is achieved for rooms between the VOA and Census data.
It was noted earlier in this document that in the VOA data collection that some local authorities’s may be gaining the information by inspection more than others. Analysis published by the census found that agreement rates between VOA and census data at the local authority level range between 70% and 95%. It was found that the local authorities with the lowest agreement rates are in rural areas. These areas are likely to contain a larger proportion of bigger properties. It is also known from the 2011 Census Quality Survey that respondents in properties with four, five, six and seven rooms had the most difficulty in understanding the census definition of a room. Given these findings, a lower proportion of data collected by inspection is not thought to impact the quality of the data collected.
Checks on ‘Total floor area’ were made against Energy Performance Certificates (EPC) data made available by the Department for Communities and Local Government. Comparisons were made by linking records by address. It was found that for detached, semi-detached and terrace properties, the floor space is between 2% and 10% larger in the VOA data compared to the EPC data. This can be attributed to differences in the data collection. For these property types, the VOA floor space is measured from the external walls of the building, while for the EPC data this is measures by the internal wall of the building. Further information on this can be found in the council tax measuring practice and Energy Performance Certificates Technical notes.
The floor space for flats however are around 40% lower in the VOA data compared to the EPC. Similarly this is due to measurement differences. Effective floor area (EFA), as described in the RICS Code of Measuring Practice, excludes bathrooms, corridors, hallways and landings. Living rooms, bedrooms and kitchens are included. As the ECP measures flats the same way as it measures all other properties (the internal face of the walls) there are areas which are excluded in the VOA area calculation but included in the EPC data. A property type variable is included in the model to estimate house prices which controls for this difference in collection practice by property type. Similar consistency checks will be conducted on an annual basis during the construction of the weights each year.
Using the data in production
Each month, HM Land Registry price data is matched to the VOA attribute data. Matching is done by address and a high match rate of 95% is achieved. Data from other data sources are also joined to the dataset prior to the modelling process.
The modelling process used in the production of house price data includes an automated assurance process that assesses modelled house prices for property with a certain set of attributes against the price for a similar property. If the modelled price is substantially different (meaning it exceeds a predefined tolerance) then the price is excluded from being used in the final house price estimate. Around 50 transactions a month (out of around 70,000 transactions) are removed as part of this process. These transactions still contribute towards the volume of transactions published.
The House Price Index modelling process used can also account for those records where a match cannot be made between the VOA data and price data provided by HM Land Registry (around 5% of cases). Each attribute used in the hedonic regression model is given a weight that represents the relative importance of that attribute in explaining house prices. If a record being used in the model has a missing attribute, then the weight of that record is adjusted downwards to represent how important the missing attribute is. This process allows the use of all property transaction data in the calculation of average house prices each month, even though some attribute data could be missing.
For example, a record with no missing attributes would receive a weight of 1, while a record which has all its attributes except rooms will receive a weight of 0.77 and so will contribute slightly less to the final modelled estimate. These weights are calculated from multiple models which are run on an annual basis to determine the importance of each variable.
Following the running of the model, test statistics are analysed to ensure the model has run correctly and fit successfully. This includes analysing the R squared of the model (model fit) and significance of the explanatory variables. An R squared of around 0.8 is achieved. This means that 80% of the variation in price is captured by the explanatory variables. An R squared of 0.8 is high. The old ONS HPI had an R squared of around 0.7.
The data is then aggregated with the resulting series analysed by various breakdowns, over time, and against other published sources of house price growth. Any unexpected movements within the series are explored through the record level data. Monthly curiosity meetings are held to review the new data and discuss any long term trends in the data and its drivers.
4. Strengths and limitations of data
There is good coverage of addresses in England and Wales (94% Census records matched to VOA data) and an agreement rate of 85% for rooms when adjustments are made for definitional differences and the quality of census data. Similarly, differences in floor area between VOA and EPC data can be attributed to definitional differences.
However, there are some acknowledged limitations in the data:
Considerable reliance is placed upon local authorities to notify VOA of any changes (including new builds, demolitions or alterations). There can be some variability in this process and lists are only updated when new information has been brought to VOA’s attention. Therefore, there are some records in the list that won’t have been updated as regularly as others. While this may contribute some volatility, particular to new builds given the fewer transactions, overall, this data source is judged to be of good quality for the use which it is being put in the UK House Price Index