Guidance

Housing Benefit speed of processing statistics: methodology note

Updated 31 January 2024

Applies to England, Scotland and Wales

© Crown copyright 2024

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: psi@nationalarchives.gov.uk.

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/housing-benefit-speed-of-processing-statistics-methodology-note/housing-benefit-speed-of-processing-statistics-methodology-note

1. Context of the statistics

1.1 What is Housing Benefit?

Housing Benefit (HB) is an income related benefit which is provided to households in order to help them meet housing costs for rented accommodation. This is available to those who are either unemployed, on a low income or in receipt of other benefits.

The claimant may get help with all or part of their rent. There is no set amount of HB for claimants as entitlement is dependent on whether they are renting privately or from a Local Authority (LA) or council.

For claimants meeting certain conditions, HB has been replaced by Universal Credit (UC).

1.2 Eligibility requirements

The eligibility requirements for making a new HB claim are:

  • they are single and have reached State Pension (SP) age
  • they are living with a partner and both have reached SP age
  • they are living with a partner and one has reached SP age and started claiming Pension Credit before 15 May 2019
  • they are in supported, sheltered or temporary housing

Previously, people in receipt of the Severe Disability Premium (SDP) were also included in eligibility requirements. However as of January 2021, the Severe Disability Gateway has closed, so claimants must meet one of the current eligibility criteria.

Existing claimants can continue to receive HB if they were in receipt of it before 15 May 2019 and had reached SP age. However, if a claimant reports a change of circumstance to an existing HB claim and it is subsequently stopped, a claimant cannot reapply for HB unless they meet the current eligibility criteria. Instead, they may be able to apply for UC if they are eligible for UC. UC has replaced HB for all other eligible groups.

Read more information about HB eligibility.

1.3 What is Housing Benefit Speed of Processing?

HB Speed of Processing (SoP) refers to how long it takes for LAs to process a new HB claim or a change of circumstances to an existing HB claim.

It is the average time taken, calculated as the average (mean) processing time in calendar days, rounded to the nearest day.

The HB SoP publication is released quarterly. Monthly, quarterly and annual statistics are provided at Great Britain (GB) level, as well as the regional and LA level.

2. Purpose of the HB SoP statistics

The Housing Benefit statistics on speed of processing provide users with the volume of new HB claims and change of circumstances to existing HB claims processed by each LA, per month or quarter, and the associated total amount of time (in days) to process them.

A simple calculation is then made to determine the average time taken to process each new HB claim or change of circumstances to an existing HB claim for each LA, known as the average speed of processing.

Year-end figures are provided to explain trends without the impact of seasonal variation and are also rounded to the nearest day.

Please see section 9 which outlines the findings from DWP’s user consultation exercise in 2019 that showed these statistics are used in debates on the UC and HB systems.

3. Data source

The Single Housing Benefit Extract (SHBE) is the most comprehensive administrative DWP dataset for HB, containing individual level claims data on all HB claimants.

The data received from SHBE is classified as ‘unmasked’; meaning it contains unencrypted information including national insurance number, full address and postcode.

In order to comply with DWP’s personal information charter, the individual level data is encrypted by DWP Data Delivery Team (DDT) and personal identifiers are removed and replaced with non-identifiable markers, for example, encrypted national insurance number, or census output area codes.

3.1 Data collection

Every month each LA will submit an extract of all claims on their HB system. Note, submissions are not mandatory however are encouraged. This raw extract includes details of all claims live or awaiting decision on the extract date, claims closed since the previous extract, and various details of activity since the last extract (for example, records for changes of circumstance, appeals and overpayments). This information is compiled into a single, encrypted data extract by DDT.

3.2 Data structure

SHBE data variables used in production of HB SoP statistics

Field Note
LASER Unique identifier for each LA
SED_DATE* Date of SHBE data extract
REC_TYPE Status of claim (for example, live, closed or withdrawn)
LCHBREF HB claim reference number
CCNINO Encrypted National Insurance number
LDHBREC Date most recent HB claim was received
LDHBDEC Date of first decision on most recent HB claim
LXHBOUT Decision outcome
LDSTARTDAT HB claim entitlement start date
CDHBMADE HB claim Treat as Made Date
CCCHGTYP Type of Change (for example, automated uprating, other)
CDCHGNOT Date LA first notified of change in claim details
CDCHGEFF Date changes are effective from
CCCHGIDT If not annual uprating, how was the change identified
CDHBSUPER Date supersession decision was made on the claim
LCLHA Local Housing Authority regulations applied
CCTEN Geographical region
PASSPORTIND Passported / standard claim indicator
CDDOB Claimant date of birth
CCSEX Claimant gender
CCHBSOURCE Source of the most recent HB claim (for example, LA form, telephone, electronic channel)
LCSOFTWARE Software company

Note: Fields marked with an asterisk (*) are derived and added by DDT before statistical production

3.3 Data reference area

Individual record SHBE data is aggregated and reported at LA, regional and GB levels.

3.4 Data coverage

Each monthly SHBE source dataset includes all system recorded new claims and changes of circumstances between the first and the last day of the month.

3.5 Frequency of data collection

SHBE source data is received and made available for statistical use on a monthly basis.

4. Methodology

4.1 Data flow

On receipt of the encrypted data scan from DWP DDT, and prior to publication, the masked SHBE data are subject to a number of methodological processes and quality assurance checks.

Flow diagram showing the movement of data from source to publication

  1. Local authorities submit data to online Single Housing Benefit Extract system.

  2. DWP Data Delivery Team take in monthly extracts of raw, unencrypted Single Housing Benefit Extract data.

  3. Encrypted Single Housing Benefit Extract extract made available for DWP analysts.

  4. DWP analyst combine 5 monthly extracts to build up latest position.

  5. Statistical analysis system (SAS) code run against data to clean, link on additional markers (including geography), and exclude, retain, or aggregate information in line with business rules.

  6. Aggregate statistics used to populate ODS data tables.

  7. ODS data tables used to populate HTML statistics bulletin and aggregate statistics used to populate interactive maps.

4.2 Data compilation

Data scans used to produce statistics

For any single, reported quarter of HB SoP statistics, 5 SHBE scans are used to capture the latest HB claim position and incorporate any retrospection beyond the main reference quarter.

This means that data from the 3 months in the publication quarter along with the 2 consecutive months following the quarter are required. This ensures that the most up to date information is included in the HB SoP calculation at the time it is made.

4.3 Data validation

DWP excludes cases that, where SHBE data indicates, have taken more than 365 days to process from all HB SoP statistics. Quality assurance (QA) work indicated that the majority of these cases are a result of processing or software errors. Therefore, their inclusion in the HB SoP statistics would distort the calculation and provide an inaccurate picture of HB SoP performance.

Statistical code is then run against the individual level data to clean, link to other data sources to attach markers (including geographical) and the HB SoP measure is calculated, before aggregating up to monthly, quarterly and annual totals.

4.4 Statistics calculations

The ‘total number of processing days’ for each new claim or a change of circumstance for an existing HB claim, is calculated by looking at the difference between the ‘date of receipt’ to the ‘date of decision’ for each claim.

The HB SoP is then calculated by dividing the ‘total number of processing days’ by the ‘total volume of claims’ for a given geography (LA, region or GB).

Quarterly average year-end calculations are also used throughout the statistical release to help describe trends. This can be thought of as a moving average and helps to smoothen any seasonality in the volume and speed of processing timeseries. Included in the following table is an example of how we calculate year-end figures.

Quarter New claims total processing days New claims volumes Average SoP Year-end average So P
Q1 2019 to 2020 1871211 100952 19 *
Q2 2019 to 2020 1669012 98623 17 *
Q3 2019 to 2020 1455059 90836 16 *
Q4 2019 to 2020 1548740 92706 17 17
Q1 2020 to 2021 1643577 91588 18 17
Q2 2020 to 2021 1543845 90987 17 16
Q3 2020 to 2021 1565375 88094 17 17
Q4 2020 to 2021 1713942 90876 19 18

The table shows how the quarterly SoP at year-end is calculated, by summing the four previous quarters of total days processed and dividing by the four previous quarters of volumes to give an average year-end SoP. The quarterly SoP is shown as a comparison. All SoP figures have been rounded to the nearest calendar day. Note, average year-end figures for the first three quarters shown have been omitted as four quarters’ worth of data are required to produce the figure.

Taking the quarter 4 2019 to 2020 SoP at year-end as an example, using the figures from the table above, this is calculated as:

  • Quarter 4 SoP at year-end = (1871211 + 1669012 + 1455059 + 1548740) / (100952 + 98623 + 90836 + 92706)

  • Quarter 4 SoP at year-end = 17.08 = 17 days (rounded to nearest calendar day)

5. Quality management

5.1 Quality assurance

Data are further analysed, before the population of statistical tables, using SAS software by producing a 40-month historical pdf summary of each LA’s HB SoP for both change of circumstances and new claims. Month 40 is the latest month in the current publication quarter.

Each LA has a ‘laser’ which is just a unique identifier (number). Any LAs that SAS considers to have fallen outside of an acceptable tolerance range are subject to human (manual) intervention.

The basic aim is to check the three latest months for each LA and pay attention to any ‘spikes’ or anything that looks unusual, indicating an extreme high or low speed of processing. The full time series can be used as an indication of any previous anomalies or trends.

For any LAs identified as potentially having outliers, the underlying data can be cross-referenced and investigated alongside the data issues log provided by the LA data experts for HB SoP for any known live issues and the SHBE user group.

Example of the average days per case for a particular LA

Example LA showing a sharp increase in their SoP in month 39, indicating that there may be issues with this data

Upon checking the claim QA outputs, this doesn’t indicate that there was an abnormally high number of cases processed by this LA in the latest month

Finally, upon checking the issues log, it was identified that this LA had severe backlogs explaining their increased SoP. Therefore, the decision was made for this LAs data to remain in the publication as there was a valid reason for their increased SoP.

Furthermore, additional QA processes are in place for the HB SoP publication that are consistent across all DWP statistical production processes. These include rigorous internal team and stakeholder quality checks and multiple layers of sign-off.

5.2 Exclusions

For any LAs that cannot be identified as having genuine outliers, there is an option of excluding them.

DWP retain the right to remove any figures that are questionable and where investigation/resolution of these figures would become a proportionate burden. This is assessed on a case by case basis.

Exclusions are made where data are not fit for purpose and DWP reserves the right to withhold any figures that are not fit for purpose in line with the code of practice for statistics. Individual explanations for the exclusion of data in each publication is not routinely published.

The existing aggregation method means that quarterly data will not be available if all underlying months of that quarter are not available. Similarly, with annual data, if all underlying quarters are not available. This is to ensure that any current data are comparable with historical data.

When DWP analysts have agreed exclusions, a final caseload dataset is created.

6. HB SoP statistical outputs

6.1 ODS tables

Here is an example of the ODS tables.

After successful QA, DWP runs code that automatically populates a set of ODS tables, containing one tab for each output measure – a quarterly summary all in one table, monthly totals by LA for total number of processing days, number of claims and the calculated average speed of processing and finally a quarterly timeseries. Furthermore, for the final quarter of each financial year, the publication includes an annual table with the monthly data aggregated to annual level for the entire financial year.

Within the ODS tables, monthly figures may not sum to quarterly totals, and quarterly figures may not sum to annual totals. This is because LAs may have suppressed or missing data at a monthly level that is then not included in quarterly total calculation. And similarly, suppressed or missing quarterly data may not be included in annual total calculations.

6.2 HTML statistics bulletin

Here is an example of an accessible HTML statistics bulletin.

This is produced by firstly compiling a draft Microsoft Word version of the release. This then goes through several iterations of QA. Finally, it is then converted into HTML format by DWP analysts and published on GOV.UK by DWP content designers. This is consistent with the standard DWP statistical release procedures.

6.3 Statistical charts and maps

Within the HTML statistics bulletin, there are several charts which are created using the raw (unrounded) output data. The latest quarter data is appended to the previous data in the case of the time series charts. These are at a GB level.

The maps are created using individual average SoP LA level data. If any LA has an average SoP of zero, then their data is omitted from the maps. The map ranges (keys or bins) are created by allocating LAs to one of five bins based on an equal proportion of LAs being allocated to each bin, where possible.

7. Overview of statistical release strategy

The table shows the months when it is published and the data included in each publication

Data to Published
September January
December April
March July
June October

The publication is released on the final Wednesday of the published month noted above.

7.1 Timeliness

There is a four-month lag in data availability, due to the way the SHBE data source is obtained and the final aggregated analytical dataset is constructed (to provide the most up to date base data to calculate accurate SoP).

Therefore, taking January’s publication for example, data would include up to the end of September of the previous year.

7.2 Data revision

In line with statistical protocols – revisions will be published transparently both here and on the HB SoP collection page.

Revisions due to conceptual or methodological changes will be announced at least three months before publication and this document will be updated accordingly.

7.3 Confidentiality

Rounding policy – the number of claims and number of processing days are unrounded in the statistical ODS tables. The average speed of processing is calculated and then presented rounded, to the nearest whole day.

In the HTML statistics bulletin, number of processing days and volumes data are rounded as follows

Range Rounded to the nearest
0 to 1,000 10
1,001 to 10,000 100
10,001 to 100,000 1,000
100,001 to 1,000,000 10,000
1,000,001 to 10,000,000 100,000
10,000,001 to 100,000,000 1,000,000

Percentages are calculated using numbers prior to rounding and are rounded to the nearest whole percentage point.

7.4 Accessibility

The bulletin and tables are now in accessibility compliant format – meeting the guidelines within DWP’s accessibility statement.

8. Changes to these statistics

Any known or new data, publication or process issues that may affect users will be highlighted on the HB SoP collection page and future changes will be announced in the previous publication, or with at least 28 days’ notice, in line with standard statistical practices.

  • impact of UC roll-out – the number of new people claiming HB is decreasing as many are now eligible for Universal Credit instead
  • these statistics should be viewed in the context of the COVID-19 pandemic
  • we intend to publish this data on Stat-Xplore in the future. Further user announcements around the availability of this data on Stat-Xplore will be made in line with our standard user notification protocols (as outlined above)

9. Relevance

9.1 User needs

Findings from our user consultation exercise in 2019 show that these statistics are used in debates on the UC and HB systems.

In addition, these statistics are used within DWP’s fraud and error publication, specifically to provide an adjustment for the underlying data.

LAs use these statistics to analyse and publicly report on their HB speed of processing performance.

These statistics are used to answer Parliamentary questions and DWP Freedom of Information requests.

9.2 User satisfaction

DWP would like to hear your views on our statistical publications. If you use any of our statistics publications, we would be interested in hearing what you use them for, and how well they meet your requirements.

Specifically views on content, presentation, and suggestions for future developments on the series. Users views on the HB SoP statistics are sought through a number of mechanisms:

DWP is also making use of improving technology to better understand and improve user experience of the landing page, downloads of the statistical release and the most popular queries, as well as the technology that users employ (desktops, mobile phones, laptops) and the type of users accessing data.

10. Status of the statistics

These official statistics are compiled following the standards of trustworthiness, quality and public value set out within the statistical code of practice.

Our statistical practice is regulated by the Office for Statistics Regulation (OSR).

OSR sets the standards of trustworthiness, quality and value in the Code of Practice for Statistics that all producers of official statistics should adhere to.

You are welcome to contact us directly with any comments about how we meet these standards.

Alternatively, you can contact OSR by emailing regulation@statistics.gov.uk or via the OSR website

11. Assessment of quality

11.1 Relevance

The statistics provide users with monthly, quarterly and annual breakdowns of LA’s SoP performance for new HB claims and change of circumstances to existing HB claims.

11.2 Accuracy and reliability

The statistical bulletin provides clear and objective commentary on LA SoP performance. The statistics are illustrated using charts and maps which allow users to identify trends. Data are sourced from SHBE which undergoes cleansing and the statistics undergo rigorous quality assurance procedures (as detailed in section 5) to ensure statistical accuracy. A list of limitations are outlined in section 12.

11.3 Timeliness and punctuality

Statistics are released quarterly which allow sufficient time for production of the statistics. There is six-week lag in data when it is obtained from SHBE before production can begin. At the point of the release, the statistics are lagged by approximately four months.

11.4 Accessibility and clarity

The statistical bulletin and ODS tables are released in accessible format and meet the guidelines set out within DWP’s accessibility statement. Future improvements will be made to the ODS tables to further improvement accessibility. The statistics are described in an objective manner and use plain English to assist users.

11.5 Coherence and comparability

Where possible, HB SoP statistics are comparable to historical data due to the use of consistent methodology. Engagement with other internal and external statisticians around data source uses and impacts, enables consistency across DWP statistical products. The format of our statistics bulletins remains consistent across subject areas, where possible, to ensure ease of reading and comparability. Further information can be found in section 12.

11.6 Trade-off between output quality components

Future improvements could be made to reduce the six-week lag in data from when it is first obtained however the production process has rigorous quality checks and therefore a trade-off exists between timeliness and accuracy of the statistics.

11.7 Assessment of user needs and perceptions

There is a clear user engagement strategy in place to ensure statistics are improved through meeting increased user needs.

11.8 Confidentiality, transparency and security

Data used is subject to DWP’s strict security standards and any individual customer identifiable data is masked. Before statistical release, only a small number of authorised internal colleagues can access the statistics, in line with Code of Practice protocols and these cannot be shared wider.

12. Limitations of the statistics

12.1 Known issues

Users of these statistics should be aware of the following limitations of the statistics:

  • the quarterly statistics are produced at a point in time, they do not include later data retrospection
  • inconsistency across tables due to suppression or omission of data. Users are advised to pick the right set of tables for their specific needs and be aware of differences in aggregated totals. Monthly figures may not sum to quarterly totals, and quarterly figures may not sum to annual totals. This is because LAs may have suppressed or missing data at a monthly level that is not included in the reported quarterly totals. If data is supressed at quarterly level then it will always be supressed in the year-end annual table totals
  • though it is encouraged, it is not mandatory for LAs to supply a monthly extract of their HB claims. As a result, there are instances in which some LA level data will be unavailable within the statistics. Similarly, situations such as changes to an LA’s administrative systems or cyber-attacks may impact an LA’s supply of data, and can result in statisticians omitting LA level data due to concerns over data quality. Although missing data from individual LAs do not materially affect the main stories, there may be a slight impact on regional trends

12.2 Coherence

  • Coherence with other DWP publications that use HB data such as the HB caseload, HB Debt Recovery and Benefit Cap statistics to ensure that all publications are consistent. Where approaches are used in one publication, these are also used across other publications, where appropriate. For example, the handling of missing data, the formatting of charts and data. In some instances, it is deemed in the best interests of stakeholders and DWP statistical producers to take a different approach to other publications – this is always outlined and explained in the publication if this approach has been taken
  • Statistical release map ranges – LAs are allocated to one of five groups (bins) based on their average speed of processing for new claims and change of circumstances to an existing HB claim. Although we aim to allocate an equal proportion of LAs to each bin, where the range of average speed of processing (in calendar days) is small across all LAs, this becomes more difficult. When these situations arise, then the best nearest split of LAs are allocated across the bins. This is more evident with change of circumstances to an existing HB claim
  • Statistical release maps resolution – since the removal of interactive maps and them being replaced with JPEG images within the statistical release itself, it is difficult to see individual LA level data from the map images themselves without referring to the underlying data

13. Other available HB statistics

The DWP publishes related statistics on:

14. Contacts

cbm.stats@dwp.gov.uk (Statistical producers), Cross-Benefit and Migration Statistics, Client Statistics, Data as Statistics, DWP Digital

Kate Walker (Lead statistician), kate.walker@dwp.gov.uk, Cross-Benefit and Migration Statistics, Client Statistics, Data as Statistics, DWP Digital

15. Definitions and terminology

15.1 Processing times

When calculating processing times all days should be counted including weekends and public holidays, not just working days. This must include any days when there are reasons for additional processing days beyond the LA’s control, for example, a customer’s failure to provide additional information and delays between offices. Speed of processing has replaced the Right Time Indicator statistics.

15.2 Change of circumstances

A change event means any notification or information received during the course of a benefit award from the customer or third party which leads to a decision on a claim whether or not entitlement is affected.

15.3 Effective dates

If a change is notified in advance, that is before the date from which the change takes effect, the count for the processing time will start from the implementation date. If a decision is made on the change on or before the date the change takes effect, the time taken to process the change will be one day. However, if a decision is not made on the change by the date the change comes into effect, processing time will continue until the date a decision is made.

The following changes are excluded from the speed of processing information:

  • automatic uprating of the HB applicable amounts and social security benefits
  • revisions to earlier decisions, for example, following a reconsideration, appeal and accuracy/management check

15.4 Accessibility

The publication adheres to DWP accessibility standards. DWP wants as many people as possible to be able to use its statistical publications.

Back to top