Guidance

HBAP Module 6: Approach and Testing Strategy for the financial year ending March 2024 (applicable to England only)

Updated 1 April 2024

Applies to England

Reporting Deadline: 30 November 2024

Queries on this document should be emailed to lawelfare.hbassurance@dwp.gov.uk

Introduction and approach

1. The Housing Benefit Assurance Process (HBAP) is the testing and reporting requirements to be followed by Reporting Accountants undertaking engagement work for a local authority’s (LAs) claim for HB subsidy from the Department for Work and Pensions (DWP) on form MPF720A.

2. HBAP is a DWP reporting framework instruction prepared in accordance with the International Standard on Related Services (ISRS) 4400 (Revised) Engagements to perform agreed-upon-procedures regarding financial information. This module of HBAP specifies the testing strategy, error type definition, case testing results and reporting requirements sufficient for HBAP reports on claims for HB subsidy. The scope of the Reporting Accountant’s work is limited to the contents and requirements of HBAP when undertaking this work in respect of an LA’s claim for HB subsidy from DWP on form MPF720A. This document is available to both accountancy firms and LAs.

3. This module is designed to be used in conjunction with Modules 1 to 5 to provide the detail on testing methodology. This module explains the steps required to complete the HBAP report on form MPF720A.

Timetable

4. The Reporting Accountant must send the final HBAP report to the email address supplied by DWP and attach the final claim form MPF720A submitted by the LA no later than the published deadline.

5. Reporting Accountants are required to complete all modules of HBAP in time to ensure the Reporting Accountant’s HBAP report is issued to the LA for comment on its factual accuracy prior to submission of the final HBAP report and final claim to DWP by the published deadline.

6. If the LA has comments on the findings of the Reporting Accountants report, they may submit a separate representation to DWP including full details of their objections to the content.

7. HBAP Modules 1, 2, 3 and 6 are issued in February of each year to allow the work required to be spread over the period to the published deadline. Module 5 is provided for each software supplier no later than the end of June 2024 as the final software position is confirmed.

8. Modules 1, 2 and 6 are available at GOV.UK. Module 3 workbooks will be maintained and supported by DWP and will be distributed directly to the Reporting Accountant and the LA. Module 5 will be issued by DWP directly to Reporting Accountants.

Advice and support

9. Advice and support is available to Reporting Accountants and LAs. Interpretations of regulations and guidance will be provided by DWP. Reporting Accountants can email lawelfare.hbassurance@dwp.gov.uk and further guidance can be found at:

• Housing Benefit Assurance Process (HBAP)

• Housing Benefit Subsidy circulars

• Housing Benefit Adjudication circulars

• Urgent and LA Welfare Direct (LAWD) Bulletins

• Housing Benefit Subsidy Guidance Manual

• The Housing Benefit Regulations 2006 (as amended)

• Supported Housing Guidance

10. Where the Reporting Accountant and LA are in dispute regarding the level of testing to be completed, the nature of errors found or the treatment of specific cases, DWP should be contacted by email to lawelfare.hbassurance@dwp.gov.uk at the earliest opportunity.

11. DWP will then determine how the issue should be resolved for the purposes of HBAP testing and provide a direction to the LA and Reporting Accountant.

Year on year changes

12. The following changes are brought to the attention of Reporting Accountants for subsidy for the financial year ending (FYE) March 2024:

The Social Security (Habitual Residence and Past Presence) (Amendment) Regulations 2023

Circular HB A6/2023.

DWP introduced legislation which, with effect from 15 May 2023, inserts a category to the list of persons who are exempt from having to satisfy the Habitual Residence Test (HRT) for HB. The category covers those who have left Sudan in connection with the violence which rapidly escalated on 15 April 2023 in Khartoum and across Sudan, and who were residing in Sudan before 15 April 2023.

HRT assurance process for UK nationals returning from Niger (not Nigeria)

Refer to the HB Assurance team.

HRT assurance process for UK nationals of Chagossian descent and the claiming of HB.

Refer to the HB Assurance team.

The Social Security (Habitual Residence and Past Presence, and Capital Disregards) (Amendment) Regulations 2023

Circular HB A7/2023.

This circular provides guidance on the Social Security (Habitual Residence and Past Presence, and Capital Disregards) (Amendment) Regulations 2023 (Statutory Instrument (SI) 2023/1144) which came into force on 27 October 2023.

SI 2023/1144 makes amendments to the HB Regulations 2006 (SI 2006/213) and the HB (Persons who have attained the qualifying age for state pension credit) Regulations 2006 (SI 2006/214).

SI 2023/1144 amends the HRT for those coming from Israel, the Occupied Palestinian Territories (OPTs) and Lebanon and also introduces a new capital disregard for payments made under the Victims of Overseas Terrorism Compensation Scheme.

For the purposes of this guidance the OPTs are the West Bank, the Gaza Strip, East Jerusalem and the Golan Heights.

Benefit cap levels increase

LAWD1/2023 lite bulletin.

In April 2023, the government increased the benefit cap levels in line with inflation by 10.1%. The benefit cap levels rose from £23,000 to £25,323 for families in Greater London and from £20,000 to £22,020 for families nationally. The levels for single households without children rose from £15,410 to £16,967 in Greater London and from £13,400 to £14,753 nationally.

Further Household Support Fund extension

LAWD12/2022 bulletin

For those who require extra support, the government is providing an additional £1 billion, to help with the cost of household essentials, for the FYE March 2024. This is in addition to what has already been provided since October 2021 and brings the total funding for this support to £2.5 billion.

In England, the extension to the Household Support Fund, backed by £842 million, will run from 1 April 2023 until 31 March 2024. Devolved administrations will receive £158 million through the Barnett formula.

State Pension underpayments

DWP is currently identifying people who have been underpaid State Pension.

There are no changes to HB regulations to disregard the arrears payments of State Pension and these payments are capital. Entitlement to HB will be affected if the claimant already has capital between £10,000 and £16,000 where the arrears payment takes them above £16,000.

Claimants who are receiving the guarantee credit element of Pension Credit will not be affected by the arrears payments if, once Pension Credit has been reassessed following the receipt of that payment, they remain entitled to Pension Credit.

Crypto-assets and HB

Crypto-assets or, as they are also known, cryptocurrencies, are a type of digital currency which are generally held as investments by people who expect their value to rise.

More information about crypto-assets can be found in a fact sheet published by HM Revenue and Customs. For HB purposes, crypto-assets should be treated as capital in the same way as other types of investments, such as ISAs (individual savings accounts), personal pension schemes, premium bonds or stocks and shares.

Detailed guidance

Step 1 - Preparation

13. Guidance on preparation is based on the premise that a Reporting Accountant has been engaged by signed contract in which fees, Standardised Engagement Terms as per Module 1, variations have been agreed.

14. DWP recommends the Reporting Accountant and the LA should have a planning meeting in advance of the implementation of HBAP to set up the detailed testing arrangement. This meeting should, as a minimum, cover:

• a discussion that will agree the timetable for the work from the initial testing to the issue of the HBAP report, including a full explanation and evaluation by the LA of any Cumulative Assurance Knowledge and Experience (CAKE) issues (see Step 3). When testing is going to be conducted (initial, CAKE, additional testing or 40+ testing and what testing will be done by the LA or the Reporting Accountant

• agreeing a timetable for re-performance testing by the Reporting Accountant for any initial testing completed by the LA, CAKE, additional testing that will be completed by the LA

• finding out what the LA has done to address previous CAKE and additional testing issues included in the previous HBAP Report (obtained from the LA where there has been a change of firm between years). Any report made under Module X does not need to be provided, as Module X engagements are stand alone and are not intended to inform subsequent years testing

• the transfer of a copy of the previous year’s final MPF720A and information on what the LA has done to address the errors that led to any amendments, exceptions, or observations

• the confirmation of a key LA contact to deal with and clear all queries in time to meet the published deadline

• arranging to obtain audit trails from the LA’s systems and select samples, and to have access to the information derived from benefits system and the Customer Information System, including a demonstration of how to view the screens on both systems for testing purposes and an awareness of data protection and security issues. In undertaking HBAP work, Reporting Accountants should observe their obligations under the General Data Protection Regulation 2018 and their data protection policies

Step 2 - Initial testing - Module 3 workbook testing

15. The Reporting Accountant must select an initial random sample of 20 cases from the audit trail supporting the headline cell (cells 011, 055 and 094) of form MPF720A for each of the 3 benefit types (non-Housing Revenue Account rent rebate, rent rebate and rent allowance). A maximum total of 60 cases (20 cases for each of the 3 benefit types) are required where the LA awards all 3 benefit types. When the test population for a benefit type is less than 100, the Reporting Accountant will reduce the initial sample pro rata, for example a population of 49 would require a sample of 10.

16. If testing is started early using a source other than the final claim form, it is essential that the details tested are reconciled to the final claim form before the claim is reported upon. Any differences must be examined and the impact on the sample, test findings and the claim must be assessed before a HBAP report is issued (see Step 6).

17. Initial testing is required on both the headline cell entries and the analysis of expenditure in the detailed cell(s) that support the headline cell entry for each initial sample. The total of the detailed cell entries should equal the headline cell expenditure (excluding any prior year overpayments).

18. An initial case is defined by its case reference number. All benefit transactions in the year against the case reference number must be tested. If a claimant has more than one case reference number in the year; for example, where the claimant returns to benefit and the LA uses a new claim reference number, only the expenditure against the single case reference number selected for testing needs to be examined.

19. If the LA changes HB software or supplier which results in a change in reference number for the sample case within the subsidy year, the Reporting Accountant will need to test all sample case transactions from the legacy system and new system so that all benefit transactions in the year made to the claimant are tested.

20. The detailed testing required is set out in the HBAP Module 3 workbooks. The testing must also comply with Appendix 3 in HBAP Module 1. The workbooks require the Reporting Accountant or LA to evidence details of the sample cases and testing results, for example, the relevant dates, the claimant’s entitlement to benefit and the subsidy coding of the payments made. The specific evidence that the Reporting Accountant or the LA must verify is listed in the tests attached to the workbooks. The workbooks record whether each case passes or fails each test. The outcome of this testing and CAKE (Step 3) must be evaluated in accordance with the guidance in Step 4.

21. Where the LA undertakes the initial testing, the Reporting Accountant will carry out re-performance testing on the LAs work. Reporting Accountants will re-perform all failed and a random 10% of the passes.

22. This will be carried out according to a timetable that allows the Reporting Accountant to review the work and findings completed by the LA, in time to enable any additional testing to be completed in order to meet the reporting deadline. If the re-performance testing identifies errors in the LA’s testing the Reporting Accountant will point out the incorrect conclusion reached by the LA and the LA is instructed to review all cases again in light of this issue. The Reporting Accountant will then pick a further sample for re-performance testing including some with the characteristics of the incorrect conclusion. If this second, LA attempt is tested without error then the Reporting Accountant will rely on the results. If there are further errors the Reporting Accountant will move to simply disclaim the LA testing and the Reporting Accountant will re-perform the full sample of 20.

Step 3 - CAKE from prior year Module 3

23. Module 1 Appendix 2 requires the LA to provide the Reporting Accountant with the HBAP report (including any supplementary reports, but excluding any report produced under Module X) from the previous subsidy year where there has been a change of accountancy firm between the previous and current subsidy years. The Section 151 Officer is required to provide the Reporting Accountant with the prior year final tested subsidy claim form (MPF720A) with any amendments made to the form initialled by the Section 151 Officer.

24. The Reporting Accountant is required to refer to the prior year final tested subsidy claim form and HBAP report (including any supplementary reports, but excluding any report produced under Module X) to identify issues that were reported which resulted in exception, observation or amendment of the claim. The Reporting Accountant will ask the LA to confirm what it has done to address previous CAKE and additional testing issues included in the previous HBAP Report.

25. CAKE testing can be generated from an observation, amendments or exception from the previous year HBAP report in respect of the previous year’s subsidy claim.

26. The purpose of CAKE is to establish whether these issues remain unresolved and test for continuing errors in the appropriate population. Where there are CAKE issues from prior year Module 3 workbook testing, the Reporting Accountant will need to carry out additional testing (see page 15, step 4.2).

27. Any additional errors identified during CAKE testing, which share the same prior year error characteristics or nature must be considered for extrapolation and inclusion in the HBAP report.

28. Unrelated errors identified during CAKE testing must be corrected by the LA and, if appropriate, the subsidy claim should be amended accordingly. Unrelated errors should be highlighted as an observation within the HBAP report but not included in any current extrapolation.

29. If the additional CAKE testing does not identify any errors and no errors were identified in the initial sample, the results must be reported in the HBAP report as an observation. DWP’s position is that the absence of any error provides sufficient assurance that the issue has been resolved. In these circumstances, the issue need not be considered under the CAKE arrangements in the following subsidy period.

Step 4 - Evaluating the results of initial testing

(Completion of Modules 2, 3, and 5)

30. Step 4 must be followed in all cases. Errors identified by testing cannot be corrected on the current form without addressing the likelihood of similar errors in the population; and an extrapolation cannot be avoided because an adjustment has been or will be made in the following year for the errors found.

31. The Reporting Accountant will complete Modules 1, 2 and 5 as well as the initial testing. Errors discovered from modules will be evaluated in the same way as an error found in initial testing.

32. Where a Reporting Accountant identifies during testing that a claim had been incorrectly assessed by an LA, but that the LA’s own quality assurance or checking procedures (including any ‘one off’ checks) has identified the error and corrective action has been undertaken by the LA prior to the Reporting Accountant requesting case population lists from which to select the current year samples, the individual case should be treated as correct, except where errors are still present in the case when it is checked by the Reporting Accountant. In these circumstances, where the case is correct at the point the Reporting Accountant or LA carry out a test under these procedures, the previous error does not need to be reported in the HBAP report, and no further action is required by the Reporting Accountant. Where errors are still present in the case, these should be reported in the HBAP report as per the HBAP procedures. This paragraph does not apply to errors identified following the Reporting Accountants request for the case population lists relevant to the current year samples in other words corrected as part of the HBAP process, where the relevant HBAP procedures will continue to apply.

33. In every case, the Reporting Accountant must critically evaluate the reason for an error and the likelihood of similar errors in the population. If it cannot be established that an error is isolated see page 14 step 4.1, or that all errors have been corrected in the year under review, the error needs to be subjected to additional testing and the results extrapolated for DWP to consider the impact on subsidy.

34. The fact that an error identified by testing is corrected by an agreed amendment on the subsidy claim form only deals with the individual error, not the likelihood of similar errors in the rest of the caseload.

35. If the LA is able to identify a whole population or sub-population relating to the error found and wishes to amend form MPF720A in relation to that error, the instructions in Step 5 of Module 6 must be followed.

36. If the LA wishes to test a sample larger than 40 cases but without testing the entire population or sub-population, this will be subject to a separate engagement, and the LA should approach DWP to agree the level of testing and re-performance testing required, as per Module X.

37. Where the LA has undertaken to amend all cases found to contain the error in the following subsidy year this will need to be reported in the current HBAP report.

38. Where initial testing identifies errors, the Reporting Accountant must examine the reasons for these errors to determine what action and additional testing is necessary to allow the Reporting Accountant to report as required by HBAP.

39. The Reporting Accountant must conclude on the type of error (see page 19 Step 5.2) to assess whether they represent an error that will require additional or 40+ testing; is the error quantifiable; or an isolated error?

40. An example of this evaluation would be where the initial sample identified one claim where earned income had been incorrectly calculated resulting in the claimant being underpaid. Despite the error being an underpayment, this error could result in an overpayment. There is unlikely to be enough data in the results of initial testing to provide the Reporting Accountant with enough information. Therefore, the Reporting Accountant should conduct additional testing to evaluate the error.

41. Having found an error, additional testing will be required unless:

• the additional testing approach is not required for errors which by their nature do not affect the amount of subsidy claimed. For example, the misclassification of HB cases between cells of the same subsidy value, with the same method of entitlement calculation, which produce the same amount of qualifying expenditure

• it relates to errors that would always result in underpaid benefit because such errors have no impact on the subsidy claimed (see error type 1). In this case the initial sample results only need be reported under the observations heading in the HBAP report. However, all overpayments are excluded from this provision; this is because of their complexity. If errors in overpayments are identified they must be subjected to additional testing

• is this an isolated error? In most cases this is unlikely as most errors tend not to be so singular that they can be classified as isolated and could never happen again

42. It may be possible to conclude immediately that errors in the initial sample are representative of the population and to agree an amendment to the claim to be made by the LA. This will be possible, for example, where system parameters are wrong, and the records can be interrogated to quantify the impact on subsidy claimed.

43. It may be possible to conclude an error is an isolated error or, through drilling down, relates to a small population or sub-population. These outcomes are covered in later sections of this Module. If such classification is not possible additional testing will be required on the headline (011, 055, 094) or detailed cell contents to help determine the significance of errors found in the initial sample.

44. It is important that the timetable for initial testing allows any additional work to be completed and any amendments or report results concluded in time to meet the published deadline. Reporting Accountants are encouraged, where possible, to carry out initial work earlier and to focus on high-risk areas and known problems first. Reporting Accountants cannot rely on the LA obtaining an extension to the reporting deadline. Extensions to statutory deadlines are granted only in exceptional circumstances. It is not acceptable to issue a holding HBAP report because work is not complete.

Step 4.1 - Isolated errors

45. Given the nature of the populations in benefit claims, ‘isolated errors’ are extremely rare. The presumption should be that errors are not isolated and that without a clear compelling case they should not be treated as such. Additional testing will usually be required.

46. If the Reporting Accountant concludes that an isolated error has been identified, the Reporting Accountant can then apply the LA amendment procedure outlined in Step 5. If an amendment is agreed, Reporting Accountants also need to consider the effect of amending one cell on other cells: other cells may also need to be amended (see Step 5 below).

47. Where it cannot be shown that the error can only have happened in this one instance only then additional testing will be applied.

Step 4.2 - Additional testing

48. Having found errors from the initial testing, additional testing is then carried out in the following steps. The additional testing will focus on the particular characteristics of the error only. These particular characteristics may relate to an expenditure type for example earnings (In this case you would select from the appropriate headline cell for that expenditure type) or to a specific detailed cell.

49. For additional testing, the HBAP Module 3 workbooks should be used, however the testing and workbook completion should focus on the nature and characteristics of the error being tested. The workbooks record whether each case passes or fails each test.

50. A case for additional testing is defined by its case reference number. All benefit transactions relevant to the error in the year against the case reference number must be tested.

51. Where testing identifies errors unrelated to the original error being tested do not undertake any further additional testing in respect of this new error. This should be reported in the HBAP report as an observation.

Step 4.2.1 - Small populations

52. Small populations are defined by DWP as 100 cases. Where test findings show there are errors affecting cells with small populations and the Reporting Accountant does not have a sufficient understanding of the cell population to agree an LA amendment, or the LA does not agree an amendment, the whole population must be tested by the LA to allow an agreed amendment to be made.

53. When considering the level of CAKE testing applicable in circumstances where the prior year CAKE error falls in a sub population under 100 but over 40, the Reporting Accountant and the LA can agree to test a minimum of 40 CAKE tests, disregarding the current small population guidance. If the LA wishes to test the full population, then this should be agreed with the Reporting Accountant at the initial planning meeting.

54. If the Reporting Accountant and the LA agree to apply the small population methodology and the LA completes the full population testing, the expected level of re-performance carried out by the Reporting Accountants should be a random 10% of the population. If the Reporting Accountant and the LA agree to apply standard CAKE testing and the LA completes the 40 case tests, then the expected level of re-performance carried out by the Reporting Accountants should be in line with current expectations, in other words, test all fails and 10% of the remaining sample. This will be carried out according to a timetable that allows the Reporting Accountant to review the further work and findings completed by the LA, including an element of re-performance testing if the LA has completed the work, in time to meet the reporting deadline.

55. When testing the full population and the results of the re-performance testing do not identify any errors in the LA’s testing, an agreed amendment must be made. If the Reporting Accountant agrees an amendment the effect of amending one cell on other cells must also be considered: other cells may also need to be amended (see Step 5).

56. If the re-performance testing identifies errors in the LA’s testing the Reporting Accountant will point out the incorrect conclusion reached by the LA and the LA is instructed to review all cases again in light of this issue. The Reporting Accountant will then pick a further sample for re-performance testing including some with the characteristics of the incorrect conclusion. If this second, LA attempt is tested without error then the Reporting Accountant will rely on the results. If there are further errors the Reporting Accountant will move to simply disclaim the LA testing and the Reporting Accountant will complete a test of 40 cases themselves and extrapolate from these results. An amendment must not be made and the issue and test findings with extrapolation must be included in the HBAP report, including any effect on other cells on the form.

Step 4.2.2 - Drilling down

57. For a population larger than 100 cases, where initial test findings indicate errors relating to a defined sub-population, the Reporting Accountant will undertake drill down testing, identifying the characteristics of the error taking their sample from a sub-population that shares these characteristics and focusing additional testing on the sub-population in question. Where the sub-population is fewer than 100, then test as per small populations (Step 4.2.1).

58. Testing samples will be 40 cases. This should allow the Reporting Accountant to quantify the error in the sub-population and calculate the extrapolation, based on this sub-population.

59. The purpose of the drill down testing is to identify the extent or significance of initial testing results within the sub-population and should therefore:

• concentrate only on the characteristics or nature of the error found. A number of tests may be applied to the initial sample, but further application is needed only of those tests which identified errors. An example of this could be where an application form is unsigned. The Reporting Accountant would not re-perform all the validity checks required by the workbook on a further sample, but instead would select an additional sample at random to see whether those forms were signed

• be biased to identify the significance of the initial testing results, hence a biased sample from the total population will need to be taken. An example of a biased sample would be where errors have been made where a problem only occurs for a particular case type. Here the additional sample would be selected from cases containing the particular case type, in other words, the sub-population

60. Drilling down is a powerful way of determining the potential impact and extent of errors within a sub-population. To drill down, the LA must be able to provide the Reporting Accountant with an audit trail that lists all cases and total values of the sub-population. Reporting Accountants must determine that only the defined sub-population detected in the initial sample contains that error and the rest of the total population does not contain that error. In these circumstances, Reporting Accountants must ensure that any initial sample cases which are not part of the defined sub-population are excluded from the sample value used to calculate an error rate and extrapolated cell adjustments (See step 5.4).

61. The Reporting Accountant should report the results of the drilled down testing.

62. The LA should carry out the work. Reporting Accountants will re-perform all failed and a random 10% of the passes. This will be carried out according to a timetable that allows the Reporting Accountant to review the further work and findings completed by the LA, in time to meet the reporting deadline. If the re-performance testing identifies errors in the LA’s testing the Reporting Accountant will point out the incorrect conclusion reached by the LA and the LA is instructed to review all cases again in light of this issue. The Reporting Accountant then pick a further sample for re-performance testing including some with the characteristics of the incorrect conclusion. If this second, LA attempt is tested without error then the Reporting Accountant will rely on the results. If there are further errors the Reporting Accountant will move to simply disclaim the LA testing and the Reporting Accountant will re-perform the full sample of 40.

Step 4.2.3 - 40+ approach

63. Where the LA is unable to isolate a sub-population or the LA is unable to establish the total value of the subpopulation, the approach determined by DWP is to select and test an additional random sample of 40 cases from the headline cell. The Reporting Accountant should concentrate only on the characteristics or nature of the error found. The 40+ approach can be undertaken by the Reporting Accountant or the LA, as agreed in their contract, preliminary meeting or at the point the need for 40+ testing is identified.

64. The sample will be a random selection of 40 cases from the headline cell.

65. The LA should carry out the work. Reporting Accountant will re-perform all failed and a random 10% of the passes. This will be carried out according to a timetable that allows the Reporting Accountant to review the further work and findings completed by the LA, in time to meet the reporting deadline. If the re-performance testing identifies errors in the LA’s testing the Reporting Accountant will point out the incorrect conclusion reached by the LA and the LA is instructed to review all cases again in light of this issue. The Reporting Accountant then pick a further sample for re-performance testing including some with the characteristics of the incorrect conclusion. If this second, LA attempt is tested without error then the Reporting Accountant will rely on the results. If there are further errors the Reporting Accountant will move to simply disclaim the LA testing and the Reporting Accountant will re-perform the full sample of 40.

66. The Reporting Accountant should report the results of the testing.

67. Where the results of additional testing do not allow agreed amendments to be made, the results of both the initial sample and any additional or 40+ testing should be set out as an exception in the HBAP report, including any effect on other cells on the form (see page 35 Step 6). An example HBAP report to show how testing results should be reported will be made available to Reporting Accountants and LAs through the DWP/LA interface system ‘Glasscubes’.

Step 5 - Agreed amendments and extrapolations

Step 5.1 - General points on amendments and extrapolations

68. The populations and the nature and values of the errors detected in HBAP testing lead inevitably to a degree of uncertainty when it comes to extrapolating findings. The Reporting Accountant’s extrapolations are based on the value of any errors identified. Where the number and nature of errors are so complicated that the Reporting Accountant cannot determine an amendment or extrapolation, exceptions should be reported as identified, together with an explanatory statement (see page 35 Step 6).

69. In cases where testing results in an amendment to a cell or an extrapolated figure is included in a HBAP report, the effect on other cells on the form must be considered and corresponding amendments made, or extrapolations provided. If corresponding amendments or extrapolations cannot be identified, the report must explain why the effect on other cells cannot be determined. Note, amendments and reported extrapolations need to satisfy the in-year reconciliation check in each section of the form: amending or reporting one cell without dealing with the impact on other cells is likely to cause the in-year reconciliation to fail and this will result in DWP returning the claim for corrective work to be done. All FYE March 2021 expenditure appears in the headline cell and in the detail cells, providing the analysis supporting it. Only prior year overpayments do not appear in either the headline cell or the in-year reconciliation cells.

70. Where testing identifies that benefit has been underpaid, it is important to note that there is no impact on subsidy because subsidy cannot be claimed on benefit that has not been awarded. Where underpaid benefit is subsequently awarded it can be included in a later subsidy claim, but in the meantime, it must be excluded from the calculation of cell amendments. Underpaid benefit should be referred to under the observations heading in the HBAP report, but its effect should be excluded from any cell amendments or extrapolations. Where the nature of the error could equally result in an overpayment, for example, an error calculating claimant income could result in an over or under payment, additional testing is required to try to establish if overpayments have been made. For HB Assurance purposes, circumstances could arise where both under and overpaid benefit has been identified, in these circumstances the overpayment element should not be netted off against any relevant underpayment.

71. Where testing identifies that expenditure has been correctly awarded but incorrectly classified and some errors have the effect of increasing the amount that should be included in a cell and others the effect of decreasing the amount in that cell, the sample error is the net figure, that is, the sum of the increases and decreases. Where testing identifies that benefit itself has been underpaid this should be treated in accordance with the error type 1, see below.

Step 5.2 - Error types for subsidy purposes

72. To calculate a cell amendment or an extrapolation for a HBAP report the Reporting Accountant needs first to determine the error type, as DWP requires the following treatment for subsidy:

Error type 1 - underpaid benefit

Where benefit has been underpaid, there is no impact on the amount of subsidy to be claimed because subsidy cannot be claimed on benefit that has not been awarded. Underpaid benefit must, therefore, be excluded from the calculation of a cell amendment or extrapolation in a HBAP report. The underpaid amount should be excluded from the sample error but the value of the case excluding the underpayment tested should be included in the sample value. The fact that there are cases where benefit has been underpaid should be included in the HBAP report in the observations section (see example report at Appendix 1).

Note: Error Type 1 applies where the error identified will always result in an underpayment of benefit: additional testing in this case is not required. Where the nature of the error could equally result in an overpayment (for example, an error calculating claimant income could result in an over or under payment), additional testing is required to try to establish if overpayments have been made. Where overpayments are identified, or additional testing does not confirm that the population is overpayment free, these should be detailed in the HBAP report (see example report at Appendix 1).

Error type 2 - missing claim forms

Subject to the application of a local risk based verification (RBV) policy, where the claim form (paper, document image or electronic format) to support the current claim is missing and the LA is unable to produce other evidence that a claim form was received, for example, letters or other records of communication from the claimant about changes in circumstances; copies of information supporting the levels of income, capital or rent; responses from the claimant to enquiries from the LA; copies of records of reviews of the claim; or information from DWP where the benefit is passported (see bulletin HB G8/2015) the case must be excluded in its entirety from the subsidy claim form. The cell adjustment or extrapolation in the HBAP report must remove all expenditure on the case from the subsidy claim form, that is, from the headline cell and detail cell(s).

Note: The checks on the assessment of benefit entitlement are unaffected by the above. The LA must hold supporting information to demonstrate that benefit was correctly awarded and calculated. If this information is missing or insufficient Error Type 3 (insufficient supporting information) applies. Local (RBV policy: a local RBV policy may not require a claim form. In such cases Reporting Accountants need to confirm the LA has complied with its RBV policy for the year in which the claim was determined. If the LA has not complied with its RBV policy, see page 21 error type 6.

Error type 3 – benefit overpaid or insufficient supporting information

Subject to the application of a local RBV), where benefit has been overpaid or supporting information for part of the claim is missing or insufficient, the headline cell is not adjusted but the cell adjustment or extrapolation in a report must remove the expenditure from the detail cell(s) in which the expenditure is reported and increase the appropriate overpayment cell(s) by a corresponding amount (for example, LA error overpayments will need to be increased where these errors are caused by an act or an omission by the LA).

Note: Where a local RBV policy operates, Reporting Accountants need to confirm the LA has complied with its RBV policy. Where the application of the RBV policy does not enable the accountant to establish the subsidy classification, the case is admitted for subsidy as long as the LA has applied its approved RBV policy. The assurance engagement covers the application of the RBV policy, not its contents or adequacy. If the LA has not complied with its RBV policy, see error type 6 on the next page.

Error type 4 – expenditure misclassification

Where benefit expenditure has been misclassified, for example, misclassifying LA error overpayments as eligible overpayments, the headline cell is not adjusted but the cell adjustment or extrapolation in a HBAP report must remove the expenditure from the detail cell(s) in which the expenditure has been incorrectly reported and increase the cell(s) in which it should have been reported by a corresponding amount. Errors which by their nature do not affect the amount of subsidy claimed, for example, the misclassification of HB cases between cells of the same subsidy value, with the same method of entitlement calculation, which produce the same amount of qualifying expenditure, do not require additional 40+ testing. The initial sample results only need to be reported under the observations heading in the HBAP report (see example report at Appendix 1). All overpayments are excluded from this 40+ exemption, because of the complicating factors of LA error and administrative delay subsidy, and the subsidy entitlement conditions which apply to prior year overpayments.

Error type 5 – impact cannot be determined

Where the consequential impact of the error on other cells cannot be determined or the testing identifies a combination of one or more of the above error types and these cannot be separated to enable an amendment or extrapolation for the consequential effect on other cells to be reported, the error(s) should be reported against the cell in which it was identified. The HBAP report should state the reason why the corresponding entries could not be established.

Error type 6 – failure to apply the RBV policy

Where the LA has not applied its RBV policy (failed to apply the specified check) or performed a check that does not meet the requirements of the RBV policy (for example, the policy states eligible rent will be verified in all cases but the LA has simply agreed the rent to the direct debit on the claimant’s bank statement, this has not established the eligible rent: the rent paid could include ineligible service charges) or cannot evidence that the check was performed (there is no record or prime documents to demonstrate the RBV policy check was performed), the headline cell is not adjusted but the cell adjustment or extrapolation in a HBAP report must remove the expenditure from the detail cell(s) in which the expenditure is reported and increase the LA error overpayment cell by a corresponding amount.

73. An example HBAP report (see Appendix 1) contains examples of the above error types and how these should be included.

Step 5.3 - Amendments to Subsidy Claims

74. The results of initial and any additional testing may allow an amendment to be made because:

• it is an isolated error
• the whole population of the cell has been tested, or
• the whole sub-population affected has been tested

75. Where the whole population or sub-population has been tested by the LA, and this does not fall under the criteria for dealing with small populations (100 cases or fewer), the Reporting Accountant should undertake re-performance testing. This will be carried out on a random sample of the population tested, and the sample size will be the greater of either 10 cases or 1% of the population or sub population.

76. Where the LA plans to carry out testing in order to calculate an amendment, they should notify the Reporting Accountant prior to commencing testing so that the Reporting Accountant can plan and adequately resource the re-performance testing.

77. If an amendment is made, consideration must also be given to the effects of amending one cell on other cells as these may also need to be amended. If the results do not allow an amendment to be made, the issue and test findings need to be included in the HBAP report, including any effect on other cells on the form.

78. All amendments must be made by the LA to a pdf version of the form (not in manuscript) because this will ensure the amendments are automatically reflected in calculations and amounts transferred between cells. The LA should make amendments to an electronic copy of the form, perform the validation check, and print a copy of the amended form. The LA should, where practically possible initial all input cells which have been amended, where a physical initial is not possible. Reporting Accountants should ensure that full details are provided Appendix C of the HBAP report: it is not necessary to initial changes to cells that contain calculations or cells that contain amounts transferred from other cells (these are updated automatically by the form). The LA should sign the LA certificate of the amended form. The Reporting Accountant should check that all amendments to input cells have been made and, where practical, initialled by the LA and check that no other input cells have been amended.

79. Amendments made to the subsidy claim form must be listed in the HBAP report (Appendix C) for LA agreement and CAKE purposes in the following year. This will list the cell amended, the amount of change and a brief explanation of the reason for the amendment.

80. For example, because it does not allow a claim where zero materiality applies, the variation in errors is such that the sample may not be representative of the population, or because the order in which amendments are made might affect total subsidy, Reporting Accountants should report why amendment is not possible (see page 25 step 6).

Step 5.4 - Extrapolations

81. The Reporting Accountant should report the results of the testing.

82. Where the results of additional testing do not allow agreed amendments to be made, the results of both the initial sample and any additional testing should be set out as an exception in the HBAP report, including any effect on other cells on the form (see page 25 Step 6).

83. If additional testing is selected from a specific detail cell due to the nature of the error(s) identified in the initial sample, the Reporting Accountant must ensure that any initial sample cases which do not reflect characteristics of the error being tested are excluded from the sample value used to calculate an error rate and extrapolated cell adjustments.

84. An example HBAP report to show how testing results should be reported is provided to LAs and Reporting Accountants via the DWP/LA interface system ‘Glasscubes’.

85. Where an amendment cannot be agreed and additional testing has been completed the results will then be used in the following extrapolations.

86. A cell extrapolation in a report should be calculated as follows:

• Cell extrapolation CA = (CT or SP) x SE/SV where:
  • CT = Cell Total that is the value in £s of the cell from which the sample is drawn
  • SE = Sample Error that is the value in £s of the errors found in the sample. Exclude cases where benefit has been underpaid (error type 1).
  • SP = Sub-population total that is the value in £s of the sub-population from which the additional sample is drawn
  • SV = Sample Value that is the value in £s of the sample tested (where benefit has been underpaid see page 19 step 5.2 and error type 1)
  • SE/SV is referred to in the HBAP report as the percentage error rate, which should be calculated to 2 decimal places. The Reporting Accountant should extend the number of decimal places if this results in a nil error rate and include the number of decimal places used in the HBAP Report

Note: SV is the value of the transactions tested that is the whole year value of the case(s) tested.

Example

In a sample where the transactions tested have a value of £1,000 (SV), you find errors whose net value is a reduction of £150 (SE) (£150 is overpaid expenditure).

The cell total (CT) is £500,000. The headline cell £10,000,000

The percentage error rate is 15% (£150 divided by £1,000 x 100)

The Cell Adjustment is £75,000 (£500,000 times 15% (£150/£1,000))

The amended cell total is £425,000 (£500,000 minus £75,000)

There is no impact on the headline cell. The LA error overpayment cell would increase by £75,000.

Note: Reporting Accountants must consider and set out the effects of amending or extrapolating one cell on the form on the other cells on the form. Further examples are provided in in the example copy of the HBAP report.

87. In principle, this calculation should be simple to apply. However, care needs to be taken where extending samples, in particular where drilling down into sub-populations. In such cases, the Reporting Accountant must only include cases that were drawn from the relevant sub-population and only apply the error rate to the sub-population (for example, any cases in the initial random sample which are not from the relevant sub-population should be excluded).

Example

Taking the same example as above but where the errors relate to a sub-population of £300,000 in the cell total of £500,000:

The percentage error rate is 15% (£150/£1,000 x 100);

The cell adjustment is £45,000 (£300,000 times 15% (£150/£1,000)); The amended cell total is £455,000.

There is no impact on the headline cell. The LA error overpayment cell would increase by £45,000.

Note: Where a sub-population is identified, the extrapolation should use SP the value of the sub-population within the cell, not CT, the cell total that includes the sub-population.

88. Extrapolations for HBAP reports should be determined by reference to the total benefit tested, in other words, the total benefit granted on the case(s) in the year. For example:

• from the initial sample (the total value: £3,280) and the additional sample (the total value: £7,100), 1 error is identified

• a claimant is awarded benefit for 2 benefit periods in the year using the same case reference number

• the claimant is awarded £100 per week for the first 4 weeks of the year (£400) before their benefit entitlement ends

• the claimant subsequently returns to benefit and is awarded £125 per week for the last 2 weeks of the year (£250)

• testing identifies a LA error overpayment of £55 per week for the last 2 weeks of the year (£110)

• the sample value is £10,380 (£3280 + £7100) – this value includes the benefit of £650 granted to the claimant (£400+£250)

• the sample error is £110

• the percentage error rate to be applied is 1.06% (£110/£10,380 * 100) for that particular claim

  • The overpayment in this case does not affect the headline cell (because the headline cell includes current year overpayments)
  • The percentage error rate is used to calculate a cell adjustment(s) for the detail cell(s) in which expenditure was included (a reduction) and therefore a corresponding amendment (an increase) to the LA error overpayment cell

89. The results of testing on each case should be combined with the results for all other cases tested in the sample and presented in an extrapolation table with the extrapolated effects on other cells as appropriate - see example HBAP report at Appendix 1.

90. Where an error is discovered that if acted upon would create a prior year overpayment or amends an amount already reported as a prior year overpayment the extrapolation calculation should only include the error based on the current year HB expenditure not expenditure which was included in previous subsidy claims.

Step 6 - The Reporting Accountant’s HBAP report

91. Before the HBAP report is started Steps 4 and 5 must be completed. The results of both the initial sample and additional or 40+ testing should be set out in the HBAP report.

92. In cases where testing results in an extrapolated figure included in the HBAP report, the corresponding effect on other cells on the form needs to be included. If corresponding adjustments cannot be identified, the HBAP report must explain why the effect on other cells cannot be determined. Note, extrapolations need to satisfy the in-year reconciliation check in each section of the form: qualifying one cell without dealing with the impact on other cells is likely to cause the in-year reconciliation to fail and this will result in DWP returning the claim for further work to be done (all 2020-21 expenditure appears in the headline cell and in the detail cells providing the analysis supporting it. Only prior year overpayments do not appear in either the headline cell or the in-year reconciliation cells).

93. The standard HBAP structure must be used for each HBAP report: an illustrative example with the prescribed content and manner of presentation is provided within the example HBAP report (available via ‘Glasscubes’ and also contained within the Module 3 help files). The HBAP report should leave DWP in no doubt about the details of the exceptions identified and, to this end, findings should be presented clearly and unequivocally. The HBAP report needs to include:

• test findings
• information on the range of error values to demonstrate to DWP why an adjustment could not be determined for amendment, subsidy implications
• any implications for the entries in other cells; and indicate whether similar issues have been included in a previous qualification letter

94. DWP will use this information to decide what further action is required. For example, it may decide to settle the claim based on the test findings or may need the LA to do more work, particularly if the problem has continued for some years. The key point is that the initial, CAKE and additional or 40+ testing provides a basis for DWP to make such decisions and achieve settlement of claims without further recourse to Reporting Accountants.

95. If the Reporting Accountant’s report is inconsistent with the requirements of HBAP, for example, if a cell adjustment or extrapolation is calculated but consideration is not given to the impact on other cells on the form, or if a 40+ sample is drawn from a sub-population but non sub-population cases in the initial sample are used in the calculation of an error rate, DWP will require re-submission of the associated parts of the report in a form consistent with HBAP requirements.

96. Where testing identifies that benefit has been underpaid, it is important to note that there is no impact on subsidy because subsidy cannot be claimed on benefit that has not been awarded. Where underpaid benefit is subsequently awarded it can be included in a later subsidy claim, but it must be excluded from the calculation of cell adjustments and extrapolations in HBAP reports. The HBAP report should observe that cases have been identified where benefit has been underpaid but that these cases are not reflected in extrapolations because subsidy is only payable on benefit that has been awarded. Note, where the nature of error could equally result in an overpayment, for example, an error calculating claimant income could result in an over or under payment, additional testing is required to try to establish if overpayments have been made.

97. The detailed information in HBAP reports should be tailored to the particular circumstances of test findings and must set out the effects of the exception on all affected cells in the form. Extrapolations for a HBAP report should be completed in accordance with paragraphs in Step 5.4. Reporting Accountants need to report the corresponding adjustments in the extrapolations. However, the requirement to set out the effects does not extend to cells containing calculations or where the form transfers amounts between sections of the form. It does not cover, for example, the effect on overall LA error subsidy of an increase in LA error overpayments in one section of the form.

Appendix 1: HBAP report

A non-pdf version of this HBAP report wording can be obtained from DWP by request to lawelfare.hbassurance@dwp.gov.uk

Housing Benefit (Subsidy)

Assurance Process [year]

Module 6 DWP Reporting Framework Instruction

(Applicable to England only)

Reporting Accountants’ report for the Housing Benefit Subsidy claim form MPF720A, year ended 31 March [year]

By email to lawelfare.lapaymentsandsubsidy@dwp.gov.uk Please ensure that the subject of the email contains [LA name] Financial Year Ending March [year] HBAP Report.

And: The Section 151 Officer of [full name of local authority………………………], Section 151 Officer.

This report is produced in accordance with the terms of our engagement letter with the [full name of local authority] dated…………………… and the Standardised Engagement Terms in Appendix 2 of HBAP Module 1 [year] issued by the Department for Work and Pensions (DWP) for the purpose of reporting to the Section 151 Officer of [insert full name of local authority………………………………………] (the ‘local authority’) and the DWP.

Our report is prepared solely for the confidential use of the local authority] and the DWP and solely for the purpose of facilitating the claim for Housing Benefit subsidy on form MPF720A dated [insert date of initial final claim].

This report should not be copied, referred to or disclosed, in whole (save as otherwise permitted by the Standardised Engagement Terms) or in part, without our prior written consent. Without assuming or accepting any responsibility or liability in respect of this report to any party other than the local authority and DWP, we acknowledge that the local authority and/or DWP may be required to disclose this report to parties demonstrating a statutory right to see it.

This report is designed to meet the agreed requirements of local authority and DWP as described in the DWP HBAP reporting framework instruction [year] [and supplemented by…………………… [Specify any additional agreed instructions and agreed deviations, e.g. written confirmation from the local authority or DWP dated………………………].

This report should not therefore be regarded as suitable to be used or relied by any other party for any purpose or in any context. Any party other than [the local authority] and DWP which obtains access to this report or a copy and chooses to rely on this report (or any part of it) will do so entirely at its own risk. To the fullest extent permitted by law, we accept no responsibility or liability in respect of our work or this report to any other party and shall not be liable for any loss, damage or expense of whatsoever nature which is caused by the reliance of anyone other than the addressees on our work or this report.

Respective responsibilities of the local authority and the Reporting Accountant

We conducted our engagement in accordance with HBAP Modules 1 and 6 [year] issued by DWP, which highlight the terms under which DWP has agreed to engage with Reporting Accountants.

As Section 151 Officer of the local authority, you have responsibilities under the Income-related Benefits (Subsidy to Authorities) Order 1998. You are also responsible for ensuring that the local authority maintains accounting records which disclose with reasonable accuracy, at any time, the financial position of the local authority. It is also the Section 151 Officer’s responsibility to extract relevant financial information from the local authority’s accounting records, obtain relevant information held by any officer of the local authority and complete the attached form MPF720A in accordance with the relevant framework set out by DWP [state titles and dates of relevant documents].

Our approach

For the purpose of the HBAP engagement we have been provided with a signed (Authorised as per Module 1 para 23) copy of form MPF720A FYE March 2024 dated [insert date of initial final claim] by the Section 151 Officer (or Authorised deputy).

The Section 151 Officer (or Authorised deputy) remains solely responsible for the completion of the MPF720A and is the signatory on the local authority’s certificate on claim form MPF720A.

Our engagement was carried out in accordance with the DWP reporting framework instruction which has been prepared in accordance with the International Standard on Related (ISRS) 4400 (Revised), Engagement to perform agreed-upon-procedures regarding financial information. The purpose of the engagement is to perform the specific test requirements determined by DWP on the defined sample basis as set out in HBAP Modules of the HBAP reporting framework instruction on the local authority’s form MPF720A dated [insert date of initial final claim], and to report the results of those procedures to the Local Authority and DWP.

The results of these are reported on in [appendices a, b, c]

Inherent limitations

The procedures specified in DWP’s HBAP Reporting framework instruction does not constitute an examination made in accordance with generally accepted auditing standards, the objective of which would be the expression of assurance on the contents of the Local Authority’s claim for Housing Benefit subsidy on form MPF720A. Accordingly, we do not express such assurance. Had we performed additional procedures or had we performed an audit or review of the Local Authority’s claim for Housing Benefit subsidy on form MPF720A in accordance with generally accepted auditing or review standards, other matters might have come to our attention that would have been reported to you. This report relates only to the Local Authority’s form MPF720A and does not extend to any financial statements of the Local Authority, taken as a whole.

[Add in following paragraph where the Reporting Accountant is also the Local Authority’s external auditor: This engagement will not be treated as having any effect on our separate duties and responsibilities as the external auditor of the Local Authority’s financial statements. Our audit work on the financial statements of the Local Authority is carried out in accordance with our statutory obligations and is subject to separate terms and conditions. Our audit report on the Local Authority’s financial statements is made solely to the Local Authority’s [members], as a body, in accordance with [Part 5 of the Local Audit and Accountability Act 2014]. Our audit work was undertaken so that we might state to the Local Authority’s [members] those matters we are required to state to them in an auditor’s report and for no other purpose. To the fullest extent permitted by law, we do not accept or assume responsibility to anyone other than the Local Authority and the Local Authority’s [members], as a body, for our audit work, for our audit reports, or for the opinions we have formed in respect of that audit.]

Summary of HBAP report

Summary of Initial Testing

Completion of Modules

Summary of testing arising from Cumulative Assurance Knowledge and Experience

In line with the requirements of HBAP Modules we have undertaken CAKE testing based upon the preceding HBAP report. Where appropriate the LA have completed testing of the sub populations for:

Summary paragraph/ending of letter

For the form MPF720A dated………. for the year ended 31 March [current year] we have completed the specific test requirements detailed in the DWP reporting framework instruction HBAP and have identified the following results set out in Appendix A, B, C and D)

Firm of accountants………………………………………………………

Office…………………………………………………………………………….

Contact details (person, phone and email) ……………………………….

Signature / stamp.………………………………………………………

Date……………………………………………………………………….

Appendix A Exceptions/errors found

[if there are no errors then state]

There are no errors to report.

[Alternatively see below]

Appendix B Observations

[If there are no observations then state]

There are no observations to report.

[Alternatively see below]

Appendix C Amendments

[If there are no amendments then state]

There are no amendments to report.

[Alternatively see below]

Appendix D Other matters