A FREE resource from the UK accountancy profession

PAYE Real Time Information, Universal Credit & your business Read More

Report Issues


HMRC’s report on RTI pilot admits some key problems

July 18, 2013


HMRC’s own report on its RTI pilot has admitted a number of the key concerns already identified by the National Audit Office. The HMRC report was published on 18 July, 2 weeks after the NAO published its findings to date on RTI. The NAO report into RTI published on 3 July 2013 raised grave concerns about HMRC’s implementation of RTI and the capabilities of HMRC’s new PAYE systems, which HMRC’s report ignores.

HMRC’s report, titled ‘The Real Time Information Pilot’, summarises its own findings of the RTI pilot and explains how information gathered in the pilot influenced HMRC’s wider roll-out of PAYE RTI, specifically guidance for employers coming into real time reporting.  However, some of the fundamental concerns raised by the NAO report of 3 July are not addressed by HMRC, who only reply in detail to the NAO’s point querying the number of PAYE schemes HMRC expected to participate in the pilot.

The report attempts to show that the management of the RTI pilot was successful in most aspects and that the expansion of the pilot and the additional costs incurred as a result were justified.  It also gives some information on how feedback from groups consulted in the planning and pilot stages was incorporated into the RTI programme, in particular on the issue of the leaver/joiner process and the use of a revised P45 form. 

HMRC deflect the concern voiced by the NAO, that the number of schemes which participated in the pilot was significantly lower than expected, by asserting that the 250,000 employer target was the pilot’s capacity rather than an expectation of take-up (a point questioned by’s 11 July news item). HMRC states it expanded the pilot because of the positive feedback HMRC received from employers during the pilot, and in order that the DWP’s Universal Credit system interface with HMRC systems had sufficient data to test.  The eventual number of records in the system was more than 6 million, which HMRC considers sufficient to ensure adequate testing was done. Although it appears clear from other information published by HMRC both prior to the pilot’s commencement in April 2012 and one month later that the scale of the pilot was not expanded after it had begun.

The key RTI issues admitted by HMRC in its report which arose during the operation of the pilot were: the creation of duplicate employments; insufficient employer use of the EPS report; PAYE payments made by employers not being credited to the correct period; the starter/leaver process; and, problems with HMRC’s Basic PAYE Tools RTI software for small employers.

However, the report omits the weaknesses in HMRC’s new RTI financial and accounting systems and the insufficient provision of disaster recovery and resilience for the RTI scheme, identified as fundamental areas of concern by the NAO.

The creation of duplicate employments the report largely attributes to ‘payroll software glitches’ or ‘employers changing payroll IDs without indicating… that they had made this change’. In instances where an employer or their payroll provider changes the internal employee payroll designation (something that has no bearing on PAYE) HMRC’s systems automatically create a new employment. Although HMRC seeks to blame employers, the report does not answer why their systems make this assumption from a miscellaneous item of data that is irrelevant to the operation of PAYE.

HMRC say they have addressed the issue of duplicate employee records (also resulting in incorrect tax codes where benefits in kind are payable) and, in some cases, demands for increased employer charges, by improving data matching within their systems and by giving further guidance to software developers and employers.

Worringly, the report admits that HMRC have demanded payment of additional sums from employers for which there was no tax liability, stating that:

‘during the pilot a number of separate issues were identified which have led to HMRC expecting, and pursuing, higher payments from employers than were actually due’,

but lists the failure by employers to submit an Employer Payment Summary within the correct timeframe as a key cause of this problem.  Although the EPS is not a mandatory submission, it serves to tell HMRC that there is no employer liability for that pay period, or that there is a reduced employer liability because of the recovery of statutory payments or similar items.  This issue is identified by HMRC as one which is likely to be ongoing, because reporting information like this is a cultural shift away from annual reporting. The report further states that:

‘there is a risk that, when automatic late payment penalties are implemented from April 2014, HMRC’s systems will create penalisable payment defaults in the absence of an EPS reporting adjustments to the employer’s liability.’ 

HMRC has not resolved this issue, but is ‘monitoring how to support the employer community further to ensure EPS are filed promptly where necessary.’

Central to the successful operation of PAYE in real time is HMRC’s ability to capture automatically from employer RTI data submissions, when an employee starts or leaves an employment. This is perhaps the single most important and challenging capability of HMRC’s RTI systems: the requirement to capture this information dynamically from employer data without it being explicitly reported as such by employers. HMRC admits the challenge and acknowledges there are issues with automatically capturing starter and leaver data. However, its report does not address how this is to happen. The report refers to “the longer term vision” for dispensing with P45s entirely and appears to suggest that problems with P45s contributing to duplicate employments is the responsibility of the ‘employer community’, who requested the retention of a reduced P45 no longer used by HMRC, but which is now used only to signal a new employment to an employer.

The report acknowledges the problem occurs when an employee with more than one job gives their P45 to a new employer, creating an incorrect tax code within HMRC’s systems. However, the report only states: “This shortcoming was corrected prior to full roll out,” but gives no information on how this was achieved.

The timing of the calculation of payroll resulted in some misalignment between the tax month and the payments which relate to it, what is known as ‘month end cross over’. HMRC correctly identifies this as an employer issue where the date the payroll was run was incorrectly used instead of the pay date when calculating deductions, something Simon Parsons, Director of Payments, Benefits & Compliance Strategies at Ceridian confirmed in May. HMRC published guidance on this issue as early as 29 April 2013.

There is no mention in the report of the lack of technical resilience and disaster recovery mechanisms highlighted by the NAO, although in a recent evidence session before the Department for Work and Pensions Select Committee Suzanne Newton (head of the RTI programme at HMRC) insisted that they had provided a ‘cost effective’ solution which would only be at risk in ‘catastrophic’ circumstances.

Nor does the HMRC pilot report cover the ‘weaknesses’ in the accounting and financial reporting systems supporting RTI.  This may be because only the transfer of starter and leaver data into the NPS (National Insurance and PAYE Service) system was tested during the pilot phase of RTI, according to the NAO, and this report only covers the operation of the pilot.  The HMRC report does say that:

Another aim of the pilot was to identify any issues with HMRC systems and processes before the main migration.  HMRC set up an Issues Resolution Forum (IRF) to analyse key issues and identify resolutions. This is an ongoing process and the RTI Programme is continuing to work closely with all stakeholders to ensure issues are resolved and communications are updated.

The report briefly mentions that the Debt Management and Banking Office of HMRC ‘were able to test the new accounting system required to monitor PAYE charges’, but gave no indication as to the outcome of this testing.  The NAO report suggested HMRC undertake urgent further investigation to assess the possible impacts of the financial accounting failures they identified.

The report goes on to say that:

HMRC recognises that the transition to in year reconciliation of employer charges is ongoing.  This is being monitored closely to improve the interaction HMRC has with employers, and to update guidance when required.”

This would appear to confirm the financial accounting issues of HMRC’s new PAYE systems and HMRC’s difficulties reconciling employer PAYE payments with the liability to PAYE as reported by their RTI data submissions.

Basic PAYE Tools, the free software provided for micro employers by HMRC, was found to be ‘less intuitive’ than commercial software by pilot employers.  HMRC responded to this issue by increasing the guidance available to employers for using the product, and developed ‘further enhancements to improve its performance’.

HMRC’s report emphasised the positive feedback they received, via “independent research”, from employers who had been in the RTI pilot, with 97% confident about dealing with PAYE in real time and 86% finding RTI ‘easy’. But the numbers of employers sampled are very low because the overall number of employers participating in the pilot was also small. Additionally, there was considerably less pressure on HMRC support staff during the pilot as there were fewer employers (only 300 in the first phase of the pilot) in the pilot, compared to the million or so employers reporting RTI from April 2013.

As yet there are no definitive statistics for employer experience after the pilot, since when HMRC support functions have been under significant strain as a result of the mass migration of all UK employers into RTI.

< Back to news

comments powered by Disqus