Strengthening the quality of HMRC’s official statistics
Today we publish the findings of our review of the principles and processes underpinning the quality of HMRC’s official statistics.
We welcome HMRC inviting OSR to carry out this review after HMRC identified a significant error in its Corporation Tax receipt statistics, which affected the period from April 2011 to July 2019. The aim of the review was to provide an independent assessment of HMRC’s quality management approach for official statistics and to identify improvements that lower the risk of errors, especially errors that have a large impact on the statistics and their interpretation by users.
HMRC produces many official statistics and most of the time it does this without issues or errors. We found much to admire in the way that HMRC manages the quality of its statistics, but several areas where it could strengthen its approach: we make nine recommendations to address these issues. Our findings highlight the issues that big operational departments working with administrative data can face with respect to quality and will ring true for other Government departments.
The most immediate of our recommendations relate to establishing and maintaining effective relationships between those producing official statistics in HMRC and other parts of HMRC they receive data from; understanding data journeys; and quality assurance of data, analysis and publications. Further recommendations focus on strengthening the leadership and oversight of quality management; building more effective routes for knowledge sharing about quality; and improving the information HMRC publishes about the quality of its official statistics. Implementing these recommendations will strengthen the quality culture in HMRC and make sure that the structure and tools are in place to manage statistical quality effectively, minimising risk.
Our report also notes that, like many other large government departments, most of HMRC’s IT systems, from which the administrative data that underpins almost all of HMRC’s official statistics, and other analysis, are drawn, have been designed for operational delivery, rather than for statistical analysis, and have been in place for many years, so may not have the same functionality or flexibility as newer IT systems. Along with the recommendations about statistical quality management, investment in underlying IT systems to ensure they keep pace with these modern expectations (while also being of wider benefit to HMRC in fulfilling its duty of collecting taxes), will be important to enable HMRC to continue to confidently exploit the value of the data it holds. I have written separately to Jim Harra, First Permanent Secretary and Chief Executive for HMRC, about the need for this wider investment.
I’d like to thank HMRC for its open and positive engagement with my team throughout this review, and I look forward to future engagement with HMRC as it takes steps to strengthen the quality management and assurance of its official statistics. I am copying this letter to Jane Whittaker, Director of HMRC’s Knowledge, Analysis and Intelligence directorate, and Sean Whellams, HMRC Head of Profession for Statistics.
Director General for Regulation
Ruth Stanier to Ed Humpherson (April 2020)