Recent and upcoming changes to public sector finance statistics: July 2022

Most methodological changes to the public sector finance (PSF) statistics are prompted by the need for statistics to keep pace with the evolving economy. These changes can alter perceptions of the government's fiscal exposure, as measured by the fiscal aggregates, such as public sector net borrowing (PSNB) or public sector net debt (PSND). This article increases transparency around the methodology work, including classification reviews, that is currently underway.

In accordance with our transparency strategy, and to provide predictability to users, we aim to package together methodological changes at a single point in the year where possible. This section presents information on the changes outlined in our Looking ahead - developments in public sector finance statistics: 2022 article, which are planned for implementation in September 2022.

Provisional estimates of how these changes will affect the fiscal aggregates are presented in Tables 14 to 15 of this month's public sector finances bulletin. We are continuing to work on these changes, and the provisional impacts presented in this article may change.

Improvements to public corporations' data

In September 2022, we will be improving the coverage and quality of our public corporations dataset.

The public sector includes a number of companies that operate on a market basis and are classified outside the general government sector. Such companies are known as public corporations. The statistical treatment of these companies poses some practical challenges, particularly with respect to data collection. We will improve our coverage of these companies by extending the use of HM Treasury's Whole Government Accounts (WGA) dataset which covers both government organisations and public corporations.

The WGA are compiled in accordance with the International Financial Reporting Standards (IFRS) - a different framework to the one used in fiscal statistics - but provide valuable source data for both purposes. We have adapted the dataset by mapping relevant flows and stocks to their statistical equivalents. As we account for the conceptual differences between the frameworks (such as the asset recognition boundary, treatment of certain flows and valuation methods), the aggregates presented in our statistics could differ from equivalent estimates presented in HM Treasury's WGA publication.

Additionally, we have identified and corrected a number of historic data issues within the public corporations data used for our fiscal statistics. These include the consistent recording of certain subsidies in both the government sector and public corporations sector accounts, as well as some improvements to the balance sheet data.

In combination, these changes are expected to increase public sector net borrowing (PSNB) by £0.1 billion in the financial year ending (FYE) March 2022.

We also expect these changes will lead to a £2.8 billion increase in public sector net debt (PSND) in the FYE March 2022 and a £2.5 billion increase in public sector net financial liabilities (PSNFL) in the same year. However, the magnitude of the balance sheet revisions will vary across the time series thanks to improvements to the recording of retroactive classification decisions and more accurate consolidation of assets held by other parts of the public sector. The largest PSNFL revision will occur in the FYE March 2006, with an increase of £10.5 billion.

We expect to continue making improvements beyond September 2022, particularly focusing on our estimates for the latest time periods, which we need to use until the WGA dataset for the relevant year becomes available.

Recording of central government leases

In September 2022, we expect to transition to the new treatment of leases for central government organisations.

The change, described in our Looking Ahead article, has been motivated by a classification review of central government property leases. The review established that because substantial risk transfer from the lessor to the central government lessee was a common feature of such contracts, it would be more appropriate to record them as "financial leases" (a statistical term for finance leases) rather than operating leases.

This means that the non-financial asset, along with a corresponding financial liability, will now be included on the government balance sheet. Because there are thousands of individual lease agreements, we estimated the total value of the leased assets by relying on the information on combined payment commitments under all the central government lease contracts.

The changes to non-financial assets are reported on a net basis, across the whole of central government, and as such, the impact of the changes on Public Sector Net Investment (PSNI) will combine the impact of new lease agreements being taken out as well as the disposal of leased assets. Movements of contracts because of the machinery of government changes will not impact the figures unless the leases are also renegotiated at a higher or lower value.

The change will affect the fiscal aggregates in two ways. Firstly, only the interest element of lease payments will now be treated as expenditure. The expenditure associated with the principal value of the asset will be imputed at the time the lease agreement is entered into. This will reprofile government spending across the entire time series, decreasing public sector net borrowing (PSNB) by £0.4 billion in the financial year ending (FYE) March 2022. It is worth noting that the impact will vary across the time series, reflecting the uneven path of lease contract start and end dates.

Secondly, the properties subject to lease agreements will now be included on the government balance sheet, as will the imputed loan liability arising from spreading the payments over the duration of the leases. Only the financial liability is included within the scope of public sector net debt (PSND) and public sector net financial liabilities (PSNFL), meaning that both will see an upward shift of £20.9 billion in the FYE March 2022.

The data source underlying this work has a lag of approximately two years between the period to which it relates and publication of the data. The latest years are therefore modelled on the available data, and as such will be subject to revisions as new data become available, as well as ongoing improvements to the modelling.

We will continue the work on leases beyond September 2022. This is because we expect the adoption of the International Financial Reporting Standard 16 (IFRS 16) will better align the source data to the needs of fiscal statistics and reduce the need for statistical modelling in the future. We also expect to transition to using IFRS 16 estimates of asset values when such data become available. We will monitor the implementation of IFRS 16 in other parts of the public sector and will transition to the new methodology at an appropriate time.

Treatment of national non-domestic rates

In September 2022, we expect to adopt a new method for measuring national non-domestic rates (NNDR), commonly known as "business rates", on an accrual basis, as explained in our Looking Ahead article.

This new methodology has been prompted by changes in business rates retention by local authorities, and the new option for businesses to pay the rates in 12 monthly instalments. The existing estimates of NNDR are calculated using a combination of cash information and an assessment of likely receipts. The new accruals methodology provides a more accurate measurement of NNDR; as it is derived from time-adjusted cash receipts, it avoids the potential inaccuracies of the assessment data that are currently being used.

It is equally important to identify the institutional sector to which the body making the NNDR payment belongs. Payments made by public sector units do not contribute to the estimates of public sector net borrowing (PSNB), being simultaneously revenue and expenditure for the public sector. Additionally, the statistical tax category under which the NNDR payments are reported depends on the sector classification of the payer, and specifically on whether it is a market or non-market entity. Where administrative data are not available at this level of detail, the new method uses Valuation Office Agency figures for rateable properties to derive a sector split of the measurement of total NNDR receipts.

Additionally, we intend to improve the presentation of central government payments of NNDR so that they are shown as both a payment and a receipt of taxes, rather than being netted off in our presentation. We will also lessen the risk of double counting of central government expenditure where an element of central government payment of reliefs was also included in intermediate consumption.

Throughout the project, we have worked with other relevant government departments to comprehensively review the recording of NNDR payments and receipts. Following our investigations and assessment of statistical guidance, we have concluded that the implementation of the new method will not affect other existing flows between central and local government, such as those relating to recording of business rate-related grants to local authorities.

As a result of these changes, we expect to make revisions to PSNB, the largest of which is expected to occur in the financial year ending (FYE) March 2011 (an increase of £0.6 billion). Updated source data in the FYE March 2021 and the FYE March 2022 will further contribute to revisions, with the FYE March 2022 seeing a total downwards revision to PSNB of an estimated £1 billion. Public sector net financial liabilities (PSNFL) will be revised upwards by £0.1 billion in the FYE March 2022. There will be no impact on PSND.

Energy profits levy

On 26 May 2022, the UK government announced a new levy on the profits of oil and gas companies in the UK and the UK Continental Shelf, intended to help fund more cost of living support. The Office for National Statistics (ONS) will review and classify this new levy in due course.