Recent and upcoming changes to public sector finance statistics: July 2021

Most methodological changes to the public sector finance 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 or planned within the next 18 months. This edition focuses on changes that will be implemented in the public sector finances (PSF) bulletin published in September 2021. Implementation of the changes discussed in this article is ongoing and estimates of the impact provided here are indicative and liable to change. More information about other recent changes, including the treatment of coronavirus (COVID-19) support schemes already implemented in statistics, can be found in previous editions.

We intend to resume the publication of a separate longer-term work plan, which was suspended during the coronavirus pandemic, later in 2021.

In accordance with our transparency strategy and to provide increased predictability to users, we aim to package together methodological changes so that they occur, where possible, at a single point in the year. This section presents information on aspects of methodology or statistical classifications that are planned for implementation in September 2021. Provisional estimates of how these changes will affect the historical series are presented in Tables 12 to 15 of this month’s public sector finances (PSF) bulletin. We are continuing to work on these changes and the impacts may change.

Alongside the development areas listed here, we will continue to make routine updates to the public sector finances dataset, such as incorporating new data from the UK National Accounts and student loan forecasts into our statistics. The changes induced by economic events (such as the coronavirus (COVID-19) interventions) will still be recorded at the earliest opportunity.

Train operating companies under Emergency Measures Agreements

Emergency Recovery Measures Agreements (ERMAs), which replaced the previous Emergency Measures Agreements (EMAs), are arrangements between government and train operating companies (TOCs) to ensure railways continue to operate during the disruption caused by the coronavirus pandemic. Following a classification review, we concluded that the ERMAs placed TOCs under public sector control.

Currently, we capture the fees paid by the government to TOCs as subsidies to the private sector. These have contributed to the overall rise in public sector net borrowing (PSNB) since the start of the coronavirus pandemic. From September 2021, we will include the full accounts of TOCs in our statistics from the point at which EMAs were effective. We expect public sector net borrowing for the financial year ending March 2021 to be revised downwards by approximately £0.1bn as a result. There is no impact on public sector net debt (PSND).

On 20 May 2021, the government announced its Plan for Rail to reform Britain's railways. This will include the establishment of a new rail body, Great British Railways. We will carry out a classification review of the new rail structure when more details become available.

Future Fund

The Future Fund offered convertible loans to eligible companies affected by the coronavirus pandemic. Convertible debt is recorded as debt securities rather than loans if it can be reasonably expected that it will be repaid. Lending under the scheme is therefore initially viewed as a financial transaction, whereby the government exchanges a cash asset for a debt security asset.

The recording of this lending as a financial transaction means that there is no PSNB impact at the point of lending. In contrast, interest accruing on the loans is considered government revenue and will reduce PSNB by £0.1bn in the financial year ending (FYE) March 2021, as it accrues continuously over the convertible loan term.

Cash is a liquid asset, whereas the convertible debt security is not; therefore, the provision of the loans contributed £1.1bn to PSND in FYE March 2021. This increase is already implicitly captured in our estimates for FYE March 2021, as is any impact of cash payments or receipts over the course of the year. Therefore, the changes we are making to the recording of Future Fund in September 2021 will have no additional impact on PSND.

However, we expect to reduce public sector net financial liabilities (PSNFL) by £1.1bn in FYE March 2021 by recognising an increase in government asset holdings in the form of debt securities (for statistical purposes, convertible loans are recorded as debt securities, differently to conventional loans).

On maturity, we expect this debt to be either repaid (reducing PSND through the acquisition of liquid cash assets) or to be exchanged for equity. The latter scenario may result in the government assuming control over the company, which may in turn lead to its public sector classification. We will review instances of debt-to-equity conversion to establish the appropriate statistical recording on a case-by-case basis.

Government loan guarantee schemes

We described the Coronavirus Business Interruption Loan Scheme, the Coronavirus Large Business Interruption Loan Scheme and the Bounce Back Loan Scheme in previous editions of this article. Using internationally agreed guidance, we have carried out a formal assessment of these schemes and concluded that the guarantees issued under all four schemes should be recorded as standardised.

We expect to fully implement this change in the PSF bulletin which will be published in September 2021. The means that we will record expenditure in the form of a capital transfer at the time the guarantees are provided, reflecting estimated losses under the schemes over their lifetime. It is this expenditure at inception that will contribute to an increase in PSNB of £20.9bn in FYE March 2021. The recording of provision for calls on the government balance sheet will increase PSNFL by the same amount in FYE March 2021, although both estimates are provisional and subject to further changes. PSND is a narrower balance sheet measure than PSNFL; it excludes provisions for calls under standardised guarantee schemes and it will not be affected by this change.

Changes to pensions data

In September 2021, we will be updating our pensions estimates to revise the discount rate assumption from 5% (nominal) to 4% (nominal) in line with international requirements. More information about this change is available in our publication Pensions in the national accounts, a fuller picture of the UK's funded and unfunded pension obligations: 2018. We will also be incorporating recent classifications and new data related to pension funds: notably, we will include the Bank of England Pension Fund in our statistics, and some changes are also expected as a result of the McCloud court case.

In combination, we expect that these changes will lead to a £6.5bn increase in PSNB in FYE March 2021 and a PSNFL increase of around £67.8bn in the same year. The main cause of these revisions is the discount rate change, which increases the value of the pension liability and affects imputed expenditure associated with changes in the liability.

While PSND does not include pension liabilities, it will also be affected. This is because of more information becoming available from our new financial survey of pension schemes and more recent scheme accounts related to holdings of gilts and others assets and liabilities by pension funds. We expect that the consolidation of gilts at a public sector level and improved information on other assets and liabilities will cause PSND to fall by around £11.8bn in FYE March 2021. We expect to see a smaller PSND change in the earlier years. This is because they will be primarily affected by the inclusion of the Bank of England’s pension scheme and not the new survey data, as there are fewer routine data updates that effect these years.

Bank of England accounts

We have improved our processing of Bank of England data to provide more comprehensive coverage and greater accuracy, with updated figures to be published in the PSF Bulletin in September 2021. Improvements include increasing the granularity of data used to calculate PSF aggregates, fully consolidating the Bank of England’s holdings of UK government securities (gilts) and some revisions to the Bank of England’s contribution to PSND.

Although the use of more granular data has resulted in minor revisions, the Bank of England’s overall contribution to PSNB will remain similar.

The impact on PSND will be somewhat greater, with the Bank of England’s contribution expected to be revised up by around £4.2 billion for FYE March 2021. The largest driver of this change is the consolidation of gilts held by the Bank of England, which is now done at face value consistent with the approach we apply to other public sector bodies. More granular data processing has also allowed for amendments to the historical series (between March 2014 and February 2018) to ensure consistency in how the Bank of England’s PSND contribution is calculated over time.

Asset Purchase Facility

Having considered the coverage and granularity of Bank of England accounts information, we are also reviewing the data sources for the Bank of England Asset Purchase Facility and Term Funding Schemes. Our quality assurance work continues, and although we aim to implement updates in September 2021, provisional estimates of the impact are not yet available

Sale of railway arches

In February 2019, Network Rail completed the sale of its Commercial Estate business, primarily consisting of railway arches. Public sector net debt at the end of February 2019 and the central government net cash requirement in February 2019 were reduced by an amount equivalent to the cash received by central government from the sale. We announced on 31 March 2020 that the agreement would mainly be treated as an operating lease with payments for market output being made over a long period of time. Further details are given in the public sector classification guide.

The new treatment will be implemented in September 2021. PSND will not be affected by the change, as any impact on cash payments or receipts will already have been reflected in the latest estimates (of public sector net debt and central government net cash requirement). However, the broader PSNFL will see an increase of around £1.5bn at the time of the initial sale, with the impact diminishing over time as the imputed payments accrue. The impact on PSNB is minor.

We continue to work on other developments detailed in the March edition of this article, including:

• improving and extending our balance sheet statistics
• reviewing the treatment of leases under International Financial Accounting Standard 16
• implementing a new accruals methodology for national non-domestic rates (business rates)

We will provide further updates on these areas in the coming months.