Official Statistics

Methodology and Quality Document - Monthly Insolvency Statistics August 2021

Published 17 September 2021

1. Data Sources

1.1 Data sources for insolvencies

Company Insolvency data for England & Wales, Scotland and Northern Ireland are sourced from Companies House, except for compulsory liquidation data for England & Wales and Northern Ireland.

Compulsory liquidation data for England & Wales are sourced from the Insolvency Service case information system (ISCIS). Compulsory liquidation data for Northern Ireland are sourced from the Department for the Economy, Northern Ireland.

Individual insolvency data for England & Wales were sourced from ISCIS; individual insolvency data for Northern Ireland were sourced from the Department for the Economy.

1.2 Data sources for non-insolvency measures

Company moratorium and restructuring plan data for England and Wales are sourced from Companies House.

Individual breathing space data are sourced from the Breathing Space register, owned by HM Treasury (HMT), for which the Insolvency Service is a custodian.

More information on the administrative systems used to compile the agency’s statistics can be found in the Statement of Administrative Sources.

2. Methodology

2.1 Data extraction and validation

Companies House routinely supply data covering company insolvencies and non-insolvency measures (moratoriums and restricting plans) on a quarterly basis. These data are used to compile the Insolvency Service’s Quarterly National Statistics. Companies House conduct a series of basic checks on the data including the removal of duplicate records prior to data supply.

In the interim months between the quarterly data supply, Companies House provides daily datasets which are combined by the Insolvency Service statistics team to create the interim monthly datasets. These data are not validated and are subsequently overwritten by the next validated quarterly dataset when available.

Individual insolvency data for England & Wales, and compulsory liquidation data for England & Wales are extracted from the Insolvency Service’s administration system approximately five working days after month end. Due to the tight turnaround for publication these data are not validated except to remove duplicate compulsory liquidations. As with the company insolvency data, there is a more robust validation process for the compilation of the quarterly statistics, as outlined in the Administrative Data Quality Assurance and Audit Arrangements.

Note that the process carried out to remove duplicate liquidation data was reviewed and changed ahead of the release of the January 2021 monthly statistics (published on 19th February). The process was improved to ensure that the correct duplicate record was retained. This resulted in a small number of revisions to the month in which the liquidation was recorded. The impact of these revisions is considered minimal since there were no changes made to the overall totals of liquidations.

Individual insolvency and compulsory liquidation data for Northern Ireland are provided in aggregate form by Northern Ireland’s Department for the Economy. The Insolvency Service does not conduct any further validation on these aggregate data.

Some automated validation checks are made on the individual breathing space data, such as ensuring that dates are formatted correctly. However, data are extracted from the breathing space register in aggregate form to protect the identity of individuals and therefore the Insolvency Service does not conduct any further validation on these aggregate data.

2.2 Seasonal adjustment of data

Insolvency data with seasonal or other calendar influences that influence patterns in the data series are typically seasonally adjusted. The process of seasonal adjustment results in a clearer picture of the underlying behaviour of the data.

However, the underlying data used to compile the Monthly Insolvency Statistics have not been seasonally adjusted since the monthly historical time series is too short to enable any seasonal trends to be determined. Additionally, the resource required to seasonally adjust the data each month would result in less timely publication of these statistics.

Since the data are not seasonally adjusted, month-on-month changes are not valid. Instead, comparisons are made in the commentary between the latest month and the same month in the previous two years.

2.3 Tabulating numbers of company and individual insolvencies

Company insolvency data, as provided by Companies House are tabulated by insolvency type (creditors’ voluntary liquidations (CVLs), administrations, company voluntary arrangements (CVAs) and receivership appointments) and calendar month in which each insolvency was registered with Companies House. Therefore, numbers reflect company insolvency registrations rather than insolvency procedure start dates.

Compulsory liquidation data are sourced by the Insolvency Service and are reported by court order date, thus providing a more accurate measure of the number of new cases in each month. However, data for the latest month are extracted from a live system approximately five working days after month end and are subject to change and therefore figures are marked provisional.

Bankruptcy and Debt Relief Order (DRO) data for England & Wales are tabulated by insolvency type and calendar month of order or agreement. Similarly to compulsory liquidations, data for the latest month are extracted from a live system approximately five working days after month end and are subject to change and therefore figures are marked provisional.

Individual voluntary arrangements (IVAs) in England & Wales are counted within these statistics once they are registered with the Insolvency Service, and they are reported by month of registration date. There is often a time lag between the date on which the IVA is accepted (known as the date of creditor agreement) and date of registration by licensed insolvency practitioners working for firms that specialise in this area and changes are often partly a result of how promptly and frequently providers register IVAs with the Insolvency Service. Therefore, the underlying data for IVAs are volatile from one month to the next (see Quality section below for information on the volatility of these data).

Three-month rolling averages are presented to smooth the data and indicate what the overall trend of IVA registrations might look like if the underlying data were less volatile. Three-month rolling averages are the mean average numbers of registered IVAs in the three months ending in the reference period. For example, the three-month rolling average estimate for January 2019 is the calculated mean average of the total IVA registrations during November 2018, December 2018, and January 2019. The mean average was calculated by summing together the actual numbers of IVAs registered in each month and dividing by three.

For transparency, both counts and three-month rolling averages are presented. Whilst three-month rolling averages are used to consider potential changes in IVA trends over time, trends are still unreliable. It is particularly important to consider longer term trends when making assessments of individual voluntary arrangements, as presented within the Quarterly Individual Insolvencies statistical series.

Aggregate counts of breathing space registrations are provided, tabulated by month of registration on the breathing space register.

2.4 Tabulating insolvency by industry tables

Company insolvency data for England & Wales and Scotland, as supplied by Companies House, contains a mixture of the 2003 and 2007 Standard Industrial Classification (SIC). Where a 2003 SIC code has been supplied, it has been converted to a 2007 code using a ‘best match’ approach using weighted tables provided by the Office for National Statistics. Where more than one SIC code is recorded for a company, only the first listed, or the primary, code is used.

For compulsory liquidations in England & Wales, companies are matched against the Companies House register where possible, and SIC data are drawn from this source. Where matching is not possible, data from the Insolvency Service’s administrative systems are used instead. The Insolvency Service’s administrative systems capture 2003 SIC codes and are therefore converted using the same method as for 2003 SIC codes captured in Companies House data.

Errors were identified in the process for matching compulsory liquidation data to the Companies House register during the production of the January 2021 company industry statistics (published on 19 February 2021). These errors were corrected and the time series back to January 2019 updated in Tables 2 and 2.1 in the accompanying xlsx and ods tables for all subsequent releases.

The impact of these errors is considered minimal. Whilst some of the numbers changed, the percentage of all insolvencies within each industry remained unchanged in most cases. Any percentage changes identified were by no more than 1%. Details can be found in the January 2021 Methodology and Quality Document.

Insolvency Service data on bankruptcy orders relating to the self-employed in England & Wales capture the SIC related to the industry of the individual’s self-employed business. Again, SIC 2003 codes are converted to 2007 codes using the ‘best match’ approach described above.

During the production of the January 2021 release it was identified that the ‘Mining support service activities’ industry had been missing from Table 6 since June 2020. This industry was added in for all subsequent publications. The impact of this omission is considered low since there was only one company insolvency in this industry between January 2019 and January 2021.

For information on SIC 2007, including its structure and more detailed information on which industries are included, please see the Office for National Statistics website.

3. Revisions

These monthly insolvency statistics are subject to scheduled revisions, as set out in the published Revisions Policy. Other revisions tend to be made as a result of data being entered onto live administrative systems after the cut-off date for data being extracted to produce the statistics. Data for the most recent month were extracted approximately five working days after month end, so there is an increased likelihood that published statistics may be revised in the future. Therefore, these monthly statistics are marked provisional. Any future revisions will be marked with an ‘[r]’ in the relevant tables.

Breathing space statistics are subject to revisions since the data are extracted from a live administrative system. Any future revisions will be marked with an ‘[r]’ in the relevant tables.

The number of IVAs presented for December 2020 were incorrect (Table 4 and subsequently Table 4.1) in the December 2020 statistics (published 15 January 2021). Compulsory liquidations, and therefore total new company insolvencies, were also presented incorrectly for Northern Ireland (Table 9). These above errors were a result of manual error during the table compilation process. From the January 2021 release (published 19 February 2021), the Insolvency Service statistics team has introduced a new improved process to prevent these errors in the future.

Compulsory liquidations can be annulled or rescinded following a court order. Prior to the March 2021 statistics (published 15 April 2021), numbers included compulsory liquidations which had been subsequently annulled or rescinded. In all subsequent releases, compulsory liquidations which have been annulled or rescinded in England & Wales have been removed. As a result of this change, 84 compulsory liquidations registered between 1 January 2019 and 31 March 2021 were removed which had previously been included.

During the production of the May 2021 statistics (published 15 June 2021), an error was identified in the process for assigning SIC codes for compulsory liquidations presented in Tables 2 and 2.1. This has been corrected for all future releases.

Until the June 2021 release, these statistics did not include Debt Relief Orders that were approved but later revoked. Subsequent releases do include these cases for all months back to the start of the series.

4. Quality

This section provides information on the quality of these monthly insolvency statistics, to enable users to judge whether the data are of sufficient quality for their intended use.

The section is structured to align with the Quality Assurance Framework of the European Statistical System for statistical outputs.

Relevance: The degree to which the statistical product meets user needs in both coverage and content.

The insolvency statistics are the most comprehensive record of the number of corporate and individual insolvencies in the UK. They include all formal types of insolvency procedure currently available.

These monthly statistics have been compiled in response to the coronavirus (COVID-19) pandemic, with the aim of publishing more frequent information to better meet user need during this time of economic uncertainty. The methodology and content will continue to be reviewed to ensure that they meet user need.

This monthly series supplements the Insolvency Service’s quarterly company and individual insolvency National Statistics to provide more up to date information, as the coronavirus (COVID 19) pandemic continues, on the numbers of companies and individuals who are unable to pay debts and enter a formal insolvency procedure.

This statistical release presents company insolvencies for England & Wales, Scotland and Northern Ireland. Individual insolvencies are presented for England & Wales, and Northern Ireland.

Individual insolvencies in Scotland are compiled and published separately by the Accountant in Bankruptcy (AiB). The Insolvency Service incorporates individual insolvencies in Scotland into the quarterly individual insolvency statistics for completeness. However, to avoid any unnecessary delay in releasing these more timely statistics, monthly individual insolvencies for Scotland were not pulled into this statistical release. Until May 2021, monthly individual insolvency statistics for Scotland were published on the Accountant in Bankruptcy (AiB) website. However, this publication has now been discontinued as per this bulletin.

Key users of insolvency statistics include the Insolvency Service itself, which has policy responsibility for insolvency in England & Wales and for the non-devolved areas within Scotland and Northern Ireland; other government departments; parliament; the insolvency profession; debt advice agencies; media organisations; academics; the financial sector; the business community and the general public. Insolvency statistics are typically widely reported in national, regional and specialist media on the day of release.

The statistical production team welcomes feedback from users of the Insolvency Statistics (current contact details are provided on the front page of the latest release).

Accuracy and Reliability: Accuracy is the proximity between an estimate and the unknown true value. Reliability is the closeness of early estimations to subsequent estimated values.

All formal insolvency procedures entered into by a company, a partnership or an individual are required by law to be reported to the appropriate body, so Insolvency Service statistics should be a complete record of insolvency in the United Kingdom.

Numbers of insolvency cases are typically based on the date they were registered onto the relevant administrative recording system, and so it should be noted when making comparisons of trends over time, that trends can be influenced by late reporting.

The impact of delayed reporting is particularly an issue for IVAs. IVAs are counted within the statistics once they are registered with the Insolvency Service and are reported by month of registration date. There can be a time lag between the date on which the IVA is accepted (known as the date of creditor agreement) and date of registration by licensed insolvency practitioners working for firms that specialise in this area. This time lag can lead to volatility in the data from one month to the next and create difficulty in constituting reliable short-term trends, since changes over time may be partly a result of IVA provider activity. Therefore, three-month rolling averages have been calculated to smooth the data and indicate what the overall trend of IVA registrations might look like if the underlying data were less volatile. For transparency, both the counts of IVA registrations and three-month rolling averages are presented in the statistics.

Checks are in place to identify and remove duplication of cases, to ensure that returns cover all reporting areas, and to check consistency within tables and between related tables.

Note that data are extracted from live administrative systems. Data for the most recent month were extracted approximately five working days after month end. Whilst there is no expectation for revisions, there is an increased likelihood that published statistics may be revised in the future. Therefore, these monthly statistics are marked provisional.

Timeliness and Punctuality: Timeliness refers to the elapsed time between publication and the period to which the data refer. Punctuality refers to the time lag between the actual and planned dates of publication.

Individual and company insolvency data were extracted approximately five working days after latest month end. The exception to this were the bankruptcy statistics by employment status, and the industry breakdowns for those who were self-employed where there is a one-month delay in reporting. This is because it can take several weeks for employment status to be recorded following the date of the bankruptcy order.

Typically, these statistics are scheduled to be released two weeks (10 working days) after month end, to provide faster indicators of the impact of the coronavirus (COVID-19) pandemic on insolvency across Great Britain. Where a scheduled release date falls on a Monday the scheduled release has been pushed forwards by a day, to the Tuesday, to ensure compliance with the Pre-release Access to Official Statistics Order.

The publication schedule for these statistics, and all other Insolvency Service statistics, can be found on the UK National Statistics Publication Hub.

Comparability and Coherence: Comparability is the degree to which data can be compared over time and domain. Coherence is the degree to which data are derived from different sources or methods, but refer to the same topic, are similar.

The Insolvency Service also publishes quarterly company and individual statistics. As detailed in the methodology section above, the sum of these monthly statistics may not equal previously published quarterly statistics due to the use of different reporting dates for bankruptcies, DROs and compulsory liquidations. In addition, the administrative systems used to capture data are live systems and are subject to amendments.

Companies House produces official statistics on companies in liquidation and in dissolution. However, these figures do not align with the insolvency statistics, since they include companies that wound up voluntarily.

The Gazette (formally the combination of three publications: The London Gazette, The Belfast Gazette and The Edinburgh Gazette) is an official journal of record consisting of statutory notices, including company and personal insolvency notices. The timings of the publication of Gazette notices and the registration of company insolvencies at Companies House may differ and therefore the numbers of insolvencies in a specified time period may not align. Additionally, it is only a legal requirement for certain insolvency procedures to be published in the Gazette. Notifications of company CVAs are not published. In terms of personal insolvency, only notifications of bankruptcies are published.

The Office for National Statistics also produces annual statistics on business “deaths” in its Business Demography publication. Again, figures do not align with Insolvency Service statistics as they relate to all registered businesses, whereas the insolvency statistics relate only to companies on the Companies House register, and business deaths include all types of dissolution in addition to insolvencies.

The AiB is required to be notified of all company liquidations and receiverships in Scotland, and publishes official statistics based on its own administrative records. These differ from the Insolvency Service statistics, which use data from Companies House as the source. Differences are due to the AiB using its own administrative system’s date rather than the date of registration at Companies House, The AiB does not publish information on the number of company voluntary arrangements or administrations, which are a reserved matter for the UK government. AiB also publishes quarterly statistics on the number of individual insolvencies, which are reproduced in the Insolvency Statistics.

Accessibility and Clarity: Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of metadata, illustrations and accompanying advice.

Insolvency Statistics are available free of charge to the end user on the GOV.UK website. They are released via the Publication Hub and they meet the standards required under the Code of Practice for Official Statistics.

The accompanying data tables are formatted in line with current guidance for producers of official statistics to help improve the usability, accessibility and machine readability of spreadsheets. The Government Statistical Service are continuing to review this guidance and so the presentation of these statistics may change in the future.

Historical insolvency data are also published for the key series, on the National Archives website.

Views on the clarity of the publication are welcomed via the contact details on the cover page of this release.

Any enquiries regarding this document/publication should be sent to us at statistics@insolvency.gov.uk.