Official Statistics

Quality Assurance Report

Published 9 July 2020

1. Introduction

This document summarises the quality assurance processes applied during production of the DCMS Sectors Economic Estimates: Business Births and Deaths release (published 9th July 2020). The release covers the number of newly-registered businesses and de-registered (ceased trading) businesses in DCMS sectors, and is based on data provided to DCMS by the Office for National Statistics (ONS). This document covers quality assurance carried out by both DCMS and our data providers (ONS). The ONS component is largely based on Business Demography Quality and Methodology Information (QMI) report from November 2019 and should be credited to colleagues at ONS.

2. Quality assurance processes at ONS

2.1 Data collection and accuracy

The data underpinning this release are taken from the Office for National Statistics (ONS) Inter-Departmental Business Register (IDBR). The information used to create and maintain the Inter-Departmental Business Register (IDBR) is obtained from five main administrative sources. These are:

• HM Revenue and Customs (HMRC) Value Added Tax (VAT) – traders registered for VAT purposes with HMRC
• HMRC Pay As You Earn (PAYE) – employers operating a PAYE scheme, registered with HMRC
• Companies House – incorporated businesses registered at Companies House
• Department for Environment, Food and Rural Affairs (DEFRA) farms
• Department of Finance and Personnel, Northern Ireland (DFPNI)

As well as these five main sources, a commercial data provider, Dun and Bradstreet, is used to supplement the IDBR with Enterprise Group information.

In addition, the Office for National Statistics (ONS) Business Register and Employment Survey (BRES) and other ONS surveys supplement these administrative sources, identifying and maintaining the business structures necessary to produce detailed industry and small area statistics. It should be noted that BRES is the only source of local unit (site) information. It is used to populate the employment and classification of a business.

A more detailed paper on the IDBR sources, structures and updating process is published on the ONS website.

2.2 Data processing and quality assurance

Since the publication is an extract from the IDBR, which is based on administrative records, there is no estimation or imputation. However, it is important to note that underlying the extracts are IDBR processing rules, which impact, for example, on the timeliness and classification of businesses.

Information on the administrative source data used to construct and compile the IDBR, the quality control checks, processes and maintenance procedures are described in Introduction to the IDBR.

The constituent data used to populate the IDBR are received from the administrative sources daily, and they are subjected to rigorous testing and quality control checks before being uploaded onto the IDBR. Checks include:

• Matching HMRC, VAT and PAYE information;
• Checking that business locations and structures match PAYE and VAT information;
• Checking that employment data are correct;
• Checking that businesses are active; and allocating businesses to correct standard industrial classifications.

These tasks are carried out via automatic system checks, with any changes or errors reported for manual investigation, checking before correction and subsequent uploading.

IDBR data are extracted for the publication and placed into a series of pre-defined tables using SAS Enterprise Guide software. Data in each of the tables are then quality checked to ensure that information has been extracted correctly and reflects the information contained on the IDBR.

Statistical disclosure control is also applied to the data to ensure that information attributable to an individual organisation is not disclosed.

The data tables in the publication are broken down by UK Standard Industrial Classification 2007 (SIC 2007) and by geography down to local authority or district level.

2.3 Time period and timeliness

The ONS publish Business Demography statistics from the IDBR. These measure business births and deaths across the UK economy and represent the closest “relative” to the DCMS Business Births and Deaths statistics, which consequently follows the same approach. For Business Demography, the reference period is December, and therefore the 2018 publication measures businesses that were active between December 2017 and December 2018. Being active means that the business had either turnover or employment at any time during the reference year. The file of active businesses is compared with previous and subsequent files in order to identify business births and deaths and up to five-year survival rates.

This is different to the related UK business; activity, size and location publication, which is a snapshot of a point in time on the IDBR.

The data have been produced in response to a European Union regulation, and there are comparable data across European Union members. These data are a useful indicator of entrepreneurship and economic growth. For the ONS to publish Business Demography estimates within a year of the reference period (i.e. publish 2018 data before the end of 2019), ONS make an adjustment to the business deaths figures to allow for reactivations.

The Eurostat and OECD manual on business demography recommends waiting for two years after the reference period to allow for reactivations before business deaths figures are calculated. In the release, ONS estimates the number of reactivations and adjust the data accordingly. The adjustment has been applied to all industries, by removing units from the business deaths data. This can lead to different percentage adjustments at the lowest level of aggregation. In order to account for reactivations, the business deaths data are not final until the third year. The preceding two years of business deaths are considered to be provisional and subject to revision. Table 9 of the ONS Business Demography dataset shows the adjustments made to the business deaths data for reactivations.

2.4 Strengths and Limitations

The comprehensive administrative sources combined with the survey data that contribute to the IDBR, ensure a wide coverage, which is one of its main strengths.

The IDBR provides comprehensive coverage of businesses registered for VAT and/or PAYE, but there exists a sizeable population of low turnover non-employing businesses that the IDBR does not include. These businesses are not included in the ONS Business Demography publication (and, by extension, the DCMS Business Births and Deaths publication). It is currently estimated that in the UK, there are 2.7 million registered businesses and 3.2 million unregistered businesses. The unregistered businesses are very small in terms of turnover, and the vast majority of UK economic activity is captured through the 2.7 million registered businesses that are on the IDBR.

The Department for Business, Energy and Industrial Strategy (BEIS) produces a publication called Business Population Estimates, which provides an estimate of the total UK business population. (It includes an estimate of the “unregistered” population). However, this does not directly include data on business births and deaths.

3. Quality assurance processes at DCMS

The majority of the quality assurance of the data underpinning the DCMS Sectors Economic Estimates Business Births and Deaths release takes place at ONS, through the processes described previously. Once the ONS checks are complete, the data are sent via secure transfer to DCMS, who apply further quality assurance measures.

Production of the analysis and report is typically carried out by one member of staff, whilst quality assurance is completed by at least one other, to ensure an independent evaluation of the work.

3.1 Data requirements

DCMS discussed our data requirements with ONS and these are formalised as a Data Access Agreement (DAA). The DAA covers which data are required, the purpose of the data, and the conditions under which ONS provide the data. Discussions of requirements and purpose with ONS improved the understanding of the data at DCMS, helping us to ensure we receive the correct data and use it appropriately.

3.2 Checking of the data delivery

DCMS check that the data delivered by ONS match what is listed in the Data Access Agreement (DAA). For this particular release we check that:

• We have received all data at the 4 digit SIC code level, which is required for us to aggregate up to produce estimates for our sectors.
• Data is available for all relevant sectors and parameters (births, deaths, regions etc.), for all years from 2010 to 2018 inclusive.

3.3 Data analysis

At the analysis stage, data are aggregated up to produce information about DCMS sectors and sub-sectors, plus the Audio Visual sector. The Business Births and Deaths lead builds in the following checks at this stage:

• Checks that the sector calculations have been correctly applied and have been correctly copy-pasted into the relevant “to-be-published” tables
• Checks that the births and deaths data are not mixed up, and ditto the different sectors.
• “Sense checks” of the data. E.g.:
  • Do the trend figures in business births and deaths look sensible? How do they compare to the number of businesses in the UK economy as a whole (like rapid growth/decline)?
  • How do the business births and deaths figures for a given year (e.g. 2018) compare with trend figures in the number of UK businesses (e.g. change from 2017 to 2018)?

3.4 Quality assurance of data analysis

Once analysis is complete, the producer hands over to the quality assurers to carry out further checks of the analytical work completed. Suggested checks are recommended by the producer in the form of a document called a quality assurance (QA) log. Within the document, the suggested checks are listed and described, and there is space for the quality assurers to indicate the outcome of the checks and give any other feedback. After the publication, the quality assurance processes are reviewed to ensure they are relevant and comprehensive. The checks listed in the QA log cover:

• Checks that the SIC codes have been correctly used and correctly allocated to each sector/sub-sector.
• Check that the data has been correctly pulled into the relevant sector and sub-sector columns.
• Check that the calculations have been correctly carried out for each SIC and, correspondingly, for each sector/sub-sector.
• Check that the figures have been rounded correctly to the nearest five, as agreed in guidance correspondence with the ONS for disclosure control purposes, and that this has been consistently applied.
• Some sense-checking of the figures, for example:
  • That the Crafts sub-sector estimate is the same for the Culture and Creative Industries for each year
  • That the Telecoms sector estimate and the Telecoms sub-sector estimate within the Digital Sector are the same for each year
• Checking that the figures and charts used in the report and submission are accurately taken from the correct underlying tables

3.5 Dissemination

Finalised figures are disseminated within Excel tables and a written report (which includes written text, graphs, tables and infographics) published on GOV.UK. These are produced by the Business Births and Deaths lead. Before publishing, a quality assurer checks the figures match between the working-level analysis, the tables and the written report. The quality assurer also makes sure any statements made about the figures (e.g. regarding trends) are correct according to the analysis.

4. Next steps

We encourage our users to engage with us so that we can improve our statistics and the documentation surrounding them. If you would like to comment on this first report on quality assurance processes, or have any enquiries please get in touch at evidence@dcms.gov.uk.