UK Manufacturers’ Sales by Product (PRODCOM): 2011 Provisional Results

In 2011 a review of the methodology for producing standard errors for PRODCOM identified that several changes needed to be made to the underlying calculation. The effect of this is that the standard errors (and coefficients of variation) are now slightly higher than they were previously for some of the product headings. It is therefore important to note that any increases to the quality measures (contained within the accompanying tables) compared with previous years quality measures do not necessarily represent a decrease in the quality of the estimates. It should also be noted that standard errors prior to 2009 have not been recalculated.

An extensive revision of NACE in 2007 resulted in changes to PRODCOM estimates for the 2008 survey onwards. All industry sectors now align exactly to the NACE classification. This differs from previous years where, in some instances, the UK published PRODCOM estimates under its own Standard Industrial Classification (SIC) which only matched the NACE classification to the first four digits.

This bulletin provides information on the value of UK product sales. The accompanying tables relating to this report contain sales estimates for UK based manufacturers of individual products covered by the PRODCOM survey. Most also contain the volume of production and imports and exports estimates supplied by HM Revenue and Customs (HMRC).

Data in the tables relating to merchanted goods, work done, waste and all other income, assist in the PRODCOM balancing process and business industrial classification.

Each divisional report contains data for the latest three years and is presented by manufacturing division. Each division is further broken down into product group then by individual products. The latest data are available from the ONS website. Due to the switch to SIC 07 in 2008 and the change in name and format of older publications, data pre-2008 are available on request from prodcompublications@ons.gov.uk.

The PRODCOM survey is conducted across all EU member states to allow comparison with other EU countries and to aid analysis in cases of anti-dumping. It is also used as a proxy in environmental statistics.

PRODCOM outputs are required as part of the National Accounts Supply Table, an integral part of the annual Supply and Use Tables balancing process which is used to reconcile the three approaches to measuring Gross Domestic Product (GDP) and Gross National Income (GNI).

PRODCOM survey information is used to create a sampling frame for the Producer Prices Index, having identified businesses that make particular products.

By combining PRODCOM with overseas trade data, users can derive various other statistics. For example: the trade balance, the UK net supply to the market and unit prices for production, imports and exports; all at the product level.

A Quality and Methodology Information Report (QMI) Report, replacing the previous Survey Quality Report describes, in detail, the intended uses of the statistics presented in this publication, their general quality and the methods used to produce them.

A report on PRODCOM EU methodology is also available from the Eurostat website

Intra and Extra EU Import and Export data is that of HMRC and is not collected by ONS. It is matched with the PRODCOM codes and included within the PRODCOM tables for the benefit of demonstrating the trade balance and UK net supply. Data are available monthly, quarterly and annually from HMRC’s Trade Statistics and vary as HMRC data collected uses a different codification.

The sum of constituent items in tables may not always agree exactly with the totals shown due to rounding.

The following symbols are used in the accompanying tables:

S Suppressed as disclosive

S* Suppressed as disclosive but included in the aggregated for UK Manufacturer Sales of “Other” products aggregated for UK in the Sales and Turnover table

N/A Data not available

* Not able to provide data due to secondary disclosure, different units of measurement or other technical issues.

Product information is available from the following Manufacturing Divisions

The latest annual EU comparative data for PRODCOM are available from the Eurostat website

The PRODCOM list is updated annually to allow for the addition and deletion of products as the market evolves; clarification of product definitions and corrections where identified. To assist our users, these changes will be documented and incorporated within the December Statistical Bulletin each year.

The survey sample is derived from ONS’ Inter-Departmental Business register. Businesses are selected by employment size and industry classification. Additional constraints are built into the sample design with a view to restricting the form filling burden on smaller businesses. Further details on sample selection are included in the QMI.

Further to the UK PRODCOM QMI, Eurostat also produce an annual PRODCOM Quality Report containing comprehensive quality information based on the latest data received from all EU members.

An EU glossary of terms can be used to interpret the technical descriptions and abbreviations used throughout the report.

The following information is available for each PRODCOM Commodity Code

• an estimate of standard error

• standard error as a percentage of the published value (the coefficient of variation)

• number of businesses providing data at the product level

• total employment of businesses providing data for the product

Standard errors are an estimate of the sampling error, which arises because an estimate is based on a survey rather than a population census. It is a measure of the precision of the estimate. A low standard error therefore indicates a precise estimate. To aid comparison and interpretation, the standard error is also expressed as a percentage of the product’s estimated total sales. This quantity is sometimes called the coefficient of variation and it allows the standard errors to be put into context.

Standard error estimates are available for most product level value estimates. The latest data are available in the accompanying publication tables.

Following a review in 2011, it was discovered that the code used to calculate the PRODCOM standard errors was incorrect. A change was implemented in the last set of results published January 2012. An article detailing the change will be published in July 2012.

This is the first (Provisional) set of data for the 2011 period and at this time no revisions are made to previous periods. A second set (Intermediate) for 2011 and a third (Final) set of data for 2010 will be published on 20 December 2012.

Provisional survey estimates are published six months following the end of the reference period; intermediate estimates 12 months after the end of the reference period and a final set of estimates 18 months following the reference period.

Revision information relating to 2011 and the two previous periods will be incorporated within the December 2012 publication.

The response rates reflect the response rates at the time of publication. Late and revised returns are included in the intermediate and final estimates.

Data collected by the Annual Minerals Raised Inquiry (AMRI) on mineral extraction are used in the PRODCOM survey to avoid duplication and to reduce the burden on business.

Intra and Extra EU Import and Export data are supplied by HMRC to measure the flow of goods into and out of the UK. These data are also used as part of the UK’s Balance of Payment Account to gauge the health of the UK. A recent correction to product data in the 2009 and 2010 trade data was announced by HMRC. A full copy of the report is available from the HMRC website. The two PRODCOM divisions affected by the changes were amended and republished on the ONS website.

The Index of Production (IOP) collects total production turnover but not data about individual products manufactured. There are, however, a significant number of respondents who return only one product figure to PRODCOM and it is, in theory, possible that this information could be collected through IOP for PRODCOM.

The Annual Business Survey (ABS) also collects total turnover plus other variables similar to the standard headings collected by PRODCOM. The feasibility therefore exists for the ABS to collect standard headings on behalf of PRODCOM. There are differences however which would create issues. For example ABS collection captures financial year-end whereas PRODCOM collects a calendar year; therefore PRODCOM would not be able meet Eurostat regulations in terms of timeliness. There would also be considerable differences in sales values for individual product headings in the smaller size-band because the ABS collect data on total sales of own manufactured products, but minimal collection on sales of individual manufactured products.

Statistical disclosure control methodology is applied to the PRODCOM estimates to ensure that information attributable to an individual business is not disclosed in any publication. However, to provide the most comprehensive dataset, ONS seek permission to publish data from businesses whose sales and/or volume data are deemed disclosive. Where permissions are granted the data is included in the final tables. The Code of Practice for Official Statistics sets out principles for how ONS protects data from being disclosed.

The recently launched ONS User Guide provides information on a range of business surveys, including Prodcom.

The scope, coverage and requirement for the PRODCOM survey are detailed in Eurostat’s PRODCOM User Guide. The interactive guide helps find statistics on UK business published by the Office for National Statistics. It covers statistics regarding the structure, content and performance of businesses across the UK, which includes measures of turnover and other financial transaction variables, employment, economic growth and productivity.

HMRC link to information on the collection of import and export data

As part of a package of measures to improve the quality of the PRODCOM results, the current methodology and systems used in the production of the data are under review. Details of work undertaken will be published in separate articles during 2012.

Planned articles in 2012 relate to

• an amendment to the method of calculation of PRODCOM Standard Errors

• the possibility of a bi-annual sample for the Prodcom Survey

• introducing selective editing within the Prodcom Survey to reduce burden on business

Users will be informed of planned work and progress through future statistical bulletins and accompanying articles. Users are encouraged to feedback any comments relating to the PRODCOM output using the email address below.

Alternatively users can contribute to varying discussions through the StatsUserNet, a community forum promoted by the Royal Statistical Society. The Business Statistics community, in particular, covers statistics about the structure, content and performance of businesses across the UK, which includes measures of turnover and other financial transaction variables, employment, economic growth and productivity.

Details of the policy governing the release of new estimates are available from the ONS press office.

National statistics are produced to high professional standards set out in the Code of Practice for Official Statistics. They undergo regular quality assurance reviews to ensure that they meet customer needs. They are produced free from any political interference.

Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: media.relations@ons.gov.uk