UK Data Archive Study Number 6644 - Annual Respondents Database, 1973 - 2008: Secure Data Service Acccess
Quick Guide for Users
Dataset:
Dates available:
Source:
Coverage:
Collected by:
Link fields:
Legal restrictions:
Annual Respondents Database (ARD)
1973-2002, annual (only from 1993 for construction and 1997 for services)
Large business census, small business rotating sample (see below)
Turnover, cost, employment, industry, investment
ONS
IDBR fields (Local/reporting units, enterprise/enterprise group refs.)
Statistics of Trade Act
Quick summary:
The ARD is formed from the Annual Business Inquiry (ABI) from 1998 onwards and previous surveys before
that (e.g. Annual Census of Production (ACOP)). It is a census of large businesses, and a sample of smaller ones.
The ARD is created for the Economic Analysis and Satellite Accounts Division for research purposes. The ABI
(and previous source surveys) is used by the Sources Directorate to produce industrial statistics and is used for a
number of other purposes including generation of the National Accounts. Northern Ireland data is held up to
2001. From 2002, the ABI is collected and stored separately in Northern Ireland. Special permission is required
to use new NI ABI data.
The ARD is constructed from a compulsory business survey. Until 1997 it was created out of the ACOP and
ACOC (Annual Censuses of Production and Construction); these were combined into the ABI (Annual Business
Inquiry) in 1998. To create the ARD, the other surveys are converted into a single consistent format linked by
the IDBR references over time.
The data prior to 1998 cover the vast majority of production and construction activities (construction from 1993
only), but from 1997 the ABI also incorporates six other previous surveys covering distribution and other service
activities. Hence from 1997 data on services are stored on the ARD. This increased coverage is reflected in the
number of individual business contributors to the ARD rising from approximately 15,000 for 1980 to 1996 to
approximately 50,000 for 1997/98 and to over 70,000 for 1999. The ONS is currently exploring the possibility of
retrospectively integrating archive data from past service surveys into the ARD.
The businesses selected for the surveys have been drawn since 1994 from the ONS Inter-departmental Business
Register (IDBR). The IDBR covers about 98% of business activity (by turnover) in Great Britain. Each year a
stratified sample is drawn for the ABI and thus the data stored on the ARD is from business respondents
returning the questionnaires that are sent out by the ONS.
The IDBR also records data from the administrative sources of VAT and PAYE records for all the 3.7m or so
businesses. These data relate to the name, location, birth, turnover and employment of the business. For the
sectors covered by ACOP/ABI most of these administrative data are also stored on the ARD in a supplementary
file. This allows the whole population to be taken into account when using the data.
The ABI is collected in two parts: ABI(1) is an employment record, collected as soon as possible after 12th
December. ABI(2) is financial information, which may be submitted up to twelve months after the financial year
end.
Smaller firms may receive a "short form". These do not require detailed breakdowns of totals. Hence for certain
variables the values may be imputed from third party sources or estimated rather than returned by respondents.
Major variables: employment, turnover/output, capital expenditure, intermediate consumption, gross value added
(derived), postcodes, industrial classification, owner nationality, acquisitions and disposals of capital goods
See IDBR quick guide for more information on IDBR unit structure.
Sampling frame:
The selected sample of small firms rotates to prevent the excessive sampling of SMEs. The sample
frame has varied from year to year:
For specific details and numbers, see the file ard_statistics.xls.
Organisation of files:
The ARD folder contains all the GB ARD data in a flat listing. For all years between 1973-2001, there exists
•
•
•
datyyyypdg.dta
nulyyyypdg.dta
snulyyyypdg.dta
selected reporting unit observations, production sector, for year yyyy
non-selected reporting units
all local units (including single units where LU=RU)
From 1994 to 1996, construction is available:
•
[dat|nul|snul]yyyycng.dta
construction sector
and from 1997 onwards all sectors:
•
•
•
•
•
•
[dat|nul|snul]yyyycag.dta
[dat|nul|snul]yyyymtg.dta
[dat|nul|snul]yyyyprg.dta
[dat|nul|snul]yyyyreg.dta
[dat|nul|snul]yyyystg.dta
[dat|nul|snul]yyyywhg.dta
catering
motor trade
property
retail trade
other services
wholesale
The “g” in the file name refers to GB. NI data is kept separately. The eight sector codes (pd, cn, ca, mt, pr, re, st,
wh) are used throughout the ARD and its derivatives.
Known data issues:
Obvious general inconsistencies should have been dealt with in the cleaning process. The Standard Variables set
of files provides an insight into particular inconsistencies or changes to popular variables.
A number of variables (particularly since the late 1990s) now have to be calculated rather than read. Thus the
more recent data looks as though it has more but different variables compared to earlier years; however, in
practice it merely has more variables, with the missing variables needing to be calculated.
E-commerce questions: variables should be 1/2 dummies but in fact contain a variety of numbers. No
explanation from Newport – probably miscoding.
There is a change in measuring units from 1992 to 1993 onwards. Variables in 1993 and after are measured in
1000’s of units compared to 1992 and before.
There are occasional problems with the register employment data (empment from 1997 onwards; previously,
sel_emp). The non-selected reporting unit files before 1980 contain no employment data (sel_emp). From 1998
onwards, there are a small number of cases were selected employment is zero, despite returned employment
being significantly positive.
Other issues
1970-1972 data is inconsistent with later years and so is only available in source files. If these are required, BDL
will be happy to decompress the data. Similarly, services data from 94-96 is potentially available.
The ARD now incorporates some financial firms too within the “other sectors” files (code st). There are around a
thousand reporting units (selected and non-selected) in 2001, and small numbers in 2000 and 1999, with sic
codes 65-67.
(ons)
ANNUAL RESPONDENTS DATABASE
USER GUIDE
Version 1.2.1
April 2002
Editor: Matthew Barnes
Business Data Linking Branch
Economic Analysis & Satellite Accounts Division
Office for National Statistics
1 Drummond Gate
London SW1V 2QQ
[email protected]
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Sections of this document are based on [refs].
[check and edit refs]
2
ANNUAL RESPONDENTS DATABASE: USER GUIDE
INTRODUCTION
5
1 BACKGROUND TO THE ANNUAL RESPONDENTS DATABASE
7
1.1
ACCESS, CONFIDENTIALITY AND DISCLOSURE
1.2
BUSINESS REGISTERS
1.2.1
BEFORE 1994: THE CSO BUSINESS REGISTER
1.2.2
AFTER 1994: THE 1DBR
1.2.3
SUMMARY STATISTICS FOR THE IDBR
1.3
UNDERLYING SURVEYS
1.3.1
ANNUAL CENSUS OF PRODUCTION (ACOP)
1.3.2
ANNUAL CENSUS OF CONSTRUCTION (ACOC)
1.3.3
ANNUAL BUSINESS INQUIRY (ABI)
1.4
ORGANISATION OF THE ARD
1.5
NORTHERN IRELAND DATA
1.6
THE RANGE OF VARIABLES
1.7
PROTECTING BUSINESSES FROM STATISTICAL BURDENS
1.8
SAMPLE STRATIFICATION
1.9
SUMMARY OF THE ARD
1.10 THE ARD IN USE
1.10.1
PRODUCTIVITY AND FOREIGN FIRMS
1.10.2
THE VARIATION OF PRODUCTIVITY BETWEEN BUSINESSES
1.10.3
PRODUCTIVITY, ENTRY AND EXIT.
1.10.4
LINKING THE ARD WITH DIFFERENT SURVEYS.
1.10.5
OTHER WORK USING THE ARD
7
7
8
9
15
16
17
18
18
18
20
20
20
21
23
24
24
27
29
30
30
2 TECHNICAL INFORMATION
33
2.1
2.2
2.3
2.4
33
37
39
40
CLEANING THE DATA
PROBLEMS USING THE DATA
GROSSING THE DATA
REGISTER CONTINUITY
3 DETAILED APPENDICES
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.7.1
3.8
3.9
3.10
3.11
3.12
3.13
VARIABLES AND DERIVED VARIABLES
COVERAGE FOR ABI
DISCLOSURE RULES
EXAMPLE FORMS
OTHER LINKABLE SURVEYS
DEFLATORS
EARLY NOTES ON THE ARD
ANNUAL RESPONDENTS DATABASE
DATA CLEANING
CLEARING RESULTS FOR RELEASE
SIC CODES?
ARD CONTACT DETAILS
COUNTRY CODES
REGION CODES
41
43
45
49
51
53
55
57
57
93
95
97
99
101
107
3
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.14
QUESTION LISTS
REFERENCES
4
109
121
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Introduction
This document introduces the Office for National Statistics Annual Respondents
Database1. This guide:
•
Introduces the Annual Respondents Database (ARD)
•
Provides background information on the data that form the ARD
•
Shows the coverage of the ARD
•
Illustrates work the ARD has been used for
•
Gives technical information necessary to use the data
•
Includes detailed appendices for reference
The remaining pages are divided into three sections. Section 1 provides
background to the ARD. Section 2 gives technical information and section 3
contains detailed appendices for reference.
1
Previously also known as the ACOP Respondents Database and ABI Respondents
Database.
5
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1 Background to the Annual Respondents Database
The Annual Respondents Database (ARD) is constructed from the micro-data
resulting from annual business surveys conducted by the Office for National
Statistics. It allows individual reporting units to be followed through time and is
therefore longitudinal. In this document we describe the ARD in more detail. In
particular summarising the work necessary to use the data, restrictions on the
use of the ARD and some of the work completed to date using this resource.
[add some more history?]
1.1
Access, Confidentiality and Disclosure
The data contained in the ARD are collected under the 1947 Statistics of Trade
Act. Under this Act all data collected from businesses are confidential and remain
confidential forever. This confidentiality includes non-ONS employees even
seeing individual observations in the data. Access for external researchers is
possible under the 1994 Deregulation and Contracting Out Act. Researchers are
contracted to ONS to conduct a specified task. Only researchers with contracts
are permitted to use the data.
At present access to the complete data set is only permitted at ONS premises at
Drummond Gate in London and Newport in South Wales, subject to available
resources. Access to the ARD is available to researchers only under specific
conditions to ensure confidentiality requirements in the Statistics of Trade Act are
met. Access is only given to researchers who are sponsored by a government
department and where the research to be undertaken is consistent with the aims
and objectives of ONS. Inquiries should be made by email to
[email protected]
In order to comply with the terms of the 1947 Act it is necessary for all work for
circulation outside of ONS and contracted researchers must be cleared by the
data custodians at ONS to ensure it is not disclosive of data about an individual
business. The rules for determining what can or cannot be disclosed have two
elements2. First is the threshold rule, which ensures there are at least 3
enterprise groups in a cell. Second, the dominance rule states that the sum of all
but the largest two values in a cell must be greater than 10 percent of the largest
observation3. Only if both of these rules are passed can work be cleared for
release. The complete disclosure rules are contained in Appendix C. It is
important to note that clearance must be sought for all forms of release including
presentations and draft papers as well as final publications. [refer to penalties?]
[SoTA in an appendix?]
1.2
Business Registers
The ONS business surveys that together form the ARD are sampled from the
ONS business register current at the time of drawing the sample. This section
2
For a background to disclosure issues see GSS (1995)
3
For example, if presenting employment numbers and the largest observation in a cell
has employment of 1000, then the sum of the employment of all the observations in the
cell except the next largest must be greater than 100 to pass the dominance rule.
7
ANNUAL RESPONDENTS DATABASE: USER GUIDE
describes the two registers that have been in use over the years covered by the
ARD. First the CSO Business Register (previously known as the BSO Business
Register) that ran until 1993. Then the Inter Departmental Business Register
(IDBR) that has run from 1994 onwards.
1.2.1 Before 1994: The CSO Business Register
The CSO Business Register (Perry 1985) ran until 1993 when it was replaced by
the IDBR (see below). The main statistical unit is the `establishment’, which is the
smallest unit for which a full range of statistics is available. Establishments
comprise one or more local units (factories or sites). Prior to 1984, it was based
on numerous sources but no one consistent source to ensure completeness and
prevent duplication. From 1984 the VAT register was used as the main source for
the register. Previously VAT data had been used as the basis for the register for
the distribution and service trades inquiries. However VAT as a source does not
provide employment information. This was instead taken from responses to
surveys in previous years or imputed from the VAT turnover. For larger units
proving forms were also sent to confirm both the local unit structure of the
company and acquire the employment data necessary for sampling. To identify
local units and their employment from VAT registrations, new registrations were
also sent proving letters. In addition local units were matched to their legal units
by matching their names and then to mail unmatched units after this. A major
problem with this register was seen to be that its employment information could
in some cases be quite old. For many units there was no systematic updating of
this data.
Inquiry cut-offs generally of 20 or more employees mean that a certain amount of
dead wood remained on the register at the lower end. Equally there was no
systematic way of identifying new businesses, especially those that are very
small. The units in the CSO Business Register can be summarise as, in
descending level of aggregation:
•
‘Enterprise Group’ – the group of all legal units under common control.
•
‘Establishment’ – the smallest group of legal units which could provide the full
range of data required for the survey
•
‘Local unit’ – the individual site or workplace (factory, shop etc) at which
activity takes place
These have also had alternative labels applied to them, depending on their
structure. Units that are part of an enterprise group are often termed ‘multiple’
establishment firms (or multiples) and those that are not `singles’. An
establishment that has no dependent local units (i.e. only reports on itself) is also
a local unit and may be termed a `single’. Establishments that report on other
local units in addition to themselves are often termed ‘parents’, while their
dependent local units are termed `children’.
In 1987, a move was made to `company reporting’. A major motivation of this
was to comply with European Communities’ Directives. In its pure form this would
have meant companies replaced all establishments as units for surveys.
However to ensure continuity of results, large mixed-activity and/or
geographically spread companies continued to be split by activity and/or region.
This compromise became known as the ‘reporting unit’ level. In practice most
8
ANNUAL RESPONDENTS DATABASE: USER GUIDE
businesses reported for the company as a whole both before and after the
change. There was therefore little difference to the main economic series as a
result. Additionally any effects of this change on results using the micro data
appear to be minimal.
In addition to the introduction of the VAT-based register and the move to
company reporting outlined above, the CSO register was subject to continuous
improvement that could be an issue for users of data based on it. In particular
there were a number of improvements in the lead up to the introduction of the
IDBR.
1.2.2 After 1994: The IDBR
The IDBR (Perry 1995) was developed in the mid-1990s to replace two separate
registers: the VAT-based register of the former Central Statistical Office and the
PAYE-based register of the Employment Department. The main drivers for
developing the IDBR were the need to improve coverage and to improve the
coherence of estimates produced from surveys that had previously used the two
registers as separate frames. In particular, the derivation of productivity and unit
wage costs would be possible using statistics generated on a more consistent
basis.
One perceived benefit was that the Register would be more complete if it used
two distinct administrative data sources. The IDBR is based on data from two
main sources: HM Customs and Excise (HMCE), which registers traders for
Value Added Tax (VAT) purposes; and Inland Revenue (IR), which registers
employers who operate Pay As You Earn (PAYE) income tax schemes. In
addition, the IDBR is updated from register quality surveys, statistical surveys
and other sources.
Information from HMCE is transferred to ONS under the VAT Act 1994, and
information from IR, under the Finance Act 1969. Both Acts, and the Statistics of
Trade Act 1947, which covers data collected directly by ONS, restrict the release
of data outside ONS.
The IDBR covers all businesses registered for VAT or PAYE. Businesses in the
agricultural sector are recorded on the IDBR, although the Department for
Environment, Food and Rural Affairs (DEFRA) (formerly the Ministry of
Agriculture, Fisheries and Food (MAFF)) is responsible for collecting agricultural
statistics. Construction businesses are also on the IDBR, and both ONS and the
Department for Transport, Local Government and the Regions (DTLR) (formerly
the Department of the Environment, Transport and the Regions (DETR)) publish
data relating to the construction sector.
Third, the IDBR covers all UK businesses4. However, separate legislation covers
the British and Northern Irish parts of the IDBR. ONS is responsible for
maintaining the British part of the Register, while the Department of Enterprise,
Trade and Investment Northern Ireland (DETINI) deals with the Northern Irish
part. Consequently, some procedures operated by DETINI are different from
4
Where the UK consists of Great Britain (England, Scotland and Wales) and Northern
Ireland only. It does not include the Isle of Man or the Channel Islands.
9
ANNUAL RESPONDENTS DATABASE: USER GUIDE
those in place at ONS. Moreover, while ONS runs an Annual Register Inquiry,
DETINI runs a two-yearly Census of Employment.
1.2.2.i
IDBR Units
IDBR units are classified into three types: administrative units, statistical units
and reporting units.
[diagram]
Administrative Units
The administrative units are the VAT traders and PAYE employers.
Statistical Units
The statistical units are the enterprise, enterprise group, local unit and kind of
activity unit (KAU), all of which are defined precisely in the EU Regulation on
Statistical Units (EEC 696/93) as follows.
Enterprise
‘The enterprise is the smallest combination of legal units that is an
organizational unit producing goods or services, which benefits from a
certain degree of autonomy in decision-making, especially for the
allocation of its current resources. An enterprise carries out one or more
activities at one or more locations. An enterprise may be a sole legal unit.’
Enterprise Group
‘An enterprise group is an association of enterprises bound together by
legal and/or financial links. A group of enterprises can have more than
one decision-making centre, especially for policy on production, sales and
profits. It may centralize certain aspects of financial management and
taxation. It constitutes an economic entity which is empowered to make
choices, particularly concerning the units which it comprises.’
Local Unit
‘The local unit is an enterprise or part thereof (eg a workshop, factory,
warehouse, office, mine or depot) situated in a geographically identified
place. At or from this place economic activity is carried out for which –
save for certain exceptions – one or more persons work (even if only parttime) for one and the same enterprise.’
Kind of Activity Unit
‘The kind of activity unit (KAU) groups all the parts of an enterprise
contributing to the performance of an activity at class level (four digits) of
NACE Rev 1 and corresponds to one or more operational subdivisions of
10
ANNUAL RESPONDENTS DATABASE: USER GUIDE
the enterprise. The enterprise’s information system must be capable of
indicating or calculating for each KAU at least the value of production,
intermediate consumption, manpower costs, the operating surplus and
employment and gross fixed capital formation.’
Reporting Unit
The reporting unit holds the mailing address for the business and is the unit for
which businesses report their survey data to ONS. In general, the reporting unit
is the same as the enterprise. In some of the more complex cases, enterprises
are subdivided into reporting units that correspond to KAUs, and are defined by
specifying the appropriate local units from within an enterprise.
Note that unlike on the CSO business register, local units and reporting units are
distinct on the IDBR. In particular reporting unit is not also a local unit. A
reporting unit and a local unit may be co-located but have distinct identities. The
local units that form a reporting unit have employment that sums to the reporting
unit, there is no residual employment accounted for by the reporting unit itself.
1.2.2.ii
Administrative Sources
VAT Traders
VAT registration is compulsory for traders with an annual turnover above the
‘VAT threshold’ (£52,000 in 2000/01) except in a small number of exempt
industries, such as health and education. Businesses register with HMCE, which
in turn provides data to ONS under the VAT Act 1994. HMCE supplies important
variables to the IDBR, specifically, name, address, postcode, industrial
classification (SIC(92)), turnover and legal status. Data are provided from HMCE
to ONS either weekly, monthly or quarterly.
The weekly data file includes details of all new registrations, VAT traders that
have closed their account with HMCE and changes to existing VAT traders.
Changes to industrial classification, turnover, legal status, address and postcode
for existing traders are also notified.
The monthly file includes members of VAT groups. In some cases, HMCE allows
a group of businesses under common ownership to register together for VAT
payment purposes. One member of the group is nominated to liaise with HMCE,
paying tax due. This member is called the representative member. The other
members of the group are called non-representative members. The monthly
update relates to the non-representatives (representatives are included in the
weekly data).
Every quarter, ONS is provided with details of the most recent year’s turnover for
all VAT-registered businesses except non-representatives.
PAYE Employers
PAYE registration with IR is compulsory for employers operating PAYE schemes.
Every quarter, IR supplies ONS with data under the Finance Act 1969, which
relates to all PAYE schemes. Changes, including births and deaths of employer
11
ANNUAL RESPONDENTS DATABASE: USER GUIDE
schemes, are detected by comparing the files provided by IR with the information
currently on the IDBR. IR supplies a similar range of data to that provided by
HMCE, the main difference being that PAYE gives details of the number of
employees within each PAYE scheme, and not the turnover. The industrial
classification used by IR is the Trade Classification Number (TCN), which is not
aligned to SIC(92).
Companies House
Every quarter, ONS uses Companies House (CH) information to help link
corporate VAT and PAYE records. CH provides information on name, industrial
classification, activity status and company number. The classification is aligned
to SIC(92) at the class level. ONS matches the name against existing corporate
records where the company number is not known by ONS. In addition, ONS has
online access through the Companies House Direct service, which is used to
provide current information to resolve queries.
Dun and Bradstreet
Dun and Bradstreet (D&B) provides data annually to ONS, so that information on
UK companies, their foreign subsidiaries and parents, can be updated at the
enterprise group level on IDBR. This is from D&B’s Worldbase system. In
addition, the ONS uses quarterly updates on CD-ROM and, occasionally, the
D&B online service.
1.2.2.iii
Register Maintenance Sources
Annual Register Inquiry
Apart from the administrative sources, the Annual Register Inquiry (ARI) is the
main vehicle for updating the IDBR. The ARI is designed, together with the
Annual Business Inquiry (ABI), to meet user requirements for the provision of
statistics for small geographical areas.
The objectives of the ARI are to:
•
provide employment data at the local unit level for ABI to calculate
disaggregated employment estimates;
•
maintain the quality of the IDBR;
•
meet the EU Regulation that requires local units to be updated every four
years; and
•
assist IDBR’s proving work.
The survey comprises two parts: ARI/1, which is an annual check of existing
enterprises; and ARI/2, the confirmation of details about the larger new
administrative units, which is carried out every quarter.
The purpose of ARI/1 is to maintain information on business structures, primarily
for those enterprises that operate at more than one location.
12
ANNUAL RESPONDENTS DATABASE: USER GUIDE
ARI/2 is designed to validate the information that has been sent to ONS from the
administrative sources, and to establish the local unit structure for new
enterprises. This process is known as ‘proving’. Proven enterprises on the IDBR
are identified by the existence of explicitly recorded local units on the Register.
Unproven enterprises are assumed to have just a single site.
Both ARI/1 and ARI/2 are for GB businesses only. DETINI carries out continuous
proving for Northern Ireland enterprises. DETINI carries out a census of
employment every two years, covering all employees (apart from in some parts
of the agricultural sector) in Northern Ireland.
Complex Businesses
Some complex businesses are maintained by a team that agrees the most
appropriate structures with the businesses concerned.
1.2.2.iv
Other Sources
Statistical Surveys
Statistical surveys conducted by ONS and other government departments can
update the IDBR in certain situations. In some circumstances, updating is carried
out immediately: the death of a business, name and address changes, and
business restructuring. Updating of industrial classification is annual in batch, but
can be more frequent for individual units. In the case of industrial classification,
those sources thought to provide the best quality data are given higher priority.
Management of the IDBR
The Business Registers Unit (BRU), which is part of ONS’s Prices and Business
Group (PBG), is responsible for managing the IDBR. A staff of around 130 is
responsible for maintaining the Register, including the ARI’s operation, and
provision of analysis for users.
1.2.2.v Uses of the IDBR
The IDBR is used mainly as a statistical tool, but it also has some limited
administrative uses.
Statistical Uses
At present, IDBR has two statistical uses: as a frame for selecting samples and
producing survey estimates, and as a direct analysis tool.
Conducting Surveys
The IDBR is used as a sampling frame and mailing list for official business
surveys. Its specific roles are:
•
to select businesses to be included in the survey samples;
13
ANNUAL RESPONDENTS DATABASE: USER GUIDE
•
as a way of contacting businesses, primarily by mailing questionnaires, and
to provide a receipting and reminder service;
•
to provide information to specify domains of analysis;
•
to provide auxiliary information to enable estimates to be made for
businesses that don’t respond to surveys (imputation), or weren’t asked to
participate in a particular survey (estimation); and
•
to facilitate the monitoring of burden on business and control of overlap
between survey samples.
At present, most surveys run from the IDBR are undertaken for ONS or for
DETINI. Surveys are also carried out on behalf of OGDs. Most surveys are
repeated regularly and are run to satisfy the demands of economic statistics.
One-off surveys are also conducted, primarily for policy-making. Data are usually
collected using mail-back paper questionnaires that are sent out to selected
businesses.
Direct Analysis
In addition to providing samples and population information for surveys, the IDBR
provides an invaluable source of direct analysis. Direct analyses have been
carried out from the UK’s central business register since the early 1980s. These
are annual snapshots, regular business demographic analyses, or ad hoc
analyses. The IDBR is invaluable because it has both the detail and the
immediacy that are lacking in survey sources. On the other hand, the IDBR does
not have the same wide range of variables as surveys, and it can be less
consistent over time than survey sources.
ONS has for many years published a regular annual snapshot of the distribution
of businesses in its Size Analysis of Businesses volume (ONS 2000). In addition,
IDBR data are also used in other ONS publications such as the Focus On series,
Regional Trends and Sector Reviews. Data are increasingly being released on
the National Statistics website.
IDBR data on enterprise demography (Small Business Service 2000a), and
business start-ups and closures (Small Business Service 2000b) are also
published annually by the Small Business Service of the Department of Trade
and Industry (DTI). There is a particular interest in this type of analysis within
Europe.
Increasingly, analyses are being requested by a wide range of users from within
ONS, OGDs, the research community, the wider public and the Statistical Office
of the European Community (Eurostat). Often, such analyses are used to support
the evaluation of changes to government policy. In addition to analysis, individual
data are made available for analysis by OGDs.
Administrative Uses
The IDBR also has some administrative purposes: to allow DTI, National
Assembly for Wales (NAW) and Scottish Executive (SE) to provide a service to
industry; for HMCE to update Standard Industrial Classification (rev 1992)
(SIC(92)) codes; and to assist local authorities in planning.
14
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Future Use
Two new developments require a high quality IDBR. These developments will
drive the way the IDBR is set up and maintained in future.
The first is as a move towards the better use of administrative data, which the
IDBR can facilitate. There are potentially significant benefits, particularly in terms
of cost savings and the reduction of burden on business, of using administrative
data better. Administrative data also provide a more complete coverage of the
economy than surveys. Better use of data from the administrative sources gives
opportunities to eliminate, or substantially reduce, the need to run statistical
surveys. Instead, resources can be used to ensure that the administrative data
can be used as the basis of statistics published by ONS.
The second development is the increased requirement for small area statistics.
This requirement comes from both central government and local authorities.
Good quality small area statistics rely on detailed information about the local unit
structure of businesses. This information is not available from any administrative
source, and can only be obtained from the IDBR.
1.2.3 Summary Statistics for the IDBR
In this section the IDBR’s size and composition are reported. Tables 1-1 to 1-3
provide counts of reporting units within the UK and within the standard survey
populations.
15
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Table 1-1
IDBR Reporting Units by Industrial Sector
Industrial Sector
Number of Units
Agriculture
153,313
Production
169,355
Construction
190,283
Distribution
521,210
Transport
79,849
Finance
27,513
Govt & other services 774,550
Total
1,916,073
Source: IDBR, October 2000
Table 1-2
IDBR Reporting Units by Structure
Unit Type
Proven
Number of Units
655,975
of which:
one local unit
two local units
more than two local units
Unproven
588,505
40,860
26,610
1,260,098
of which:
VAT and PAYE
PAYE only
VAT only
Total
328,720
150,398
780,980
1,916,073
Source: IDBR, October 2000
Table 1-3
IDBR Reporting Units by Legal Status
Legal Status
Company
Sole proprietor
Partnership
Public corporation
Central Government
Local authority
Non-profit-making
body or association
Total
Number of Units
822,811
647,328
390,668
524
499
2,526
51,717
1,916,073
Source: IDBR, October 2000
1.3
Underlying Surveys
The Annual Respondents Database (ARD) stores the data collected by the Office
for National Statistics (formerly the Central Statistical Office) from the Annual
16
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Census of Production and the Annual Census of Construction5. From 1998 these
surveys have been incorporated into and replaced by the Annual Business
Inquiry (ABI) and hence the ABI data are now added to the ARD each year. The
data prior to 1998 cover the vast majority of production and construction activities
(construction from 1992 only), but from 1998 the ABI also incorporates six other
previous surveys covering distribution and other service activities. Hence from
1998 data on services are stored on the ARD. This increased coverage is
reflected in the number of individual business contributors to the ARD rising from
approximately 15,000 for 1980 to 1996 to approximately 50,000 for 1997/98 and
to over 70,000 for 1999. The ONS is currently exploring the possibility of
retrospectively integrating archive data from past service surveys into the ARD.
At the time of writing, final data are available up to 2000. In addition provisional
data for 2001 have been made available, though a number of responses are yet
to be included in the data.
The businesses selected for the surveys are presently and since 1994 drawn
from the ONS Inter-departmental Business Register (IDBR)6. The IDBR covers
about 98% of business activity (by turnover) in Great Britain. Each year a
stratified sample is drawn for the ABI and thus the data stored on the ARD is
from business respondents returning the questionnaires that are sent out by the
ONS. Under the 1947 Statistics of Trade Act it is a legal requirement that
businesses fill these out and return them to the ONS. The IDBR also records
data from the administrative sources of VAT and PAYE records for all the 3.7m
or so businesses. These data relate to the name, location, birth, turnover and
employment of the business. For the sectors covered by ACOP/ABI most of
these administrative data are also stored on the ARD in a supplementary file.
This allows the whole population to be taken into account when using the data.
1.3.1 Annual Census of Production (ACOP)
Data available: 1970 to 1997, annual
Timing: Calendar year. But businesses may report on any 12-month period up to
the end of the financial year. So the 1995 ACOP contains returns covering 1
January 1995 to 31 December 1995 up to 5 April 1995 to 6 April 1996.
Coverage: Production
Other information: Asks for year-average employment.
Replaced by: Annual Business Inquiry from 1998
5
Note that despite their names, ACOP and ACOC were in reality only a census for larger
businesses, and were a stratified random sample for smaller observations. In addition in
many years the very smallest businesses were exempt to reduce their administrative
burden.
6
Prior to 1994 the surveys were drawn from the CSO Business Register. See below for
details.
17
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1.3.2 Annual Census of Construction (ACOC)
Data available: 1992 to 1997, annual
Timing: Calendar year. But businesses may report on any 12-month period up to
the end of the financial year. So the 1995 ACOC contains returns covering 1
January 1995 to 31 December 1995 up to 5 April 1995 to 6 April 1996.
Coverage: Construction
Other information: Asks for year-average employment.
Replaced by: Annual Business Inquiry from 1998
1.3.3 Annual Business Inquiry (ABI)
Data available: 1998 onwards, annual
Timing: Calendar year. But businesses may report on any 12-month period up to
the end of the financial year. So the 1998 ABI contains returns covering 1
January 1998 to 31 December 1998 up to 5 April 1998 to 6 April 1998.
Coverage: Whole market economy with some exceptions (notably financial
intermediation). See appendix 3.2
Other information: Has two parts. ABI/1 covers employment and ABI/2 covers
financial data. ABI/1 replaced the Annual Employment Survey and has wider
coverage than ABI/2. Asks for employment on 12 December.
1.4
Organisation of the ARD
The ARD is a large data set. To make it manageable, it is stored in annual cross
sections, split into a number of groups of data. The present version of the ARD
has the following component files:
18
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Table 1-4
ARD file splits 1970-1997 (Production) and 1992-1997 (Construction)
File
DAT
Description
Contributors who were
selected and returned data
NUL
Contributors who were
Reporting unit
selected but did not return
level data
data & contributors who were
not selected
Contributors who were
Local Unit detail
selected and returned data
SDAT
(See note1)
SNUL
Level of Data
Reporting unit
level data
Contributors who were
Local Unit detail
selected but did not return
data & contributors who were
not selected
Details
Returned question
data and indicative
data
Indicative data
only
Returned question
data and indicative
data
Indicative data
only
Note 1 - This file only exists until 1992 as no local unit information was requested from
1993 on. Only employment and capital expenditure were collected by site.
The DAT file is also known as the ‘selected’ file and NUL as the ‘non-selected’ or
‘universe’ file. Files are also split by sector production (P) or construction (C) and
by their origin: Great Britain (G) and Northern Ireland (N).
Table 1-5
ARD file splits from 1998
File
Description
DAT Contributors who were selected
and returned data
Level of Data
Reporting unit level
data
SND Contributors who were selected
but did not return data
XND Contributors who were not
selected
SNL All local units ,ie selected and
returned; selected but not
returned; not selected. Believed
to cover all industries selected for
ABI 1(ABI 2 covers a sub-set of
ABI 1 industries)
Reporting unit level
data
Reporting unit level
data
Local unit detail
Details
Returned
question data and
indicative data
Indicative data
only
Indicative data
only
Indicative data
only - this is the
Local Unit (site)
information for all
reporting units in
the universe. The
reporting unit
information may
appear in DAT,
SND or XND files.
As previously, files are split by GB (G) and NI (N) and by form (ABI/1 and ABI/2).
ABI/1 employment data is stored in a single file (A!). ABI/2 is split by sector:
19
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Table 1-6
ARD sector splits from 1998
Code
CA
CN
MT
PD
Description
ABI/2 Catering
ABI/2Construction
ABI/2Motor Trades
ABI/2 Production
PR
RE
ST
ABI/2 Property
ABI/2 Retail
ABI/2 Service Trades
WH
ABI/2 Wholesale
1.5
Sectors (SIC(92))
55.1 to 55.5
Section F
50.1 to 50.5
Sections C, D, E plus
group 020, 050 from 2000
70.1 to 70.3
52.1 to 52.7
60.1 to 62.2, 63.1 to 64.2,
71.1 to 74.8, 80.1 to 93.0
51.1 to 51.7
Northern Ireland data
The Statistics of Trade Act only covers Great Britain. In Northern Ireland, surveys
are conducted under local legislation. To date ONS has continued to design the
forms for Northern Ireland, but the sample has often been larger than it would be
using the rules for Great Britain. In addition the survey process is managed by
the Department of Enterprise, Trade and Industry for Northern Ireland (DETINI)
and they own the data. Therefore all work using NI data must be cleared by
DETINI and it is their decision to allow access.
1.6
The range of variables
The range of variables collected since 1970 (unfortunately most of the data
before 1968 was destroyed) on the ARD has varied over the years and the same
variable names can sometimes hide changing definitions or elements included in
questionnaires and derived variables. Records of most of these changes were
kept, see Appendix 3.1 for more details. The central variables collected are
measures of employment, turnover/output, capital expenditure and intermediate
consumption. The data from these direct responses are used to calculate derived
variables such as per head measures and gross value added. Postcodes and
industrial classification (Standard Industrial Classification codes) are included
from the business register along with the nationality of the ‘ultimate owner’ of the
enterprise (as supplied by Dunn and Bradstreet). Although the register lists
business names and full addresses the ARD does not. However, the ARD data
can be linked to the register by using the unique business contributor codes.
Acquisitions and disposals of capital goods7 are recorded but there is no
information on the scrapping of capital stock.
1.7
Protecting businesses from statistical burdens
Osmotherly, compliance limits, etc
7
Plant and machinery, vehicles, and buildings.
20
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1.8
Sample stratification
The Annual Census of Production was not in fact a census other than for the
highest size-band businesses. For smaller size-bands a sample was selected
each year. For most years of the ARD the census size was 100, but this was
raised to 200 for more recent years preceding the ABI. For the ABI this has been
increased to 250. The sample of businesses for the other employment sizebands is drawn from the business register (i.e. the CSO Business Register and
then since 1994 the IDBR) and has changed from year to year. Table 1-7 below
summarises the how the sampling proportions have changed over time. Prior to
1995, the very smallest businesses (generally those employing less than 20
people) were not sampled at all. To limit the impact of ONS surveys on very
small businesses, they do not receive another survey for three years after they
have completed one. Note also that these rules mean the register data on
smaller companies is more dependent on administrative sources than for larger
companies where responses to other surveys are also a source of revision. Note
finally that stratification is also based on industry and region.
21
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Table 1-7
Sampling in ARD source data, 1970-2000
Survey
year
1970-1971
1972-1977
1978-1979
1980-1983
1984
1985-1988
1989
1990-1994
1995-1997
1998
onwards
Employment size
band
<25
Sampling
fraction
0 (exempt)
25 or more
All
<20
20 or more
<20
20-49
50 or more
20 or more
<20
20-49
50-99
100 or more
<20
20-49
50 or more
0 (exempt)
All
0 (exempt)*
0.5
All
All
0 (exempt)
0.25
0.5
All
0 (exempt)
0.5
All
<20
20-49
50-99
100 or more
<20
20-49
50 or more
0 (exempt)
0.25
0.5
All
0 (exempt)
0.5
All
<20
20-49
50-99
100 or more
<10
10-49
50-99
100-199
200 or more
0 (exempt)
0.25**
0.5
All
0.2
0.25
0.5
0.75
All
<10
10-99
100-249
250 or more
0.25
0.5
All or <= 0.5
All
Comments
In some industries,
<11
In some industries 11
was lower limit.
All industries
In 68 industries
In 68 industries
In all other industries
All industries
In most industries
In most industries
All industries
All industries
England only
20 or more outside
England
All industries
In most industries
In most industries
All industries
All industries
England only
20 or more outside
England
All industries
In most industries
In most industries
All industries
50% of industries,
others with smaller
thresholds
Varies by industry
Source: Oulton (1997) and Barnes and Martin (2002)
Note: For 1997 and earlier years these are sampling frames for ACOP. From 1998
onwards they refer to ABI.
*
In 1978 a small sample of establishments employing less than 20 was also
drawn.
**
0.2 in 1993.
The data for each year are essentially a snapshot of those businesses selected
for that year from the register. While the data were not collected or compiled with
22
ANNUAL RESPONDENTS DATABASE: USER GUIDE
the intention of creating a longitudinal database, this possibility is created thanks
to the unique ONS reference number assigned to the reporting units. It is worth
noting that prior to the IDBR the CSO Business Register was subject to
substantial improvements in the 1980s and in the run up to the creation of the
IDBR. These improvements present an additional challenge for researchers
using the data.
1.9
Summary of the ARD
In table 1-9 below, we summarise the data according to the definitions given in
table 1-8 below.
Table 1-8
Definitions of demographic status
Status
Continuer
Entrant
Exitor
One-Year
Definition
Present in t-1, t and t+1
Present in t and t+1, not present in t-1
Present in t-1 and t, not present in t+1
Present in t only
Table 1-9 presents some raw data. We begin in 1980 since that is the earliest
year we can calculate entry, exit and representative employment. Unfortunately
the universe files in the 1970s do not contain any employment data making it
difficult to move from the sample to the population. The number of reporting
units jumps in the second half of the 1990s with the introduction of the IDBR.
The introduction of the VAT-based register in 1984 is also very evident in the
large number of one-year observations that year.
23
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Table 1-9
Number of reporting units and employment by status
Year
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
Number of reporting units
Continuers Entrants Exitors One-Year
102,667
3,476
93,262
3,579
9,405
213
95,000
4,253
1,841
114
42,881
1,506
56,372
710
39,428
9,549
4,959
82,991
46,476
78,217
2,501
15,294
106,994
19,535
17,699
4,260
107,365
20,106
19,164
3,878
112,239
7,132
15,232
20,098
103,480
27,655
15,891
10,917
116,659
16,451
14,476
4,307
96,081
4,094
37,029
10,856
88,550
8,795
11,625
51,403
85,062
41,437
12,283
9,282
104,938
22,699
21,560
8,813
112,477
33,735
15,178
14,122
131,427
17,740
14,785
3,832
133,773
17,128
15,394
4,442
150,901
18,455
Continuers
6,179,728
5,436,345
5,179,553
4,605,333
4,420,175
4,466,677
4,585,082
4,580,044
4,651,077
4,588,500
4,539,950
4,182,156
3,913,568
3,628,195
3,555,510
3,692,887
3,652,517
3,744,431
4,044,728
Employment
Entrants Exitors One-Year
256,103
125,065 192,768
7,941
113,430
82,872
4,087
116,557 516,433
12,980
264,085 258,971 327,451
412,394 135,453
85,102
186,175 225,206
22,301
176,725 270,849
21,943
142,282 233,632
58,402
183,170 273,371
35,745
148,201 236,866
29,320
119,665 333,240
31,618
150,563 264,279 162,188
197,864 292,444
33,962
267,890 258,403 173,571
344,143 316,611 138,671
188,032 352,182
73,645
280,922 273,649
65,950
220,612
Source: Author’s calculations using the ARD
Note: Data for manufacturing on ISIC rev. 3 (UK SIC 1992) definition only.
1.10 The ARD in Use
[edit this]
There are two broad ways of using the ARD. First, cross-sections of the data
may be compared either within a single time period or over time. Second, the
individual identifiers for the businesses can be used to follow each business
through time, creating a longitudinally linked database. Such linking is subject to
the various difficulties outlined in section 1. Examples of both methods are given
below organised under different themes of research that has been undertaken.
The results in this section relate to labour productivity only, there is also on-going
work on the more problematic measurement of Total Factor Productivity.
1.10.1 Productivity and foreign firms
The ARD has variables for gross value added and employment and a field
showing nationality of ultimate ownership. It is therefore possible to compare
labour productivity by country of ownership. Indeed, GVA per head by nationality
of ownership has been published for total manufacturing for many years (see e.g.
ONS 1999 for recent data) but this does allow detailed comparison of particular
firms.
24
ANNUAL RESPONDENTS DATABASE: USER GUIDE
One way of examining whether foreign firms are more productive than domestic
ones is to compare domestic and foreign enterprises which are as alike as
possible. As a simple way to do this, the businesses in the 1998 ABI were
tabulated into 5-digit SIC activities and employment size-bands (i.e. 0-9, 10-19,
20-49, 50-99, 100-249 and 250+)8. UK and foreign owned firms present in the
same 5-digit and employment cell can be compared. Using the ARD, foreignowed businesses were split into their countries of origin and, for each cell,
average foreign and UK productivity were calculated. Table D1 shows, for each
country and size band, the number of cells for which foreign productivity exceeds
UK productivity or vice versa.
With the close run exceptions of non-EU and Australia and New Zealand, the
results indicate that a non-UK owned firm is on average more likely to have a
higher productivity than a UK-owned matched competitor, but this is not true for
every business. The results of this simple tabulation for 1998 are consistent with
results from regression analysis using the ARD for the 1980s, see e.g. Griffith
(1999b), Oulton (1998). Griffith notes “Looking across all manufacturing
establishments we see that foreign-owned establishments are larger and
produce more per worker than domestic-owned establishments.”
Whilst therefore foreign businesses are generally more productive than their UKowned counterparts, this is not always the case. For example, USA-owned
enterprises had the third highest figure for average GVA per head according to
1997 ONS figures (ONS, 1999) and the highest percentage margin over the UK
out of France, Germany and Japan in 1992 in Griffith’s analysis. The USA also
scores the highest differential over UK competitors in Table D1, but even here
there are over 1/3 of cell-pairs where UK owned firms were outperforming their
USA owned close rivals in 1998. This begs the question of why some UK firms
do better than other UK firms and whether or not this is because they possess
characteristics associated with USA-owned businesses.
8 For reasons of disclosure control the smaller size bands are combined in the table.
25
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Table D1
UK and foreign ownership ‘like with like’ comparisons in 1998
Employment Size Band
0-99
66
123
100-249
45
77
250+
67
110
Total
178
310
1.74
UK > Germany
UK < Germany
Ratio: Germany ‘better’/UK ‘better’
25
57
19
23
32
39
76
119
1.57
UK > Canada
UK < Canada
Ratio: 'Canada ‘better’/UK ‘better’
4
13
5
8
13
11
22
32
1.45
UK > East & South East Asia
UK < East & South East Asia
Ratio: E&SE Asia ‘better’/UK ‘better’
26
42
21
22
23
24
70
88
1.26
UK > EU exc. UK & Germany
UK < EU exc. UK & Germany
Ratio: EU exc. ‘better’/ UK ‘better’
86
131
65
65
78
75
229
271
1.18
UK > Australasia (Australia & NZ)
UK < Australasia (Australia & NZ)
Ratio: Australasia ‘better’/ UK
‘better’
10
9
7
4
3
7
20
20
1.00
UK > Non EU
UK < Non EU
Ratio: Non EU ‘better’/UK ‘better’
50
51
26
27
33
30
109
108
0.99
UK > USA
UK < USA
Ratio: USA ‘better’/UK ‘better’
Source: Business Data Linking Project, ONS.
Notes: Results obtained using 1998 ABI data. Columns 2 to 4 show the number of cell
pairs for which the condition in column 1 is met. A cell pair exists only where a UK and a
foreign firm are both present in a particular 5-digit SIC industry and size combination.
Results for smaller firms have been combined into and employment size band of 0-99 for
disclosure reasons. Comparisons are based on total GVA per head. Column 5 shows the
totals of columns 2 to 4 and the ratio of the UK ‘better’.
The above analysis could easily be repeated in cross-section for other years, but
allowance would have to be made for sample stratification. Of course, the above
illustration would be more telling if it used observations for a longitudinally linked
set of businesses so that relative performance over a period could be compared
rather than simply looking at a snap shot. This can be done but it requires
dealing with the problems of longitudinal linking mentioned above.
Further work on foreign ownership using the ARD can be seen in Griffith (1999a)
and Oulton (2000). These papers compare ownership productivity estimating
production functions and controlling for inputs. They generally find higher
average labour productivity in foreign businesses but only slightly higher average
foreign total factor productivity.
26
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1.10.2 The variation of productivity between businesses
The ARD can be used to compare distributions of business related variables over
time. As it is only investigating the features of these distributions such work does
not require longitudinally linking particular businesses. As an example, Barnes
and Haskel (2000b) apply price and cost deflators to ARD longitudinally linked
contributor data to reveal a wide distribution of productivity among UK
businesses. Figure D1 shows the wide spread in the productive performance of
plants in UK manufacturing over the period 1994 to 1997. Plants at the 90th
percentile are around 5 times as productive as the plants at the 10th percentile.
Equally, within individual sectors these large variations in productivity persist with
the best plants being between 3½ and 6 times (depending on the sector) as
productive as the worst plants in most cases.
Figure D1
Distribution of Productivity in UK Manufacturing
90th
Percentile
Frequency
10th
Labour Productivity
Source: Barnes and Haskel (2000b).
Note: Selected ARD data for 1994 to 1997. Distribution is number of enterprises at each
value of average productivity, where the values are rounded to the nearest 100.
Productivity measured as GVA per head. Data have been Winsorised to correct for
outliers.
27
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Table D2 shows data for each year plus 1998 along with the number of
contributors.
Table D2
Comparing percentile points of the productivity distribution
Year
1994
1995
1996
1997
1998
Number of
contributors
2,875
2,892
2,661
2,631
2,483
10th
Percentile
11.2
10.3
11.5
12.1
11.1
50th
Percentile
24.2
25.1
27.1
28.4
28.2
90th
Percentile
53.6
59.0
61.4
61.8
64.2
10th/90th
0.208
0.175
0.187
0.196
0.173
Source: Business Data Linking Project, ONS.
Notes: Columns 3 to 5 show percentiles of the distribution of GVA per head. Column 6
shows the ratio of the 10th percentile (column 3) divided by the 90th percentile (column 5).
Table D3 reproduces work by Oulton (1997) for the 1980s. His work shows a
remarkably consistent standard deviation, but with some suggestion that the
skew has been rising since the early 1980s.
Table D3
Moments of log labour productivity distribution and output growth:
all manufacturing, 1980 - 1992
Year
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
Number of
contributors
14,714
14,483
14,253
13,893
18,123
13,643
13,056
13,179
13,315
18,754
13,881
13,666
13,222
Standard
Deviation
0.65
0.66
0.67
0.66
0.67
0.65
0.66
0.66
0.67
0.68
0.66
0.66
0.66
Skewness
-0.86
-1.08
-0.10
-0.70
-0.56
-0.66
-0.69
-0.64
-0.89
-0.84
-0.86
-0.84
-0.74
Kurtosis
5.83
8.80
6.27
6.25
4.22
5.92
6.07
4.76
6.38
5.54
6.41
5.57
4.53
Output Growth
% p.a.
-9.00
-6.36
-0.13
2.04
3.72
2.76
1.29
4.57
6.80
4.39
-0.20
-5.55
-0.64
Source: Oulton (1997).
Notes: Columns 3 to 5 show the moments of the distribution of log GVA per head.
Column 6 shows the output growth rate in manufacturing expressed as percentage
change per annum.
28
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1.10.3 Productivity, entry and exit.
An interesting question is how much productivity is increased by surviving
businesses raising productivity and how much by entry and exit of new firms. To
investigate this requires longitudinal linking of data. The Centre for Economic
Studies of the US Bureau of the Census has been engaged in this work since
1982, but only recently has this been done for the UK. Price deflators are also
required and, if TFP is to be examined, so are estimates of capital stock. Capital
stock estimates introduce a raft of conceptual and measurement issues. Harris
and Drinkwater (2000) discuss some ideas of capital stock generation at the local
unit level and TFP. Other researchers generate capital stock series at the
establishment level, such as Disney, Haskel and Heden (2000) and Griffith
(1999a).
The BDL Project has recently provided data for the OECD Firm Level Study, part
of an OECD project looking at the sources of productivity growth in member
states. Using the ARD data for the UK, and working with participants in other
member states with access to equivalent data, Barnes, Haskel and Maliranta
(2001) (BHM) produced labour productivity decompositions for eight OECD
countries.9 Table D4 shows some of the results from the project. Column 1
shows the 5-year productivity growth rate for each country. Columns 2 to 5 give
the share of this growth rate due to each of the components of the
decomposition. The within effect is existing firms getting more productive,
between is exiting firms gaining market share, cross is a co-variance term and
net entry is the effect of entry and exit (column 6 minus column 7).
Table D4
Labour productivity decompositions, Manufacturing, 1987-1992
1
Labour Prod
Country
Growth (5 yr)
Finland
23.7
Netherlands 11.8
Portugal
29.5
UK
9.0
USA
8.1
2
3
4
Within
61
90
82
64
85
Between Cross
22
-1
19
-25
-4
-16
10
-8
31
-44
5
Net
Entry
18
17
37
34
28
6
7
Entry
2
41
7
21
-16
Exit
-15
24
-30
-13
-44
Source: Barnes, Haskel and Maliranta (2001).
Note: The productivity growth rate quoted is over a five-year period. Method used is FHK
(see Foster, Haltiwanger and Krizan, 1998). Manufacturing is defined as the 2-digit ISIC
sectors 15 to 37. All values in columns 2 through 7 are per cent of total change in column
2. Net entry = entry – exit: figures may not add up due to rounding.
BHM note there is interesting variation in the composition of the net entry term
with the USA having a negative entry term but a particularly high (in absolute
value) negative exit term. The entry and exit terms are correlated with measures
of (lack of) product market regulation. This issue is explored further in their
paper. The results of this project will also be included in the OECD’s industry
9 See the appendix for more technical details of the decompositions.
29
ANNUAL RESPONDENTS DATABASE: USER GUIDE
level STAN database to allow their use by other researchers. This type of
decomposition analysis has also been undertaken in greater depth for the UK
alone in Disney, Haskel and Heden (2000).
1.10.4 Linking the ARD with different surveys.
As samples for all ONS business surveys are drawn from the IDBR, this presents
the opportunity for linking together the data from different ONS surveys using the
unique enterprise business register numbers. An example of this is that the ONS
R&D survey has been linked with ARD data. Recent work with the DTI has also
demonstrated that it is possible in some cases to achieve useful linkages
between surveys even where the units of measurement of businesses are not so
perfectly matched. There is thus considerable scope for extending such crosssurvey data linkages. Surveys that are being investigated for linking include R&D,
innovations, and producer price data. Hildreth and Pudney (1998, 1999) carried
out work to link the new earnings survey establishment level data to the ARD
establishment data with some success.
It is also possible to link R&D data and the ARD. To illustrate the sort of simple
calculations that can be done using this linked data, the 1994-97 data were split
into 37 industries and median GVA per head calculated for each industry. R&D
per head was then calculated for businesses with above and below median
productivity in each industry. Table 5 shows that in 95% of the industries, the
above median productivity group of businesses also had above median R&D per
head.
Table D5
Links between productivity and R&D
Industries with:
Above median productivity
and R&D
Above median productivity
and below media R&D
R&D per
head
35
(95%)
2
(5%)
Source: Business Data Linking Project, ONS.
Notes: Linked ARD and R&D Surveys, 1994-1997. Results for business in 37
manufacturing 3-digit SIC industries.
1.10.5 Other work using the ARD
Other work undertaken using ARD data has looked at a wide variety of issues.
Harris (2000) looks at profitability, efficiency and market share over the period
1978 to 1995, looking at a cross-section of six industries. Haskel and Heden
(1999) look at the effect of computers on the demand for skilled labour using
establishment level data from the ARD in combination with industry data. Barnes
and Haskel (2000c) look at job flows in UK manufacturing and in particular the
role of small firms in the job creation and destruction process. Barnes and Haskel
(2001) give an overview of productivity, competition and downsizing. Productivity
growth and downsizing are also analysed in Barnes and Haskel (2000a) and
Haskel (2000).
30
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Finally, the inclusion of post-codes in the ARD also allows spatial analysis. For
example, cross-sectional analysis to study the detailed location patterns of
manufacturing industries is currently being conducted by Duranton and Overman
(2001). They develop distance-based measures of localisation and apply them to
the ARD data set for 1997. This approach allows an assessment of the statistical
significance of departures from randomness.
31
ANNUAL RESPONDENTS DATABASE: USER GUIDE
2 Technical information
2.1
Cleaning the data
Issues specific to certain years, see also notes below on longitudinal linking and
overcoming the IDBR introduction.
•
In the universe files in 1980, 1981, and 1982 the returned employment
(ret_emp) and register employment (sel_emp) have been swapped around.
•
In some years reporting units that report on a number of local units are
wrongly interchanged with one of their local units
Table 4 below summarises the initial data cleaning process and the extent of
gaps in the data.
Table 4
Initial data cleaning process: reporting units
Year
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
Reporting units
Yearly cross sections
Panel
Total in
Selected Total after selected
Units
raw data in raw
first
after first
created to fill
data
cleaning
cleaning
gaps in time
109,182
15,108
109,165
15,100
807
108,608
14,922
108,608
14,922
1,218
102,676
14,620
102,676
14,620
206
104,276
14,204
103,840
14,079
1,114
139,275
18,546
138,806
18,424
663
144,333
13,986
143,990
13,875
3,586
147,466
13,417
147,118
13,328
4,304
147,465
13,514
146,570
13,250
4,986
149,271
13,664
148,462
13,393
3,281
153,608
19,196
152,455
18,821
5,413
145,184
14,242
144,218
13,921
4,519
139,065
14,143
138,161
13,879
14,164
141,394
13,661
140,700
13,434
54
149,388
13,579
149,388
13,579
159,438
13,116
144,726
11,414
7,082
172,212
12,682
172,212
12,682
4,160
166,623
13,220
166,623
13,220
1,248
171,640
12,771
171,640
12,771
569
171,271
13,264
171,271
13,264
909
172,323
13,365
172,184
13,365
177,008
12,864
176,892
12,864
Source: Data cleaning by Ralf Martin
Note: Manufacturing sector only
To link over the introduction of the IDBR between 1993 and 1994, we match at
the local unit level over time, using postcode and sector to match if necessary. In
addition to using the existing ONS lookup table. We then link the matched local
33
ANNUAL RESPONDENTS DATABASE: USER GUIDE
unit references to the reporting unit references in 1997 and track this back. This
creates new local and reporting unit keys that are necessary to avoid duplicates
being created by the matching process. A long panel can then be created and
gaps in the data filled in as necessary.
From BDL notes database
•
In 1995 the gross output variable does not contain any non zero entries.
Solved the problem by calculating gross output as total turnover + stocks
eop-stocks bop+work of capital nature i.e. q156+q177-q176+q252.
•
Surveys such as the Annual Inquiry into foreign direct investment (AFDI)
often have survey specific reporting units typically having 60 as the first 2
digits. To match them to ABI we used an idbr lookup to get their egrp_ref and
then match on enterprise group level. This is problematic in earlier because
the idbr lookup is a current snapshot.
•
Andrew Walton has this way round: The 60 refs typically refer to an
enterprise group already. The egrp_ref can be derived by chopping of the 60.
I.e. a reporting unit with the number 60896165388 reports about enterprise
group number 896165388
•
Enterprise group reference numbers changed from 1997 to 1998. This
reflects a move from an earlier register to the D&B one. The enclosed
document contains a lookup table
•
There has been a change in the region classification after 1996. Here is the
relevant lookup table
The regions have changed over the years and from 1996 to 1998 we moved from
Standard Statistical Regions (SSR) to Government Office Regions (GOR). The
table below shows what variables were included in the ARD files:
Year
Standard Statistical
'Old' Government
Regions (SSR)
Office Regions
Possible Values 1 - 11 Possible Values 1-13
1996 Code = REGION
Code = GOREG
1997 Not used
Code = GOREG
1998 Not used
Not used
1999 Not used
Not used
Appendix 3.10 shows, for each option, the possible
text description of the area.
Current Government Office
Regions (GOR)
Possible Values AA - ZZ
Not used
Code = GOCODE
Code = REGION
Code = REGION
values with more detail eg
•
After 1992 all variable which report values are denominated in 1000s
•
Info about the 1999 data load:
1. The question values used are WQ values i.e. the values include 'expanded
values'.
Expansion occurs where a question is asked to a limited number of contributors
but is then created for all other contributors, e.g. on the 'short forms' only Total
Turnover (WQ399) is asked, on the 'long forms' a breakdown of Total Turnover is
34
ANNUAL RESPONDENTS DATABASE: USER GUIDE
asked (WQ301, 308, 309, 310, 311). Expanded values will be created for all
contributors who received a short form, i.e. values will be created for WQ 301,
308, 309, 310, 311.
(Within an employment/ industry strata combination (cell) the ratio of the
components to the total, for the returned long forms, is applied to the total
variable for the short forms in that cell. This is subject to there being a minimum
number of long forms in the cell.)
2. Capital Expenditure breakdown and derived values.
ABI does not request a detailed breakdown of Capital Expenditure. Values for
WQ 510 -WQ533 are derivations. These are calculated using totals from ABI
and additional data from the Quarterly Capital Expenditure Inquiry. Estimated
data is also included for units Not Yet in Production (NYIP), again this data is
taken from the Quarterly Capital Expenditure Inquiry.
The exception to this is the Property Sector where a breakdown is requested
(WQ 680 - 699).
3. For Production and Construction Sectors only, a breakdown of Stocks is
given. These are derivations and are calculated using totals from ABI and
additional data from the Quarterly Stocks Inquiry.
4. The SNL files (local unit information for all reporting units in the universe) are
created for each sector for ABI 2 and for ABI 1 as a whole. Any unit selected
for ABI 2 must also be selected for ABI 1, because of this there will be
duplication between the ABI 1 and ABI 2 sector SNL files.
5. Question 9 (Research and Development - Yes/No question) has been
included where possible. Due to processing problems this question may be
missing for some contributors in Production, Construction, Services and
Wholesale sectors.
6. Question 9 (R&D) will not be included in 2000 data as it is only asked every
other year. It will return in 2001.
7. Question 143 (Trade in Overseas Services). Due to processing problems
this question is missing for the Wholesale sector.
8. Question 143 (Trade in Overseas Services) will not be included in the 2000
data. From 2000 onwards four questions relating to trade in services with
other countries have been included in ABI.
•
Info on 2000 data load
1. The question values used are WQ values i.e. the values include 'expanded
values'.
Expansion occurs where a question is asked to a limited number of contributors
but is then created for all other contributors, e.g. on the 'short forms' only Total
35
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Turnover (WQ399) is asked, on the 'long forms' a breakdown of Total Turnover is
asked (WQ301, 308, 309, 310, 311). Expanded values will be created for all
contributors who received a short form, i.e. values will be created for WQ 301,
308, 309, 310, 311.
(Within an employment/ industry strata combination (cell) the ratio of the
components to the total, for the returned long forms, is applied to the total
variable for the short forms in that cell. This is subject to there being a minimum
number of long forms in the cell.)
2. Capital Expenditure breakdown and derived values.
ABI does not request a detailed breakdown of Capital Expenditure. Values for
WQ 510 -WQ533 are derivations. These are calculated using totals from ABI
and additional data from the Quarterly Capital Expenditure Inquiry. Estimated
data is also included for units Not Yet in Production (NYIP), again this data is
taken from the Quarterly Capital Expenditure Inquiry.
The exception to this is the Property Sector where a breakdown is requested
(WQ 680 - 699).
3. For Production and Construction Sectors only, a breakdown of Stocks is
given. These are derivations and are calculated using totals from ABI and
additional data from the Quarterly Stocks Inquiry.
4. The SNL files (local unit information for all reporting units in the universe) are
created for each sector for ABI 2 and for ABI 1 as a whole. Any unit selected
for ABI 2 must also be selected for ABI 1, because of this there will be
duplication between the ABI 1 and ABI 2 sector SNL files.
5. For 2000 two new divisions have been included, these are Forestry - SIC's
02010, 02020, and Fishing - SIC's 05010, 05020. These will be published in
the Final 2000 results in June 2002 for the first time.
6. Also for 2000, SIC 51475 - Wholesale of musical instruments - has had
contributors selected this year.
7. Questions 163 and 164 are new for 2000 - these relate to amounts received
for imports and exports.
8. Question 610 (Investment in purchased computer software included in 600) is
new for 2000. However 610 is historically a derivation code used within the
ABI SAS system. Therefore code 610 on the form is loaded for ARD
purposes as 788.
9. Question 250 (Other non retail turnover) is new for 2000. However 250 is
historically a derivation code used within the ABI SAS system. Therefore
code 250 on the form is loaded for ARD purposes as 787.
10. On the retail sector from 2000, the commodity breakdowns now conform to
COICOP (Classification of Individual Consumption by Purpose standards).
The breakdowns are different but the overall totals are comparable.
36
ANNUAL RESPONDENTS DATABASE: USER GUIDE
11. Question 143 (Trade in Overseas Services) has been expanded to collect
actual data (Questions 1033, 1034, 1035, and 1036). These four questions
relate to trade in services with other countries.
12. Question 9 (R&D) is not included in 2000 data as it is only asked every other
year. It will return in 2001.
•
Foreign ownership marker
Before 1994 UK RUs are marked in the data as 801. From 1998 onwards they
are marked as 783. Between 1994 and 1997 inclusive they are marked as both.
•
1970-1972 contains respondent data only. 1970 data must be treated with
caution. Most businesses are duplicated on the file. A few are triplicated.
Only some occur once.
•
Constructed and Local Unit data has been added to the respondent data for
1984-1992. Constructed data after 1992 was always present.
•
In 1993, the construction sector was included in the ARD. However, in this
year, many businesses were amalgamated in order for grossing to be
performed. All businesses in this sector were singles.
2.2
Problems using the data
Some researchers are concerned that changes in the composition the Enterprise
through Local Units closing or changing ownership may distort the data from
Enterprises. Hence they apportion the Enterprise data across the Local Units
recorded on the business register for that Enterprise. This is done using
employment recorded for that unit on the IDBR as provided by VAT or PAYE
returns or from other ONS surveys10. These IDBR size data are not considered
to be as reliable as directly returned data partly because there can be a delay in
its updating.11
Obviously directly returned data are only available for contributors that are
selected in the ACOP/ACOC/ABI surveys and that returned a form. This
returned data held in the ARD is sometimes called the raw data12, but it is more
accurate to call it ‘contributor’ data as the returned data may have been adjusted
or used to impute some variables where contributors have not been asked for a
10
The employment numbers on the register for local units are likely to be imputed. From
1994 the IDBR also contains turnover data.
11
In addition it is likely that recent changes in the composition of an enterprises local
units which may be included in the returned survey data may be absent from the local
units on the register. This is almost certain when looking at the snapshot of the register
from which the sample was drawn (known as the ‘non-selected’ file).
12
These data have also been termed as the ‘selected’ data.
37
ANNUAL RESPONDENTS DATABASE: USER GUIDE
full range of data. For example, when businesses fail to return the questionnaires
the contributor data for that unit may be constructed if ONS has sufficient
information from other sources. Thus constructed data may use returned data
such as turnover, stocks and capex from other inquiry sources and will generally
be based on what ONS already knows about that particular company and its
history (e.g. from follow-up or other current and past surveys). Also constructed
data are usually used only for the 1:1 cells i.e. where a census has taken place.
This is therefore more accurate than straight pro-rata allocations based on
measures of size from the register. Therefore for many purposes the ‘processed’
data can be treated as real returned data even when constructed or imputed, but
researchers should be aware of these constructed, imputed or outlier adjusted
contributor data where this could affect results.
The main need for imputation arises from the use of 'short-forms'. These are
used to reduce business compliance costs. A percentage of firms, other than in
100% sampled cells13, are sent shorter forms than the full survey questionnaire.
These forms ask for the full range of data, including sectoral specific questions.
However they generally ask for totals only for a number of sections where the
long form asks for a breakdown, e.g. breakdown of intermediate consumption.
These short form values are expanded to give the full breakdowns on a cell by
cell basis using ratios derived from long form returns in the cell.
ONS regards a response as an outlier if it is outside certain limits of what to
expect for that Enterprise i.e. when compared with previous surveys or
administrative data.
Where this cannot be reconciled through follow-up
enquiries, smoothed outliers are added to the original and constructed
respondent data to produce a set of variable sets prefaced by ‘WQ’. The ‘W’
refers to the method of smoothing outliers known as ‘Winsorisation’. This was
the method of dealing with outliers used between 1993 and 1997 on the ‘new’
ACOP Results system. In the ABI Results system alternative outlier correction
methods are used. The view of ONS is that the processed data are for most
purposes a reliable measure of the individual business variables. In the ARD
data files, winsorised data are not supplied but imputations have been done.
As all the contributors in the ARD are identified by their unique business register
reference number, it might seem straightforward to link the data for individual
businesses through time. In practice there are many problems that arise. For
example, apart from the size-bands where there is a yearly census, the firms
selected for surveys change year to year. Thus longitudinal linking may return
missing values for the smaller businesses across the years where they were not
included in the ACOP/ABI surveys. Changes in SIC structures e.g. from SIC 80
to SIC 92, also complicate longitudinal linking of industries. In addition changes
in the register due to both gradual improvements and complete revisions (e.g. the
move to the IDBR) present extra problems. When this happened it can be
unclear where previous contributors ended up on the new register. The
problems of allowing for changes in local units associated with an enterprise and
changes in ownership have already been mentioned.
Particular problems with register changes occur in 1983 to 1984 when the VAT
register was added to the CSO Business Register. This resulted in large amounts
13
In 2001 and beyond some completely enumerated cells do contain short forms for
compliance control reasons.
38
ANNUAL RESPONDENTS DATABASE: USER GUIDE
of entry and exit including large numbers of observations only present for one
year. Mostly this occurred in the universe files rather than the selected sample.
A further potential problem was the change to ‘company’ reporting in 1987. In
practice this doesn’t seem to be a big problem. Perhaps the greatest challenge
of all was posed by the introduction of the IDBR in 1994. Unfortunately a
comprehensive look-up table between the old and new registers was not kept14.
To link the data longitudinally more sophisticated techniques were required. The
process of cleaning and linking the reporting unit level data is described in more
detail below.
2.3
Grossing the data
Because we have wage, sales output etc data on the survey files only it is
necessary to weight the data so that summary statistics represent the “universe”.
We can population weight the data either according to the sampling rules used
by the ONS reported in table 1, or using the sampling probabilities given by the
actual data.
•
Using ONS’s sampling probabilities. In this case weights would reflect the
probabilities given in table 1. The weight is 1 where a census is performed
(for the very largest reporting units) and something higher for the smaller
units, e.g. 2 if the probability is 0.5 (the inverse of the probability of a
business of a certain size appearing in the survey). To weight employment
we multiply actual employment by relevant weight. However this method
does not allow for the fact that some selected large businesses may not
respond or may have a reporting unit change that means they are no longer
selected15. The second method given below is therefore preferred.
•
Using sampling probabilities given by the survey and universe files. If
we have both the survey and universe files we can count how many reporting
units of a certain size/industry/region appear in the survey files, and how
many of that size/industry/region appear in both the survey and the universe
together. Suppose there are 700 reporting units with 30 employees and 100
of them are in the survey file, the sampling probability is 1/7. We can then
multiply the employment in firms with 30 employees in the survey files by 7
so that these 100 firms will represent the true 700 firms. For the survey files
we have “returned” actual employment but for the universe files we have to
use the indicative employment that is on the register. Comparing total
employment rather than the number of observations and using the same
process as above is an alternative method.
14
A lookup table only exists for the selected sample. In order to link observations that
enter and exit due to sampling (i.e. smaller units) a complete mapping is needed as there
may be a number of years between observations.
15
It is possible for a selected reporting unit to change the way it reports after it receives
the form, or in the case of a merge or acquisition where the reporting unit has changed.
In such cases contributors can be asked to return the form in an appropriate manner but
response cannot be enforced as the new reporting units were not selected for the survey.
39
ANNUAL RESPONDENTS DATABASE: USER GUIDE
2.4
Register continuity
1994 was the first year of using the new business register - the IDBR.
However, the inquiry processing system was not replaced and the IDBR
referencing system was made to fit. (This continued until ARD98.)
On the ARD 'S' for Site datasets, the field CSO_REF will hold the IDBR LUREF
correctly for the individual sites of multi-site businesses.
However, elsewhere and on the other ARD datasets, the field is a meaningless
duplicate of CSO_REF2 which now holds the IDBR RUREF. This is because with
the old register system CSO_REF was always the same as CSO_REF2 except
at the individual site level. A single establishment did not have a distinct separate
site; with multi-site businesses, one site was nominated as the parent local unit
and this was given the establishment reference. In both cases, CSO_REF ended
up holding the same CSO_REF2.
This means that the new LUREF has been lost for singles.
To overcome this and to ensure some register continuity, a special exercise was
mounted for 1993/94 datasets only.
On ARD93 datasets, the field NEXT_REF holds either a RUREF or a LUREF for
matching against 1994. RUREFs have more digits than LUREFs being always 11
digits and start with a ‘4’ or ‘5’.
On ARD94 datasets, the field PREV_REF holds an old register reference for
matching against 1993. Establishment or Site references are indistinguishable
since they share they same referencing system.
It was intended to find the LUREF for single site businesses. This should have
been stored in the field CSO_REF. By mistake, the exercise placed it in
CSO_REF2. Since this field was always defined as numeric, those minority
LUREFs which are prefixed by the letter 'X' would not fit. A field CSO_REF2C
was added to hold these.
40
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3 Detailed Appendices
41
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.1
Variables and derived variables
This appendix describes common variables that are used and lists changes to derived
variable definitions over time.
43
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.2
Coverage for ABI
In this appendix we list the sectors covered by the ABI (and hence the ARD) from 1998
onwards:
Section Division Description
A
1
Agriculture, Hunting And Related Service Activities
2
Forestry, Logging And Related Service Activities
B
5
Fishing, Operation Of Fish Hatcheries And Fish Farms;
Service Activities Incidental To Fishing
C
D
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Mining Of Coal And Lignite; Extraction Of Peat
Extraction Of Crude Petroleum And Natural Gas; Service
Activities Incidental To Oil And Gas Extraction Excluding
Surveying
Mining Of Uranium And Thorium Ores
Mining Of Metal Ores
Other Mining And Quarrying
Manufacture Of Food Products And Beverages
Exclusions
ALL
1998, 1999
1998, 1999
10.2
ALL
13.2 in 2000
14.13 in 1998
Some of 15.9 in
all years
Manufacture Of Tobacco Products
Manufacture Of Textiles
Manufacture Of Wearing Apparel; Dressing And Dyeing Of
Fur
Tanning And Dressing Of Leather; Manufacture Of
Manufacture Of Wood And Products Of Wood And Cork,
Except Furniture; Manufacture Of Articles Of Straw And
Plaiting Materials
Manufacture Of Pulp, Paper And Paper Products
Publishing, Printing And Reproduction Of Recorded Media
Manufacture Of Coke, Refined Petroleum Products And
Nuclear Fuel
Manufacture Of Chemicals And Chemical Products
Manufacture Of Rubber And Plastic Products
Manufacture Of Other Non-Metallic Mineral Products
Manufacture Of Basic Metals
Manufacture Of Fabricated Metal Products, Except
Machinery And Equipment
29
30
31
Manufacture Of Machinery And Equipment n.e.c.
Manufacture Of Office Machinery And Computers
Manufacture Of Electrical Machinery And Apparatus n.e.c.
32
Manufacture Of Radio, Television And Communication
Equipment And Apparatus
33
Manufacture Of Medical, Precision And Optical Instruments,
Watches And Clocks
34
Manufacture Of Motor Vehicles, Trailers And Semi45
ANNUAL RESPONDENTS DATABASE: USER GUIDE
E
F
G
H
I
J
35
36
37
40
41
45
50
Manufacture Of Other Transport Equipment
Manufacture Of Furniture; Manufacturing n.e.c.
Recycling
Electricity, Gas, Steam And Hot Water Supply
Collection, Purification And Distribution Of Water
Construction
Sale, Maintenance And Repair Of Motor Vehicles And
Motorcycles; Retail Sale Of Automotive Fuel
51
Wholesale Trade And Commission Trade, Except Of Motor
Vehicles And Motorcycles
52
Retail Trade, Except Of Motor Vehicles And Motorcycles;
Repair Of Personal And Household Goods
55
60
61
62
Hotels And Restaurants
Land Transport; Transport Via Pipelines
Water Transport
Air Transport
63
Supporting And Auxiliary Transport Activities; Activities Of
Travel Agencies
64
65
Post And Telecommunications
Financial Intermediation, Except Insurance And Pension
ALL until 2002
Funding
Insurance And Pension Funding, Except Compulsory Social ALL until 2001
Security
Activities Auxiliary To Financial Intermediation
ALL until 2001
Real Estate Activities
Renting Of Machinery And Equipment Without Operator And
Of Personal And Household Goods
66
K
67
70
71
72
73
74
Computer And Related Activities
Research And Development
Other Business Activities
L
75
M
80
Public Administration And Defence; Compulsory Social
Security
Education
N
85
Health And Social Work
O
90
Sewage And Refuse Disposal, Sanitation And Similar
Activities
Activities Of Membership Organisations n.e.c.
Recreational, Cultural And Sporting Activities
91
92
46
40.3 in 1998
51.24/1 in 1998
(Fur), 51.47/5 in
1999 (Musical
instruments)
62.3 (Space
Transport)
74.15/2, 74.15/4
and 74.81/1 in
1999
ALL
Private/Commerci
al only
Private/Commerci
al only (excludes
85.11/1, 85.12,
85.13, 85.31/1
and 85.32/1)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
P
Q
93
95
99
Other Service Activities
Private Households With Employed Persons
Extra-Territorial Organisations And Bodies
ALL
ALL
47
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.3
Disclosure rules
ONS BUSINESS DISCLOSURE RULES
The current rule (post 1995) is as follows.
Current rule (post 1995)
1.
There must be at least three enterprise groups in a cell (threshold rule).
2.
If so, the sum of the returns of all reporting units excluding the largest two must
be at least 10 per cent of the value of the largest (dominance rule).
A cell meeting both these criteria is non-disclosive.
Note. Modification in those cases where sampling fraction in a cell is less than
100%. If there are exactly two distinct enterprise groups, the threshold rule is relaxed by
treating any non-sampled units as if they were a third enterprise group.
Therefore, if there are two sampled enterprise groups and at least one non-sampled unit,
the threshold rule is passed.
(If on the other hand, the cell is not sampled and there are only two enterprise groups in
the cell, then the cell fails the threshold rule.)
History underlying rule changes in 1995
Introduction
SMQ division put a proposal for a change to the CSO rules for disclosure testing to
CSOEX in June 1995. The main purpose was to reduce the costs and complexity of
disclosure testing. The proposal was to apply the rules at enterprise rather than at
enterprise group level. CSOEX considered the changes, but was uneasy about dropping
the enterprise group criterion; this mirrors an unease among several inquiry statisticians.
Dislcosure Rule in 1995
The disclosure rule operating up to 1995 had two parts:
1
there must be at least three enterprise groups in a cell (threshold rule);
2
if so, the sum of the returns of all enterprise groups excluding the largest
two must be at least 10 per cent of the value of the largest (dominance
rule).
A cell meeting both these criteria is non-disclosive.
In order to determine the outcome in each case, it is necessary to combine reporting units
into groups, determine which are the two largest values and then do the calculation. This
process is complex to program, requires substantial additional file-store and takes a lot of
processing time.
New proposal
49
ANNUAL RESPONDENTS DATABASE: USER GUIDE
The new proposal seeks to retain as much of the protection of the current rules as possible
while greatly reducing processing costs. There are also other reasons relating to releasing
as much information as possible.
In summary, it was proposed in 1995 to retain the threshold rule at enterprise group
level, but to apply the dominance rule at the reporting unit (enterprise) level only.
At the same time, it is possible to improve the efficiency of the processing by introducing
simpler intermediate tests which will decide a majority of cases one way or the other, while
retaining the more complex test for borderline cases.
Effect of the changes
The changes only affect the case where one or two enterprise groups account for more
than 90-95 per cent (but less than 100 per cent) of the total in a cell, but the largest two
constituent reporting units do not.
By ensuring there are more than two enterprise groups in a cell, the new proposal means
that no enterprise group can deduce precisely the figures of any other (except by collusion
which is an unavoidable risk under the current system).
The worst case is where one enterprise group is known to dominate to the extent of 99 per
cent (say) of the particular cell. Its figures would then be known to within one per cent. On
the other hand, it would not be possible to know for certain without certain prior knowledge.
It is argued that this situation is rare and involves a combining of separate data; that it is
within the law; and that, if the enterprise group in question were to complain, a sympathetic
and ad-hoc decision could be taken to ensure this enterprise group was given special
treatment thereafter.
Nonetheless, when the new rule is in operation for the first time, it will be prudent to
examine cells which previously had been considered disclosive (under the old rules) but
were not under the new rules, to see if there was a particular risk.
50
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.4
Example forms
51
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.5
Other linkable surveys
All linking work will be separately documented. When complete, linked data sets will be
made available to other researchers.
a)
Community Innovation Survey / UK Innovation Survey (CIS). Linking under way by
the Centre for Research into Business Activity (CeRiBA).
b)
E-commerce survey. Linking under way by the Centre for Research into Business
Activity (CeRiBA).
c)
Annual Inquiry into Foreign Direct Investment (AFDI). Linking under way by the
Centre for Research into Business Activity (CeRiBA).
d)
Business Enterprise Research and Development (BERD). Linking under way by the
Institute for Fiscal Studies (IFS).
e)
Workplace Employment Relations Survey (WERS). Legal problems presently
preventing linking by the Centre for Research into Business Activity (CeRiBA).
f)
Quarterly Fuels Inquiry. Linking under way by the Centre for Research into
Business Activity (CeRiBA).
g)
New Earnings Survey. Linking under way by the Centre for Research into Business
Activity (CeRiBA).
53
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.6
Deflators
55
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.7
Early notes on the ARD
Edited and included for reference. Contains useful tables of codes. Filenames and
structures have since been changed. See main text for details.
3.7.1
ANNUAL RESPONDENTS DATABASE
3.7.1.i
Introduction
The Annual Respondents Database (ARD) is a timeseries of ACOP data for the
production sector ranging from 1970 to 1997. It is a complete series of returned
data but excludes estimated values, and all local unit values. The database does
however hold register data (Selected Employment, Postcodes etc) for the whole
ACOP universe. This document explains how the data from the ACOP inquiry has
been incorporated into the ARD.
3.7.1.ii
Data sources used to create the ARD
A separate file of disaggregate data. was used for each year of the ACOP inquiry.
The data required for the ARD has been held in 4 different formats since 1970:
1970 - 1979
Data held on the Extended Forms File (EFF).
1980 - 1992
Data held on the Census File (CF)
The above files were COBOL files held on an ICL 3980 mainframe which has now been
decommissioned.
1992 - present
Data held as INGRES tables in a UNIX environment.
Data for 1993 - present is also held in SAS.
3.7.1.iii ARD ASCII Files
The files described above were used to create two ASCII format files for each year
(yy = year):
NONSELyy.txt - contains indicative data only for :all local units
non-selected parents and singles (including NYIP contributors), (for 19701972 see para 3.7.1.vi)
constructed contributor returns (pre-93)
deleted contributors
57
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Selection conditions
1970 - 1972 (EFF-STRUCT<>2)
1973
(EFF-STRUCT<>2 or EST-MARKER-X NOT IN($A# $D#))
1974 - 1979
(EFF-STRUCT<>2 or EST-MARKER-X NOT IN(1 2))
1980 - 1992
(STRUCTURE=9 or SEL_STATUS <> 8)
1993 – 1997
(STRUCT=9 or SEL_STAT not in(3 4 5)
SELyy.txt - contains indicative data and data values for :selected parents and singles (including NYIP contributors),(for 1970-1972
see 3.7.1.vi)
constructed contributor returns (post 1992)
Selection conditions
1970 - 1972 (EFF-STRUCT=2)
1973
(EFF-STRUCT=2 and EST-MARKER-X IN($A# $D#))
1974 - 1979 (EFF-STRUCT=2 and EST-MARKER-X IN(1 2))
1980 - 1992 (STRUCTURE <> 9 and SEL-STATUS = 8)
1993 - 1997 (STRUCT <> 9 and SEL_STAT IN (3 4 5)
3.7.1.iv Creation of SAS datasets
Each ARD ASCII file has been converted into an equivalent SAS dataset:
NONSELyy.sd2 - indicative data only
SELyy.sd2 - indicative data and data values
When a data extract is requested it will normally be provided via SAS.
The SAS datasets and ASCII files are held in compressed format (using WinZip file
compression software) in D:\AMD\ on PC A201 in Room D.230 and on Novell
Server Npt7 on R:\AMD. The copies held on the server are backed up daily with
permanent tape backups taken monthly. The server copies of ARD files should be
regarded as the original versions. Any authorised amendments to ARD SAS
datasets should be copied to R:\AMD. The ASCII files should be regarded as
archive material and never be edited.
3.7.1.v
58
Data Format
The indicative data was stored in a standard format for all years even though some
variables are missing/blank in certain years. See Appendix A for a complete list of
ANNUAL RESPONDENTS DATABASE: USER GUIDE
indicative data items used from 1970 - 1994 and a list of files created for the
timeseries. This involves an extra storage overhead but simplifies conversion and
documentation of the series. The number of changes of questions asked over the
period made this approach impractical for the data values held, so the question
value variables vary year to year. The questions asked in any particular year can
be found by reference to Lotus Spreadsheets QUES7079.wk4 (1970-1979) and
QUES8094.wk4 (1980-1994). Prints of these tables are held at Appendix B.
3.7.1.vi Conversion of 1970-1972 data
Data for 1970 -1972 was derived from files that predate the EFF though the record
layout was the same. In these years EST-MARKER-X field was always = 0 and
could not be used in the selection condition. Therefore non-selected files for this
period contain local units only whilst the selected files contain only parents and
singles. The resulting 1970 file appeared to contain no local unit records. However,
parents and singles with common establishment references exist on the file
suggesting further matching would identify which $singles# were actually local
units. It is thought that this was the last year when a $full# census was undertaken
3.7.1.vii .Conversion of 1993-1997 data
The selection criteria used for inclusion in the ARD $selected# file for 1993
onwards differs in that constructed returns (SEL_STAT in(4 5)) have been included
to be consistent with the contents of CONTRIB1 dataset in the ACOP results
system. For previous years only actual returned data (SEL_STAT=3) is held ( the
logic being that only $real# data is suitable for microdata research).
Some extra variables are available for some contributors from the ACOP SAS
system so these have been added. Most are extra regional variables, the
remainder are population counts and industry/sizeband data. There is also a
PAR_EMP field to indicate parent employment for a local unit.
Deleted contributor records (SEL_STAT in 7 8) are held as in previous years on the
non-selected file.
3.7.1.viii Constructed returns and returned Local Units
Prior to 1993 constructed data for non-returns are not held on the $selected# ARD
file. Apportioned and real (CAPEX only) data for returned local units is also
excluded. Except for the CAPEX data, these values were excluded as they were
not $real# data and thought not to be of interest to microdata research. Since then
it has become clear that this data (especially the CAPEX values) would be useful.
A decision to rerun back years of ACOP data through the current grossing method
provided an opportunity to retrieve data for constructed and local unit contributors
for 1980-1992. The retrieved data is held in the following datasets (1984 and 1989
only at present) :
CSTRyy.sd2
contains data values for constructed contributors
LUSELyy.sd2
59
ANNUAL RESPONDENTS DATABASE: USER GUIDE
contains data values for returned local units
Indicative data is already available for these contributors on NONSELyy.sd2. At a
later stage these contributors may be removed from NONSELyy.sd2 and
CSTRyy.sd2 and LUSELyy.sd2 merged with SELyy.sd2. Until then it is important
when combining these additional datasets with NONSELyy.sd2 to avoid doublecounting these contributors.
From 1993 onwards indicative data and data values for constructed returns are
held in the $selected# file. Selected local units continue to be held as indicative
data only on the $non-selected# file. As in previous years, the data values for
these returned local units will be held on a separate dataset LUSELyy.sd2 until a
decision is taken to add them to SELyy.sd2.
3.7.1.ix
Summary of changes affecting ARD data since 1970
1970
Last “full” census.
1979
Benchmark year.
1980
SIC80 activity classification replaced SIC68. Reclassification on register done by
analysis of returns to quarterly inquiries into manufacturers sales or on a pro-rata
basis in line with the reclassification pattern by industry of establishments for which
actual product sales data was held.
Questions relating to standard cost stocks and road transport costs were dropped.
Sampling increased to :1 -19 exempt
20-49 1 in 4
50-99 1 in 2
100+
1 in 1
18,965 forms mailed.
Publication of BM PA1000 showing provisional results discontinued.
1981
Questions relating to capital expenditure on motor cars dropped.
Sampling as per 1980.
18,670 forms mailed.
60
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1982
Sampling as per 1980.
18,260 forms mailed.
1983
Sampling as per 1980.
17,500 forms mailed.
1984
Register updated making fuller use of HM Customs and Excise VAT records,
increasing by 1.3% overall for manufacturing. Some reclassification of
establishments occurred resulting in some cases to large changes in employment
(eg group 210 +38%, 231 -18%).
Additional questions (for larger establishments only), introduced for road transport
costs, postal and telecommunications costs.
Coverage extended to include activity headings 4121 (slaughterhouses), 4126
(animal by-products processing), and 4930 (photographic and cinematographic
processing laboratories)
Benchmark year.
Sampling increased to:1-19
exempt
20+
(in England)
1 in 2
20+
(excluding England) 1 in 1
24,200 forms mailed.
1985
questions for larger establishments on road transport costs, postal and
telecommunications costs dropped.
Sampling decreased to :1 -19 exempt
20-49 1 in 4
50-99 1 in 2
100+
1 in 1
16,800 forms mailed.
61
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1986
Additional questions asked for number of computer employees, costs of computer
equipment purchased and (for larger establishments only), costs of hiring, leasing
or renting computer equipment.
Sampling as per 1985.
16,200 forms mailed.
1987
Questions introduced in 1986 on computer expenditure (see above) dropped.
Company reporting replaced establishment reporting, with some exceptions (eg
large mixed activity companies which were asked to make separate returns to the
census for each of their production activities on an establishment basis). Little
effect on the main economic series since in practice most businesses both before
and after the change reported for the company as a whole.
Sampling as per 1985.
16,200 forms mailed.
1988
Additional questions asked for number of computer employees, costs of computer
equipment purchased and (for larger establishments only), costs of hiring, leasing
or renting computer equipment. Additional questions asked for cost of assets
leased under finance leasing arrangements.
From 1988 contributors were asked to include the value of assets acquired as
lessees under finance leasing arrangements.
Sampling as per 1985.
16,050 forms mailed.
1989
Additional questions (for larger establishments only), introduced for road transport
costs, postal and telecommunications costs. Questions introduced in 1988 on
computer expenditure (see above) dropped.
Benchmark year.
Sampling increased to:1-19
exempt
20+
(England only) 1 in 2
20+
(excluding England) 1 in 1
23,300 forms mailed.
62
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1990
questions for larger establishments on road transport costs, postal and
telecommunications costs dropped. Additional questions asked on capital and
current costs associated with pollution abatement.
Sampling decreased to :1 -19 exempt
20-49 1 in 4
50-99 1 in 2
100+
1 in 1
16,800 forms mailed.
1991
Additional questions on breakdown of capital and current costs associated with
pollution prevention and solid waste management.
Sampling as per 1990.
16,600 forms mailed.
1992
Additional question identifying those businesses with employees engaged in
research and development work included. From 1992 The breakdown of questions
on capital expenditure and stocks was excluded. Data for these variables for 1992
results was estimated for using information gathered in the CSO Quarterly Capex
Expenditure and Stocks inquiries.
Sampling as per 1990.
15,700 forms mailed.
1993
SIC92 activity classification replaced SIC80 based on NACE Rev1 to guarantee
comparability between national and EC classifications and therefore between
national and Community statistics.
Data for stocks breakdown variables (excluded from 1992) estimated from
information collected in the CSO Quarterly Stocks Inquiries. No individual asset
types published for capital expenditure variables.
Sampling:extended to the 10-19 sizeband, and the 1-9 sizeband in some industries
20-49 1 in 5
63
ANNUAL RESPONDENTS DATABASE: USER GUIDE
50-99 1 in 2
100+
1 in 1
15,700 forms mailed.
Construction included in ARD from 1992.
1994
Transition from CSO Register to IDBR (Inter Departmental Business Register). As
well as potentially increasing the overall number of contributors in the ACOP
universe, this has resulted in a change of contributor references for the first time
since at least 1970 .
1995
Sampling continued in the 1-9 sizeband and will continue for the foreseeable future.
From 1995 1 in 1 sampling for a fixed size band ceased and varied from industry to
industry.
20,790 forms mailed.
1996
[From 1996 the Annual Census of production became known as the Annual
Business Inquiry.]
The layout and content of forms changed resulting in a complete change of variable
code numbers.
20,759 forms mailed.
1997
Introduction of point in time employment to replace average employment.
To maintain consistency, average employment was estimated using STES (Short
Term Employment Survey) data.
20,760 forms mailed
3.7.1.x
Documentation
See Appendix I for references used for this guide and a list of further
documentation available.
See Appendix J for explanations to 20 questions/anomalies highlighted by visiting
academics.
64
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.7.1.xi Description of Data Items held
The remainder of this document describes each data item in detail, and explains
how it has been incorporated into (or excluded from) the ARD. For convenience the
items have been grouped into the sections shown overleaf and when viewed in
WordPerfect may be selected as hypertext.
3.7.1.xii List of Data Items by Category (click the topic you want to view)
xii.1. Contributor References
xii.2
Industry Classifications
xii.3.
Dates
xii.4.
Employment
xii.5.
Structure Markers
xii.6.
Deletion Markers
xii.7.
Selection Markers
xii.8.
Organisation Markers
xii.9.
Activity Markers
xii.10. Record Status Markers
xii.11. Establishment and Unit Reference Check Digit Markers
xii.12 Directory of Manufacturing Business Markers
xii.13. Industry Team Markers
xii.14. Response Chasing Markers
xii.15. Late Return Markers
xii.16. Regional markers
xii.17 Foreign Ownership Markers
xii.18. Financial Markers
xii.19. Reporting Unit Markers
xii.20. Misc
xii.21. Question Values and Vet/Estimation Status Indicators
xii.1.
CONTRIBUTOR REFERENCES
Purpose - uniquely identify individual contributors and provide information on
establishment structure, company and enterprise group ownership.
65
ANNUAL RESPONDENTS DATABASE: USER GUIDE
ESTAB-REF8 1970 - 1979
ESTAB-REF 1980 - 1992
CSO_REF2
1993 - present
UNIT-REF8
1970 - 1979
UNIT-REF
1980 - 1992
CSO_REF
1993 - present
ENT-REF5
1970 - 1979
ENT-REF
1980 - 1992
ENT_GROUP_REF 1993 - present
LU-COUNT
1980 - present
VATKEY
1986 - 1992
COMPANY_REF
1993 - present
Notes.
See also POST-CODE described in REGIONAL DATA MARKERS which also
provides a unique reference to a contributor.
ESTAB-REF8, ESTAB-REF and CSO_REF2 all refer to the standard CSO
reference number, a 7 digit number distinguishing an establishment from all
others. It is usually used in conjunction with a check digit (see ESTAB-CD
described in ESTABLISHMENT AND UNIT REFERENCE CHECK DIGIT
MARKERS). Local units within an establishment will all have the same reference
number. For singles the reference will be unique. For parents and singles, the
estab-ref will be the same as the unit-ref (see below).
UNIT-REF8, UNIT-REF and CSO_REF are all used to hold the local unit level
reference. For a parent or single establishment, the local unit reference will be the
same as the estab-ref. For a local unit, the unit reference will be different and
unique to allow each unit to be separately identified. It is usually used in
conjunction with a check digit (see UNIT-CD described in ESTABLISHMENT AND
UNIT REFERENCE CHECK DIGIT MARKERS). Local units should always have
the same VAT Trade Classification, foreign ownership code, status code and
enterprise number as their parents. Selected and returned employment of LUs
must agree with selected and returned employment of parent. Also totals for codes
201,202,205,206,835,836, and 845 must tally for parent and local units.
ENT-REF5, ENT-REF and ENT_GROUP_REF are all used to hold a 9 digit
reference used to identify enterprise group to which an establishment belongs. It is
this level at which disclosure is tested for and at which many of the results are
calculated for publication.
66
ANNUAL RESPONDENTS DATABASE: USER GUIDE
LU-COUNT holds the number of local units for the parent local unit (including
itself). The value for singles and locals will always be 0 and for parents will always
be >= 2 (there must be at least 2 local units in a multi-establishment group.
VATKEY and COMPANY-REF are used to hold VAT registration information on the
contributor. The first 7 digits hold the 7 digit VAT ref allocated by Customs & Excise
when a business registers for VAT. The remaining 5 digits hold the VATSUBCODE
(3 digits) and VATACTIVCODE (2 digits).
In 1994 the ACOP universe was obtained for the first time from the IDBR (Inter
Departmental Business Register). As well as potentially increasing the overall
number of contributors in the ACOP universe, this has resulted in a change of
contributor references for the first time since at least 1970 . It remains to be seen
how easy it will be to match IDBR references to previous years. There is a lookup
table available for 1993 but this only includes selected contributors. It is likely that
the POS_CODE field may well be the best way of matching 1993/1994
contributors.
Action taken.
Date retained in existing format.
xii.2.
INDUSTRY CLASSIFICATION
Purpose - classify contributors by economic activity.
SIC-68
1970 - 1979
SIC-80
? - 1994
SIC-68ORDER
1970 - 1979
ORIGINAL-SIC-80
1970 - 1979
NACE-CODE
1970 -1979
SIC-80-SOURCE-CODE
1970 -1979
OTHER-CLASS
1980 - 1983
VTC
1986 - present
COMPANY-INDUSTRY
1987 - 1990
ACOP-FORM-TYPE 1970 -1990
FORM-TYPE
1970 - present
Notes.
The exact purpose of ORIGINAL-SIC80, ACOP-FORM-TYPE, NACE-CODE and
SIC-80-SOURCE-CODE are currently unknown.
The EFF years (1970 - 1979) were classified to SIC-68, which had 27 orders, each
divided into a number of 'Minimum List Headings' denoted by 3 digit numbers.
67
ANNUAL RESPONDENTS DATABASE: USER GUIDE
SIC-80 adopted a decimal structure of 10 divisions (1 digit), subdivided into classes
(2 digits) , groups (3 digits) and activity headings (4 digits). SIC-80 was introduced
to align the UK classification with that of the EC (NACE).
SIC-92 was introduced following revision of NACE.
SIC-68ORDER holds the order number of the contributor.
OTHER-CLASS is a spare field for alternative classification. For the 1983 Census
the field held the SIC68 classification. It was unused for 1984. From 1988 to 1992,
OTHER-CLASS redefined VTC. The ASCII files for these years have the VTC
inserted in the OTHER-CLASS data item, whilst the intervening years (ie 1984-87)
are filled with '9999'.
VTC (Vat Trade Classification) is a 4 digit code used to classify legal units by trade
or industry. When registering for VAT, the legal unit must select the VTC which
most closely fits the main business activity undertaken. With some exceptions the
VTC corresponds closely to the Minimum List Headings of the 1968 SIC.
COMPANY-INDUSTRY holds the industry to which the contributors company ( is
classified. This need not be the same as the establishment reference (ESTAB-REF
and CSO_REF2) in cases where the major activity of the company differs from that
of the establishment. Use of this field was only used from 1987 - 1990, although the
data item appears on the record description from 1986 until 1992, .
FORM-TYPE indicates which form should be sent to a contributor based on their
industry and employment. The FORM-TYPE is needed to identify what questions
were asked of a contributor in any particular year. The FORM-TYPE consists of a 3
digit numeric code eg 920. Appendix C lists the form-types used from 1980-1994
(form-types for 1970-1979 are currently undocumented). Some form-types with
character codes (920E and 920G) appear on the 1980 form-type matrix produced
by ACOP section. However the forms were input using a standard 3 digit form-type
(920). The codes refer to industries requiring special forms (E - Electricity, G-Gas).
If analysis is required an extract can be done using the SIC80.
Action taken.
Data retained in existing format. The changes from SIC68 to SIC80 and
subsequently to SIC92 will complicate timeseries analysis that span the transition
years. At a later stage it may be worth converting the earlier classifications.
xii.3
DATES
Purpose - record dates of processing contributor returns.
PERIOD
1970 -1979
EFF-BATCH-DATE 1970 - 1979
68
VET-DATE
1980 - present
RECEIPT-DATE
1993 - present
TAKEON-DATE
1993 - present
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Notes.
The exact purpose of PERIOD and EFF-BATCH-DATE are currently unknown.
RECEIPT-DATE records the date the form was receipted having been returned by
the contributor. TAKEON-DATE holds the date that the form details were entered
onto the system. VET-DATE records the date of the last update or record
amendment. The VET-DATE is not relevant for local units.
Action taken.
Data retained in existing format. In addition, to simplify creating extracts of more
than one year, a 4 digit YEAR field has been added to all data sets.
xii.4.
EMPLOYMENT
Purpose - to hold employment totals supplied from the register and from contributor
returns.
REGISTER-EMP
1970-1979
SEL-EMP
1980-PRESENT
RET-EMP
1980-PRESENT
Notes.
REGISTER-EMP and SEL-EMP hold the employment totals supplied from the
register. RET-EMP holds the employment totals supplied by the contributor and
thus is only available for selected contributors. There appears to be no equivalent
item for returned employment for the EFF years (1970-1979) although it may have
been held as one of the data values.
Action taken.
Data retained in existing format. In addition, sizebands based on selected data
have been added to the entire series (1970 - present). The data item is called
STRATUM and has the following values:1970 to 1994
All industries (except Construction)
1
Selected Employment < 10
2
Selected Employment < 20
3
Selected Employment < 50
4
Selected Employment < 100
5
Selected Employment < 300
6
Selected Employment > 299
1970 to 1994
69
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Construction
1
Selected Employment < 10
2
Selected Employment < 20
3
Selected Employment < 50
4
Selected Employment < 300
5
Selected Employment > 299
1995 to present (all industries including construction)
xii.5.
1.
Selected employment < 10
2.
Selected employment < 20
3.
Selected employment < 50
4.
Selected employment < 100
5.
Selected employment < 200
6.
Selected employment > 199
STRUCTURE MARKERS
Purpose - identify singles, parents and local units.
EFF-STRUCTURE
1970-1979
2
Parent or Single
3
Local Unit
PARENT-INDIC(4)
0
Other
1
Parent
1970-1979
STRUCTURE 1980 - present
7
single
8
parent
9
local unit
Other codes were sometimes used for late returns in earlier years (pre-1989) :-
70
1
disclosive late return single
2
disclosive late return parent
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3
disclosive late return local unit
4
non-disclosive late return single
5
non-disclosive late return parent
6
non-disclosive late return local unit
Notes.
PARENT-INDIC(4) was used in conjunction with EFF-STRUCTURE in the
Extended Forms File to indicate if a contributor with EFF-STRUCTURE of 1 was a
parent or single.
Action taken.
Codes for EFF-STRUCTURE were converted to those used by the STRUCTURE
data item. The PARENT_INDIC(4) data item was then no longer needed and
excluded from the ARD. Where late return codes ( STRUCTURE codes 1-6 from
1980-1989) were found these were converted to those of the main series (ie 7,8,9).
The information on disclosive late returns is unlikely to be needed.
The ARD STRUCTURE codes are shown below:-
xii.6.
New Code
STRUCTURE
EFF-STRUCTURE/PARENT-INDIC
7 Single
7 or 1 or 4 (disc/non-disc lr)
2 and 0
8 Parent
8 or 2 or 5 (disc/non-disc lr)
2 and 1
9 Local Unit
9 or 3 or 6 (disc/non-disc lr)
3
DELETION MARKERS
Purpose - indicate if contributor has been deleted and why.
DELETION-TYPE
1980 - 1992.
0
not deleted
1
dead letter/ineffective address
2
ceased trading
3
duplicate reference
4
out of scope of census
5
dormant or NYIP
71
ANNUAL RESPONDENTS DATABASE: USER GUIDE
6
structure change
DEL_TYPE
1993 - present
Codes as for DELETION-TYPE (except that 3 & 6 will cease to be used)
Notes.
See also SELECTION-STATUS and SEL_STATUS described in SELECTION
MARKERS which also provide information on deletion. No deletion markers are
available in the EFF years (1970-1979).
Recommendation.
Data retained in existing format.
xii.7.
SELECTION MARKERS
Purpose - indicate whether contributor has supplied data and whether it has been
deleted.
EST-MARKER-X 1970 - 1979 (except 1973)
1 or 2
Returned
3 or 4 or 5
Meaning unknown but assumed to be returned (selected) data
E
Estimated
EST-MARKER-X 1973 only
A or D Returned
C or B Meaning unknown
E
Estimated
SELECTION-STATUS 1980 - 1992
0
not selected, not received
3
not selected, deleted
6
selected, not received
7
selected, constructed
8
selected, received
9
selected, deleted
SEL_STATUS 1993 - present
72
0
not selected
1
selected, not receipted
ANNUAL RESPONDENTS DATABASE: USER GUIDE
2
selected, receipted
3
selected, receipted, taken on
4
selected, not receipted, constructed
5
selected, receipted, constructed
6
selected, receipted, left for estimate
7
selected, receipted, deleted (removed from census file)
8
not selected, deleted
Notes.
Codes of 3,4,5 for 1970-1979 have been retained until their meaning can be
established. They do not therefore correspond with codes 3,4,5 for later years.
See also EST-MARKER, ESTIMATION-STATUS and RESULT-CODE described in DATA
VALUE MARKERS.
Action taken.
A single set of codes was created for all years by matching early codes to post-93
codes. No action has yet been taken on codes of 3,4,5 in 1970-1979. Unknown
codes B and C for 1973 have also been retained :The ARD SEL_STATUS codes are shown overleaf :-
SEL-STATUS
SELECTIONSTATUS
EST-MARKERX (except 1973)
EST-MARKERX (1973)
B
n/a
n/a
B
C
n/a
n/a
C
0
0
E
E
1
6
n/a
n/a
2
n/a
n/a
n/a
3
8
1,2,3*
A,D
4
7
4*
n/a
73
ANNUAL RESPONDENTS DATABASE: USER GUIDE
5
n/a
5*
n/a
6
n/a
n/a
n/a
7
9
n/a
n/a
8
3
n/a
n/a
*Codes of 3,4,5 for 1970-1979 have been retained until their meaning can be
established. They do not correspond
with codes 3,4,5 for later years.
xii.8.
ORGANISATION MARKERS
Purpose - indicate type of organisation
EFF-ORGANISATION
1970 - 1979
0
Unknown
1
Co-op
2
Incorporated
3
Local authority
4
Public corporation
5
Nationalised industry
6
Central Government Department
7
Other
ORGANISATION
1980 - 1983
Codes as for EFF-ORGANISATION
ACOP-STATUS-CODE
74
1984 - present
1
Company
2
Sole proprietor
3
Partnership
4
Public corporation
5
Central government body
6
Local authority
7
Non-profit making body
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Notes.
The EFF-ORGANISATION code 0 may be due to missing data. In 1979, codes
20,40 and 79 were found for a small number of contributors.
Action taken.
A single set of codes was created for all years. Where codes of 0,20,70,40 were
found they were left unchanged. ORGANISATION and EFF-ORGANISATION
codes 2 and 7 were combined with 1 and 7 from ACOP-STATUS-CODE
respectively.
ARD ACP_STAT codes are shown below:-
New Code
EFF-ORGANISATION
& ORGANISATION
ACOP-STATUS-CODE
0 Unknown
0
n/a
1 Incorporated OR Company
2
1
2 Sole proprietor
n/a
2
3 Partnership
n/a
3
4 Public corporation
4
4
5 Central government body
6
5
6 Local authority
3
6
7
7
8 Co-op
1
n/a
9 Nationalised industry
5
n/a
7 Other OR
making body
xii.9.
Non-profit
ACTIVITY MARKERS
Purpose - indicate activity of contributor
PROD-STAT 1980 - present
75
ANNUAL RESPONDENTS DATABASE: USER GUIDE
1
Production
2
Transport
3
Other uses
4
Head office
5
Other office
6
Not yet in production
Notes.
For ACOP purposes, codes 1 & 2 and codes 4 & 5 are defined as Production units
and Office units respectively. The value should always be 1 for singles and parents
because they must be production units to be in the Census in the first place. No
Activity markers are available in the EFF years (1970-1979).
Recommendation.
Data retained in existing format.
xii.10. RECORD STATUS MARKERS
Purpose - indicate processing status of contributor.
DATA-VET-FLAG
1970 - 1979
RECORD-STATUS
1980 - present
0
Valid record
1
Suspect record (vet errors)
5
Not yet run (default at initiation)
6
Deleted
7
Suspect Override
8
Cleared Suspect Value
9
Vet required (not yet vetted)
Notes.
The exact purpose of DATA-VET-FLAG is uncertain. Although described in early
documentation, the position it occupied (char 54) is described in the EFF record
description as filler.
RECORD-STATUS indicates the worst vet status allocated to any of the data
values on the record. Codes 6,7 & 8 exist on the Data Dictionary (DDS), but are
not used by ACOP.
76
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Action taken.
RECORD-STATUS retained in existing format. DATA-VET-FLAG was not included
in the ARD.
xii.11. ESTABLISHMENT AND UNIT REFERENCE CHECK DIGIT MARKERS
Purpose - Check digit for establishment and unit references to ensure correct
selection of contributors before being accepted for processing.
ESTAB-CD
1970 - 1992
A through to Z
UNIT-CD
1970 - 1992
A through to Z
Notes.
None.
Action taken.
Data retained in existing format.
xii.12. DIRECTORY OF MANUFACTURING BUSINESSES MARKERS
Purpose - indicates agreement (or otherwise) to inclusion in classified list of
business addresses. The question appears as number 8 in the ACOP inquiry form.
CONFIDENCE
1970 - 1992
0
initial setting
1
indicates agreement
2
permissions not given
Notes.
From 1993 onwards this data item has been held as a data value for question no.
8.
Action taken.
Data retained in existing format. For 1993 onwards question no 8 will contain the
marker.
xii.13. INDUSTRY TEAM MARKERS
Purpose - route prints to appropriate ACOP team (according to 3 digit group).
INDUSTRY-TEAM
1980 -1992
77
ANNUAL RESPONDENTS DATABASE: USER GUIDE
B
130,140,411-429(less 412&419) ,475,483
C
257-314,320-327,329,330,461-466
D
111,120,152-170,328,341-374(less 3480)
E
431-456,472,481-482,491-494
F
210-256,316,467-471,495
X
ACOP Ops Control
Notes.
The only likely use of this item would be a convenient way of selecting these
particular sets of 3 digit groupings.
Action taken.
Data retained in existing format.
xii.14. RESPONSE CHASING MARKERS
Purpose - indicate whether an establishment warrants special response chasing.
KEY-RESPOND
1980 - 1990
0
not a key responder
1
key responder
ENFORCEMENT
1991 - present
0
not necessary to enforce
1
enforce
Notes.
ENFORCEMENT and KEY-RESPONDER
differ somewhat in definition.
ENFORCEMENT = 1 where a responder has not responded for a number of years.
KEY_RESPONDER = 1 where employment is > 300 OR selected employment is >
5% of the sizeband total of selected employment for that industry. For the purposes
of the ARD they may be regarded as equivalent. No response chasing markers are
available in the EFF years (1970-1979).
The marker could be useful in identifying large companies and/or important
contributors. Information related to enforcement may however be commercially
sensitive.
Action taken.
Data retained in existing format.
xii.15. LATE RETURN MARKERS
78
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Purpose - indicate late return to estimation system
LATE-RET-MKR
0
normal
1
late
LATE_RETURN
1980 - 1992
1993 - present
Codes as for LATE-RET-MKR
Notes.
No response chasing markers are available in the EFF years (1970-1979). See
also STRUCTURE which may also provide information of late returns (1980-1989
only).
Action taken..
Data retained in existing format.
xii.16. REGIONAL MARKERS
Purpose - to indicate geographical location of contributors.
EFF-AREA-CODE
1970 - 1979
Can be broken down as:Chars 1- 1 ASSISTED-AREA
Chars 2 - 3 REGION
Chars 4 - 8 LO-CODE
LOCAL-AUTH-CODE 1970 - 1979
Can be broken down as :Chars 1 - 2 REGION
Chars 3 - 6 LO-CODE
REGION
1980 - present
1
South East
2
East Anglia
3
South West
4
West Midlands
5
East Midlands
79
ANNUAL RESPONDENTS DATABASE: USER GUIDE
6
Yorkshire & Humberside
7
North West
8
North
9
Wales
10
Scotland
11
Northern Ireland
12
England (England(Gas, Coal, Electricity) in some early documentation)
13
Great Britain (Wales(Gas, Coal, Electricity)in some early documentation)
14
Scotland(Gas, Coal, Electricity) in some early documentation
Nb. In the years 1970-1973, other codes were found for a small number of
contributors which are currently unknown:1970 - 23,30,72
1971 - 82
1972 - 20
1973 - 23,27,32,62,72,82,92
GOVERNMENT OFFICE REGION 1996 to present
1 South East
2 Eastern
3 London
4 South West
5 West Midlands
6 East Midlands
7 Yorkshire & Hunberside
8 Merseyside
9 North West
10 North East
11 Wales
12 Scotland
13 Northern
80
ANNUAL RESPONDENTS DATABASE: USER GUIDE
14 England
15 Great Britain
LA-CODE
1980 - present
see Appendix D for list of codes.
LO-CODE
1980 - present (described as TTWA + FILLER for 1986 & 1987)
see Appendix E for list of 1984-1985 codes. See Appendix F for list of 1970-1983
codes matched to 1984-1985 codes.
ASSISTED-AREA
0
Not assisted
5
Intermediate
6
Development
ASSIST-AREA
1980 - 1992 (See Appendix G for more details)
1993 - present (See Appendix G for more details)
Codes as for ASSISTED-AREA
ENT-ZONE
1980 - 1992
0
Not in enterprise zone
1
Located in enterprise zone
POST-CODE 1984 - present
Notes.
The lowest level of regional data available is POST-CODE.
ASSISTED-AREA indicates whether or not an establishment or local unit is in one
of the government assisted area categories. The information is picked up from the
Locational Code Directory from the relevant postcode.
ENT-ZONE indicates whether or not an establishment falls within a government
enterprise zone. Establishments in such a zone should not be sent a Census form.
LA-CODE is a sub-division of the region to which the establishment belongs; in
England and Wales, local authority, county and district, in Scotland, region and
district.
The Local Authority Code comprises the following:Region eg
County
District
03 = South West
01 = Avon
05 = Bath
(held as REGION)
(leftmost 2 digits of LA-CODE)
(rightmost 2 digits of LA-CODE)
81
ANNUAL RESPONDENTS DATABASE: USER GUIDE
The Local Office Code was held on the Census file until 1985 and comprises the
following:Region
03
= South West
Sub-region (ESD code)
27
Dept of Emp Exchange
LO-CODE)
(held as REGION)
= South West
(leftmost 2 digits of LO-CODE)
205 = Penzance
(rightmost 3 digits of
LO-CODE comprises a further regional subdivision for establishments based on
Dept of Employment catchment areas.
The 1984 brief refers to LO-CODES and LA-CODES "being replaced by a
locational code derived from postcodes". However the change in LO-CODE did not
take place until 1986 and the LA-CODE does not appear to have changed at all.
In 1984 the LO-CODES changed to a new series (still in LO-CODE format) which
were also used in 1985 before being replaced by the TTWA code series in 1986.
From 1986 onwards the Travel to Work Area Code was held instead of the Local
Office Code. This holds the same regions in a more compact code (occupying the
leftmost 3 digits of LO-CODE). For example the Local Office and Travel to Work
Area codes for Tiverton are 03-27-222 and 128 respectively.
In the ASCII files for 1986 and 1987 the TTWA codes all have 2 trailing zeroes.
Action taken.
Regional data held for the EFF years (1970 - 1979) was converted to the data
items used in later years. REGION, LO-CODE and ASS_AREA were obtained
from EFF_AREA. LA_CODE was obtained from LOC_AUTH. In some cases the
resulting LA_CODE and LO_CODE are 0.
The LO_CODE codes are held in the TTWA variable in the ARD. At a later stage it
might be possible to standardise the codes held in TTWA if the Local Office Codes
used prior to 1986 could be converted to their equivalent TTWA codes. For 1984 to
1985 this should be possible using the POST_CODE table currently held in
INGRES which has details of TTWA etc for each postcode. The difficulty will arise
in trying to convert the pre-84 data for which there are no postcodes and a different
set of LO-CODES which are currently undocumented..
ASSIST-AREA, ASSISTED_AREA, ENT-ZONE and POSTCODE were retained in
existing format. Trailing zeroes were removed from 1986 and 1987 TTWA codes.
xii.17. FOREIGN OWNERSHIP MARKERS
Purpose - identify foreign owned units and country of ownership.
FOREIGN-OWN-CODE
1970 - 1979
see Appendix H for a list of codes
AFO-CODE
1980 - 1983
see Appendix H for a list of codes
82
ANNUAL RESPONDENTS DATABASE: USER GUIDE
FO-CODE
1984 - present
see Appendix H for a list of codes
Notes.
The foreign ownership codes were changed to 3 digit codes after 1983 to
accommodate many more countries. They were revised again for the 1988 ACOP
inquiry and again for 1994 onwards. The 1988 revision consisted of code changes
for some countries and the addition of new companies. The 1994 changes
comprised additional countries being added and some minor updates of country
names (eg New Hebrides became Vanuatu).
It is worth noting that in the current list there are two possible codes for UK
ownership.
See also RECEIPTS-OVERSEAS and PAYMENTS-OVERSEAS described in
FINANCIAL MARKERS.
Action taken.
Foreign ownership codes were converted to the 1994 version. The main problem
with this approach was where countries had changed in names (eg Yugoslavia to
Former Yugoslavia). Such changes where made have been documented in the
listings of foreign ownership codes (Appendix H) using footnotes to indicate name
changes etc.
xii.18. FINANCIAL MARKERS
Purpose - provide information on leasing and overseas payments/receipts.
RECEIPT-MKR
1984
1
Total receipts from overseas > 1,000 for royalties or copyright, sound
recordings, patents, licences, trademarks, manufacturing rights, research and
technical knowhow.
2
Total receipts from overseas <= 1,000 for royalties or copyright, sound
recordings, patents, licences, trademarks, manufacturing rights, research and
technical knowhow.
RECEIPTS-OVERSEAS
1987 - 1989
1
Total receipts from overseas > 10,000 for overseas services excluding
royalties.
2
Total receipts from overseas <= 10,000 for overseas services excluding
royalties.
PAYMENTS-MKR
1984
1
Total payments to overseas > 1,000 for royalties or copyright, sound
recordings, patents, licences, trademarks, manufacturing rights, research and
technical knowhow.
83
ANNUAL RESPONDENTS DATABASE: USER GUIDE
2.
Total payments to overseas <= 1,000 for royalties or copyright, sound
recordings, patents, licences, trademarks, manufacturing rights, research and
technical knowhow.
PAYMENTS-OVERSEAS
1987 - 1989
1
Total payments to overseas > 10,000 for overseas services excluding
royalties.
2
Total payments to overseas <= 10,000 for overseas services excluding
royalties.
LEASING-MKR
1984
1
Capex does
include goods for letting out on hire or leasing.
2
Capex doesn't include goods for letting out on hire or leasing.
FINANCIAL-LEASING
1987 - 1989
0
Not selected, not responded or question not completed on form.
1
Capex does
2
Capex doesn't include assets for leasing out on a finance lease.
include assets for leasing out on a finance lease.
Notes.
These data items appeared as questions on the ACOP inquiry. Owing to the format
of the reply (1 or 2), the response was stored in the indicative portion. RECEIPTMKR and RECEPTS-OVERSEAS appeared as questions 803 and 805
respectively in the relevant years, PAYMENTS-MKR and PAYMENTSOVERSEAS appeared as questions 804 and 806, and LEASING-MKR and
FINANCIAL-LEASING appeared as questions 801 and 802.
Although RECEIPT-MKR, PAYMENTS-MKR and LEASING-MKR are listed on
record descriptions from 1984 to 1986, the questions were only asked in 1984.
RECEIPT-MKR and RECEIPTS-OVERSEAS are NOT equivalent, neither are
PAYMENTS-MKR and PAYMENTS-OVERSEAS. However for convenience they
occupy the same character position in the ARD ASCII files. The items LEASINGMKR and FINANCIAL-LEASING can be treated as equivalent although the
wording is slightly different.
Action taken.
Data retained in existing format. Although NOT equivalent RECEIPT-MKR and
RECEIPTS-OVERSEAS and PAYMENTS-MKR and PAYMENTS-OVERSEAS
were stored in RECS_OS and PAYS_OS variables respectively in the ARD..
xii.19. REPORTING UNIT MARKERS
Purpose - To monitor effects of change to company reporting in 1987.
COMPANY-ESTREPSTAT
84
1987- 1990
ANNUAL RESPONDENTS DATABASE: USER GUIDE
0
(not known at present)
1
(not known at present)
COMPANY-RESPONSE
1987- 1990
0
No response (not selected, not returned etc)
1
Yes figures would have been different
2
No figures would not have been different
COMPANY-ESTAB
1987- 1990
0
(not known at present)
1
(not known at present)
Notes.
Use of these fields was limited to 1987 - 1990, although the data items appear on
the record description from 1986 until 1992. See also COMPANY-INDUSTRY
described in INDUSTRY DATA ITEMS which also provides information on the
change in reporting unit.
COMPANY-ESTREPSTAT indicates whether a responder completes a form as an
establishment or a company.
COMPANY-RESPONSE contains the response to Question 807 which asked
whether the change to company reporting had made any difference to the figures
returned by the contributor.
COMPANY-ESTAB indicated that this establishment should be regarded as the
company establishment for a multi-establishment company. This means that its
reference would be used when creating a company record.
Action Taken.
Data retained in existing format.
xii.20. MISC
Purpose - Data items that do not fit any particular category.
NIL-CODE
1970 - 1979
SUBMISSION-CODE 1970 - 1979
DUMMY-MARKER
1970 -1979
EFF-BATCH-NO
1970 -1979
POS-IN-BATCH
1970 -1979
UNIT-SEQUENCE-NO
1970 - 1979
85
ANNUAL RESPONDENTS DATABASE: USER GUIDE
NO-OF-DATA-GROUP
1970 -1979
LINK-1
1980 - 1992
LINK-2
1980 - 1987
LINK-3
1980 - 1987
ATYP-MARKER
1988 - 1992
1988-1992
(codes unknown at present)
1992 - present
POLLUTION-EXPEND
1990 - 1992
0
No response, initial setting
1
Yes - expenditure has been incurred
2
No - expenditure has not been incurred
RD-EMPLOYMENT 1992 only
1
Yes
2
No
PA-LOCK
0
not locked
1
locked
ALT-SIC
1991 - 1992
AUDIT-NO
1988 - 1992
STOP-REMINDER
1993 - present
1
Stop first Reminder
2
Stop second Reminder
3
Stop third Reminder
A
Stop all Reminders
PURCHASES-SEL-MKR
(codes unknown at present)
Notes.
86
1990 - present
1994 - present
ANNUAL RESPONDENTS DATABASE: USER GUIDE
The exact purpose of SUBMISSION-CODE, DUMMY-MARKER, EFF-BATCHNO, POS-IN-BATCH, UNIT-SEQUENCE-NO are currently unknown, though it
seems likely that they contain data processing information, and are not required in
the ARD.
NO-OF-DATA group holds the number of questions held in the contributor record.
This is required since the EFF contains variable length records, unlike the Census
file where the record length is fixed and up to 170 data values are held.
LINK-1, LINK-2 AND LINK-3 were used in company reporting to all establishments
within a company. Although linking establishments together into enterprises will be
of interest to researchers, this can be done using the ENT-REF5, ENT-REF, and
ENT-GROUP-REF items. The link items are therefore not be included in the ARD
ASCII files.
ATYP-MARKER - set to mark establishments thought to be atypical, ready for
highlighting at estimation.
POLLUTION-EXPEND - tick box for indicating expenditure on pollution.
RD-EMPLOYMENT - New heading for 1992 only to show whether a company
employs Research & Development Staff.
ALT-SIC - Alternative SIC used to store previous SIC. Introduced for 1991 when
the Fur Trade (4560) merged with 4533. SIC-*) holds 4533 and ALT-SIC holds
4560.
AUDIT-NO - incremented by 1 each time a contributor record is amended, so that
every action on the record is referenced uniquely and in sequence on the audit file.
PA-LOCK indicates that an establishment form has been "locked" to prevent any
further amendment by the online update system.
STOP_REMINDER - controls the despatch of reminders to contributors.
PURCHASES-SEL_MKR - Indicates a contributor's selection status for the
Purchases Inquiry.
Action taken.
The following codes were all excluded from the ARD:NIL-CODE
SUBMISSION-CODE
DUMMY-MARKER
EFF-BATCH-NO
POS-IN-BATCH
UNIT-SEQUENCE-NO
LINK-1
87
ANNUAL RESPONDENTS DATABASE: USER GUIDE
LINK-2
LINK-3
The other codes described in this section were retained in existing format.
xii.21. DATA VALUES AND VET/ESTIMATION STATUS INDICATORS
Purpose - to hold values returned by contributors together with associated status
markers and question numbers
DATA-VALUE 1970-present (see Appendix B for a list of codes held for each year)
Notes.
This item holds the value for a question for a contributor. The value may be
returned, estimated, constructed apportioned or derived.
In the EFF (1970-1979), the question number was held in an adjacent field (see
QUESTION_NO below). The ASCII format files for 1970 - 1979 also have this
format and therefore differ from the format of files for 1980-1992.
The question numbers for data values held on the Census file (1980 - 1992) are
indicated by the position of the value in the record. There are tables available for
each year from which the question number can be obtained.
From 1993 onwards the data values have been held in a separate table linked to
the indicative portion of the file by CSO_REF. The question number is held in an
adjacent field (see QUESTION-CODE below).
Action taken.
Each ARD data value is held a variable named according to the question number.
Eg. The value related to Question 11 will be stored in field Q11. The tables of
question numbers were used to create the appropriate fields for the Census file
years. The values held in QUESTION-NUMBER and QUESTION-CODE were used
to create field names for data held on the EFF and INGRES systems respectively.
QUESTION-NUMBER
QUESTION-CODE
1970-1979
1993-present
Notes.
As described above, the above markers indicate the question number relating to
the adjacent data value. They are no longer be required in the ARD since the
question number is identifiable from the field name.
EST-MARKER
0
Returned
<> 0
Estimated (No detailed description of codes found yet)
ESTIMATION-STATUS
88
1970-1979
1980-1992
ANNUAL RESPONDENTS DATABASE: USER GUIDE
0
Inappropriate code for this form type - no value expected therefore none
estimated
1
Manual exclusion of value from averages - manual exclusion of good (vet
stat 9) data
2
Estimated value - for short forms, non-responders and non-selected
3
Derived value from other codes - calculated from parent values &
apportioned to Lus
4
Apportioned value from parent record - for atypical returns etc
5
Value excluded automatically - for atypical returns etc
6
Forced inclusion of value after automatic exclusion of value regarded as
atypical
7
Non-compulsory; not estimated for if not returned - any returned value may
not to be used to calculate averages
8
Non-compulsory; not estimated for if not returned - any returned value will
be used to calculate averages
9
Must contain a returned value - else error message. Any returned values
are used in estimation
RESULT-CODE
1993-present
0
Initial value, no action taken
1
Possible atypical (marker only)
2
Computer exclusion from averages
3
Manual exclusion from averages
4
Manual exclusion (was marked)
5
Manually forced inclusion in averages
6
Manually included was excluded
7
Manually included was excluded (marked)
8
Value has been estimated
9
Value has been apportioned
Notes.
This marker indicates what action has been taken on a question value during
processing. It is useful if grossed data is required since the atypicals can be easily
identified. It also reveals whether the value concerned is estimated, derived or
actually returned.
89
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Action taken.
Data retained in existing format. Ideally one would create a single set of codes, but
this will be difficult because:
detailed non-zero codes for the EFF years are unknown at present
it is difficult to match the codes used in the Census file with those of the
INGRES held data
millions of edits would be involved to standardise the whole timeseries
VET-STATUS 1980-present
0,2-4 Not used
1
Suspect value; RECORD-STATUS will be set to 1
5
No data present as yet; initial setting
6
Value previously present now deleted
7
Suspect value override; data found to be reasonable
8
Cleared suspect value; ie value amended or resubmitted
9
Good data
Notes.
Indicates the type of data value currently held. No vet marker exists for the EFF
years.
Action taken.
Data retained in existing format.
Appendix I
REFERENCES
ACOP User guide (1992) - held by ACOP IS Support.
FORMTYPES folder held in Rm D.230. (Covers 1980 - 1992).
DDS DATA ITEMS folder held in Rm D.230 (prints from Data Dictionary held on mainframe
including EFF data items).
LOTUS SPEADSHEETS held in D:\AMD\METADATA
INGRID.WK4 - Indicative data table (Appendix A)
PROGRESS.WK4
90
- Record of file conversions (Appendix A)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
QUES7079.WK4
- Question list 1970 - 1979 (Appendix B)
QUES8094.WK4
- Question list 1980 - 1992 (Appendix B)
FORMTYPE.WK4
- Formtype by Question matrix 1980 - 1992 (Appendix C)
FOCODES.WK4
- Foreign Ownership codes (Appendix H)
FURTHER READING
PAPERS WRITTEN FOR ARD PROJECT held in D:\AMD\METADATA
A comparison of the requirements of EUROSTAT and ACADEMIC Researchers for CSO
microdata COMPARE.WPD
An investigation of alternative hardware options to hold a timeseries of ACOP
disaggregate contributor data. - HARDWARE.WPD
An examination of requests for special analyses of CSO contributor data received by
ACOP publications and analysis. - SPECIAL2.WPD
Report on a visit to CES (Center for Economic Studies) at the Bureau of the Census,
Washington DC. - USATRIP.WPD
PAPERS ON MICRODATA (copies available from Rm. D.230)
Longitudinal Research Database Technical Documentation Manual 1992', Center for
Economic Studies, US Bureau of the Census.
The Importance of Establishment Data in Economic Research , R.H.McGuckin, CES-93
August 1993, Center for Economic Studies, US Bureau of the Census.
Establishment Microdata for Economic Research and Policy Analysis: Looking Beyond the
Aggregates, Journal of Business &Economic Statistics, January 1995, Vol. 13, No 1.
Overview of CES Research Programs 1993-94', December 1994.
Technology, Economic Growth and Employment: New research from the Department of
Commerce, US DEPARTMENT OF COMMERCE, Econonics and Statistics Administration,
Office of the Chief Economist December 1994.
VERSION CONTROL
Document name
: ARDGUIDE.wpd
Latest version held in
: C:\AMD\METADATA\DOCS\ARDGUIDE.WPD
PC mnemonic
: AP074
Room
: D.230
Backup
Rom
: q:\amd\live\metadata\docs\ARDGUIDE.wpd
NPT6 CD91
ANNUAL RESPONDENTS DATABASE: USER GUIDE
(current version held= 23/7/96)
Updated on
Amendments
affecting ARD data
: 23 July 1996
: Added section 9 - $Summary of changes
since 1970".
Updated on
:13 January 1997 (12:15pm)
Amendments
: Amended 12.16 - $Regional Markers# to alter description
of Regions 12
-14
Updated on
: 23 April 1999 (10.00am)
: Text updated to include references to 1995 and 1996
plus other
micellaneous amendments as listed in
Appendix L
92
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.8
Data cleaning
93
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.9
Clearing results for release
95
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.10 SIC Codes?
Refer to web for SIC92? What about SIC80?
97
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.11 ARD Contact details
99
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.12 Country Codes
Country of Ownership Codes: from 1994
Foreign ownership code
000
005
009
013
014
017
021
025
027
028
029
033
034
035
037
041
049
053
061
065
069
071
072
073
077
081
084
085
089
093
096
100
101
113
117
121
127
131
135
137
141
149
150
153
157
161
165
166
169
173
Country of ownership
Not known
Afghanistan
Albania
Algeria
Alderney
American Samoa
Andorra
Angola
Anguilla
Antigua
Egypt
Argentina
Aruba
Ascension Island
Australia
Austria
Bahamas
Bahrain
Bangladesh
Barbados
Belgium
Belize
Benin
Bermuda
Bhutan
Bolivia
Bophuthatswana
Botswana
Brazil
Virgin Islands (UK)
Brunei
Bulgaria
Burkina Faso
Burundi
Cameroon
Canada
Cape Verde
Cayman Islands
Caroline Islands
Cayman Islands
Central African Republic
Chad
Channel Islands
Chile
China People’s Republic
Taiwan (China Rep of)
Colombia
Ciskei
Comoros Islands
Congo
101
ANNUAL RESPONDENTS DATABASE: USER GUIDE
175
177
179
181
185
190
197
198
199
201
209
213
217
219
221
225
229
233
237
241
245
249
253
261
263
269
273
277
285
289
291
293
297
301
302
303
305
309
313
317
321
325
329
333
337
341
345
349
351
353
357
361
365
369
371
377
379
381
383
102
Cook Islands
Costa Rica
Croatia
Cuba
Cyprus
Czech Republic
Denmark
Djibouti, Republic of
Dominica, Commonwealth of
Dominican Republic
Ecuador
El Salvador
Equatorial Guinea
Estonia
Ethiopia
Faero Islands
Falkland Islands
Fiji
Finland
France
French Guiana
French Polynesia
Gabon
Gambia
Gazankula
Germany
Ghana
Gibraltar
Greece
Greenland
Grenada
Guadeloupe
Guam
Guatemala
Guernsey
Guinea-Bissau
Guinea
Guyana
Haiti
Honduras
Hong Kong
Hungary
Iceland
India
Indonesia
Iran
Iraq
Republic of Ireland
Isle of Man
Israel
Italy
Ivory Coast
Jamaica
Japan
Jersey
Jordan
Kangwane
Kenya
Kerghelen Island
ANNUAL RESPONDENTS DATABASE: USER GUIDE
385
387
389
393
397
398
399
401
405
406
407
409
413
417
421
425
429
433
441
445
449
453
457
461
469
473
477
481
489
491
497
499
501
505
509
510
511
513
517
519
521
525
529
537
541
545
549
551
552
553
561
565
569
573
581
585
589
597
601
Kampuchea
Kiribati
North Korea
South Korea
Kuwait
Kwandebelle
Kwazulu
Laos
Lebanon
Lebowa
Leeward Island
Lesotho
Liberia
Libya
Liechtenstein
Lithuania
Luxembourg
Macau
Madagascar/Malagasy
Malawi
Malaysia
Maldives
Mali
Malta
Marshall Island
Martinique
Mauritania
Mauritius
Mexico
Micronesia, Federal State of
Monaco
Montenegro
Montserrat
Morocco
Mozambique
Myanmar (Burma)
Namibia
Naura
Nepal
N’kwavoma
Netherlands
Netherlands Antilles
New Caledonia
New Zealand
Nicaragua
Niger
Nigeria
Northern Mariana Islands
Norfolk Islands
Norway
Oman
Pakistan
Panama
Papua New Guinea
Paraguay
Peru
Philippines
Poland
Portugal
103
ANNUAL RESPONDENTS DATABASE: USER GUIDE
609
612
613
614
617
620
622
623
626
629
633
635
637
641
649
653
657
661
663
665
669
677
678
679
680
681
685
693
701
705
709
713
717
721
725
727
729
733
741
744
745
749
757
761
763
765
767
769
771
777
783
805
807
813
816
819
821
829
833
104
Puerto Rico
Qwa-Qwa
Qatar
Redonda
Reunion Island
Romania
CIS (Russian Federation)
Rwanda
St. Helena
Saint Kitts-Navis
St Lucia
St Pierre
Saint Vincent
Samoa Western
San Marino
Sao Tome and Principe
Saudi Arabia
Senegal
Serbia
Seychelles
Sierra Leone
Singapore
Slovenia
Solomon Islands
Slovakia
Somalia
South Africa
Spain
Sri Lanka
Sudan
Suriname
Swaziland
Sweden
Switzerland
Syria
Tahiti
Tanzania
Thailand
Togo
Transkei
Tonga
Trinidad and Tobago
Tunisia
Turkey
Turkish Rep of N. Cyprus
Turks and Caicos Islands
Tuvalu
Uganda
Ukraine
United Arab Emirates
United Kingdom
United States of America
United States Virgin Islands
Uruguay
Vanuatu
Vatican City
Venezuela
Vietnam
Virgin Islands (US)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
841
853
861
865
869
873
897
898
899
900
901
905
910
915
916
917
918
919
920
925
930
935
937
938
940
942
943
944
945
946
950
952
955
965
970
Wallis and Futu Islands
Yemen Arab Republic
Former Yugoslavia
Zaire, Republic of Zaire
Zambia
Zimbabwe
Abu Dhabi
Ajman
Al Ain
Armenia
Admiralty Islands
Azerbaijan
Bosnia and Herzegovina
Byelorussia
Christmas Island
Dubai
Fujairah
Eritrea
Georgia
Kazakhstan
Kirghizia
Latvia
Mayotte
Midway Island
Moldavia
Ras Al Khaimah
Sharjah
South Georgia
Tajikstan
Tokelau
Turkmenistan
Umm Al Qaiwain
Uzbekistan
Macedonia
Wake Island
105
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.13 Region codes
County
Old IDBR Standard Statistical Region
Region Code
08
North (A)
Others
Cumbria
07
North West (B)
Others
Merseyside
06
Yorkshire & Humberside (C) All
05
East Midlands (C)
All
04
West Midlands (E)
All
02
East Anglia (F)
All
01
South East (G)
Essex, Beds,
Herts
00
(Incl Greater London (H)
Greater London
01
Others
03
South West (J)
All
09
Wales (W)
All
10
Scotland (x)
All
11
Northern Ireland (Y)
All
12
Non UK (Z)
N/A
For 1996 = REGION
From 1997 N/A
IDBR GOR Code*1
AA
BA
BB
CB
DC
ED
FE
GF
GG
HH
JG
KJ
WW
XX
YY
ZZ
For 1996 N/A
For 1997 = GOCODE
For 1998 = REGION
For 1999 = REGION
OLD GOR*2
Government Office Regions
North East (A)
North West (B)
10
9
Mersyside (C)
Yorkshire and the Humber (D)
East Midlands (E)
West Midlands (F)
Eastern
8
7
6
5
2
London (H)
South East (J)
South West (K)
Wales (X)
Scotland (X)
Northern Ireland (Y)
Non-UK (Z)
3
1
4
11
12
13
N/A
For 1996= GOREG
For 1997= GOREG
From 1998 N/A
*1 GOR is equivalent to NUTS 1
107
ANNUAL RESPONDENTS DATABASE: USER GUIDE
3.14 Question lists
Also add 1970-1979 table
No.
Question
First
Last
1
Units not yet in production (y/n) *
1993
1997
2
3
4
5
6
7
8
1994
1995
1993
1996
1996 (103)
1980
1980
1997
1997
1996 (101)
1996 (102)
1992
1992
1980
1992
1992
1995
1995
1988
1995
1995
1980
1995
1989
1995
23
93
Showroom activity (nb used in 1975 for 'New LA code')
Enterprise number
Industry number (SIC 1968)
Activity heading (SIC 80)
NACE Number
Foreign Ownership Code
Directory of manufacturing business marker * ('nb. used for
'Confidence Code' prior to 1986)*
Local office area code
Indicative Employment
Period covered by the return - from
Period covered by the return - to
Organisation code
Period covered by the return - from (STOCKS)
Period covered by the return - to (STOCKS)
Capital expenditure - date
Period covered by the return - from (CAPEX)
Period covered by the return - to (CAPEX)
Operation of retail shops by bakers
Generating stations
Time taken to complete form : hours (nb. used for UNKNOWN
prior to 1980)
Time taken to complete form : minutes
Net capex per head
1995
1980
1995
1997
130
137
138
139
140
Net capex per head: point in time (1997 only)
Total costs
Net NYIP capex - Land & Buildings
Other NYIP net capex
Net NYIP net capex
Gross wages and salaries (% is code 920)
1997
1995
1980
1980
1980
1980
1997
1995
1995
1995
1995
1997
141
142
Gross wages and salaries (including outworkers remuneration)
Total labour costs (excluding Jobbers)
1980
1980
1995
1997
143
Sales of products manufactured, work done and industrial
services rendered (derivn ch 93)
Total sales (derivn ch 93)
Purchases of raw materials (drpd 93 equiv 734 )
Purchases of raw materials and industrial services received
(derivn ch 92)
Intermediate consumption of materials and services (derivn ch
92)
Gross value added at market prices (excluding VAT)(derivn
chgd 93)
Gross value added excluding all taxes (derivn chgd 93)
1980
1995
1980
1980
1980
1995
1992
1997
1980
1997
1980
1997
1980
1997
1997
1997
9
10
11
12
13
14
15
17
18
19
20
21
22
144
145
146
147
148
149
Gross value added excluding all taxes per head: average
during year (1997 only)
Number
Changes
1996 (801)
1997 (801)
1996 (802)
1997 (802)
1997 (942)
1996 (299)
1997 (299)
1996 (861)
1997 (861)
1996 (809)
1997 (809)
1996 (810)
1997 (810)
1996 (811)
1997 (811)
1996 (812)
1997 (812)
1997 (940)
109
ANNUAL RESPONDENTS DATABASE: USER GUIDE
150
151
Gross value added excluding all taxes per head: point in time
(1997 only)
Total plant and hire
1997
1997
1997 (946)
1980
1997
1996 (813)
1997 (813)
1996 (814)
1997 (814)
1996 (867)
1997 (867)
1996 (816)
1997 (816)
1996 (817)
1997 (817)
1996 (818)
1997 (818)
1996 (399)
1997 (399)
1996 (819)
1997 (819)
1996 (820)
1997 (820)
1980
1997
152
Increase in year for work in progress and goods on hand for
sale
Increase in year for total stocks (% is code 933)
1980
1997
153
Increase during year, materials, stores & fuel **
1993
1997
154
Gross capex
1980
1997
155
Capex disposals
1980
1997
156
Total turnover (derivn ch 93)
1980
1997
157
Increase during year - Goods on hand **
1993
1997
158
Increase during year - WIP (derivn ch 92)**
1993
1997
159
160
176
Increase during year - WIP - new vessels (drpd 92) **
Increase during year - other WIP (drpd 92)**
Total stocks (beginning of year) (not derived from 92)
1980
1997
177
1980
1997
1995
1997
1995
1997
181
Total stocks (end of year) (% is code 934)(not derived from
92)
Related to other than dwellings : Total B/Y (nb. used for
UNKNOWN prior to 1980)
Related to other than dwellings : Total E/Y (nb. used for
UNKNOWN prior to 1980)
Gross value added as a percentage of net output
1980
1997
182
Gross value added as a percentage of gross output
1980
1997
183
Wages and salaries as a percentage of gross value added
1980
1997
184
Gross value added as a percentage of Total Sales and work
done
Related to dwellings : Total B/Y (nb. used for UNKNOWN prior
to 1980)
Related to dwellings : Total E/Y (nb. used for UNKNOWN prior
to 1980)
Cost of non-industrial services recd, other than rent, insurance
and bank charges (drpd 93 equiv 630)
Work in progress - end of year (drpd 92 equiv 404)
Work in progress - start of year (drpd 92 equiv 403)
Working proprietors (for CCL)
Administrative, technical and clerical employees (% is code
202)
Average number of full time employees (1996 only)
OP's and ATC's employeed in the bakehouse
Average number of part time employees (1996 only)
Operatives (% is code 928)
Jobbers - printers only (drpd 90)
Computer employees
Ratio of operatives to ATC's **
Total number of employees: point in time (1997 only)
Number of male full-time employees: point in time (1997 only)
1995
1997
1995
1997
1995
1997
1980
1992
1980
1980
1980
1980
1991
1991
1996
1995
1996
1980
1996
1980
1980
1986
1993
1997
1997
1996
1988
1996
1995
1989
1988
1995
1997
1997
178
179
185
186
190
196
197
201
202
203
205
206
207
208
110
1996 (500)
1997 (500)
1996 (599)
1997 (599)
1996 (41)
1997 (41)
1996 (42)
1997 (42)
1996 (821)
1997 (821)
1996 (822)
1997 (822)
1996 (864)
1997 (864)
1996 (863)
1997 (863)
1996 (49)
1997 (49)
1996 (50)
1997 (60)
1996 (202)
1996 (203)
1997 (50)
1997 (51)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
211
248
249
250
251
252
255
256
257
258
259
260
261
Number of male part-time employees: point in time (1997 only)
Number of female full-time employees: point in time (1997
only)
Number of female part-time employees: point in time (1997
only)
Number of working proprietors, partners & executive directors:
point in time (1997 only)
Other unpaid workers(eg family workers but excluding
voluntary workers): point in time (1997 only)
Total number of people in your employment (including unpaid
workers): point in time (1997 only)
Full-time employment: average during the year (1996/7 only)
1997
1997
1997
1997
1997 (52)
1997 (53)
1997
1997
1997 (54)
1997
1997
1997 (55)
1997
1997
1997 (58)
1997
1997
1997 (59)
1996
1997
Part-time employment: average during the year (1996/7 only)
1996
1997
Number of male full-time employees: average during the year
(1997 only)
Number of male part-time employees: average during the year
(1997 only)
Number of female full-time employees: average during the
year (1997 only)
Number of female part-time employees: average during the
year (1997 only)
Total number of employees: average during the year (1997
only)
Total employment: average during the year (1997 only)
Total employees adjusted for part-time: point in time (1997
only)
Total employment adjusted for part-time: point in time (1997
only)
Full-time employment: point in time (1997 only)
Part-time employment: point in time (1997 only)
R & D marker *
Sales of goods produced (derivn ch 93 equiv 261, used for top
10 prt) ***
Sales of goods produced **
1997
1997
1996 (928)
1997 (928)
1996 (929)
1997 (929)
1997 (930)
1997
1997
1997 (931)
1997
1997
1997 (932)
1997
1997
1997 (933)
1997
1997
1997 (934)
1997
1997
1997
1997
1997 (935)
1997 (936)
1997
1997
1997 (937)
1997
1997
1993
1993
1997
1997
1997
1997
1993
1997
1993
1997
1993
1980
1996
1997
1997
1997
Value of vessels and floating equipment completed for sale
during the year
Sales of Electricity
1980
1997
1993
1997
1997 (938)
1997 (939)
1997 (9)
1996 (826)
1997 (826)
1996 (827)
1997 (827)
1996 (828)
1997 (828)
1997 (602)
1997 (602)
1996 (303)
1997 (303)
1996 (304)
1997 (304)
1996 (302)
1997 (302)
Work done on textiles on commission
Sales of waste products, work done and industrial services
rendered
Other sales of goods of own production
1980
1980
1987
1987
1993
1997
Work done industrial services rendered, repairs etc. **
(dummy code for 1996/7)
Work of a capital nature by own staff **
Work of a capital nature by own staff
Sales of gas (1996/7 only)
Value of new construction of buildings, reservoirs etc. and
other capital goods
Sales of goods of own production (equiv 289 from 93)
1980
1995
1980
1997
Water supply charges (1996/7 only)
1996
1997
1996 (312)
1997 (312)
1996 (301)
1997 (301)
1996 (305)
1997 (305)
111
ANNUAL RESPONDENTS DATABASE: USER GUIDE
262
264
266
267
268
269
Sales of gas and other goods of own production (1995 Gas
industry only)
1995
1997
Retail turnover (inclusive of repair and installation) of goods to
public (1997 only)
Work done & industrial services rendered
Other work done on customers materials (1996 Mineral Oil
industry only)
Amounts included for work done (augments 262 or 286)
1997
1997
1995 (261)
1996 (868)
1997 (868)
1997 (300)
1980
1996
1995
1996
1996 (315)
1993
1997
1980
1997
1980
1997
Sales of goods bought and resold without processing
(merchanted)
Other services rendered
1987
1987
1990
1990
270
271
Work done on own textiles
Sales for waste products, residues, work done & industrial
services rendered (drpd 91)
Work done on textiles on commission (drpd 91)
Sales of petroleum products
1987
1980
1990
1997
272
Charge for commission refining
1980
1997
273
1980
1997
274
Other sales of goods for own production, work done &
industrial services rendered
Commission refining, work done & services rendered **
1993
1997
276
277
Road vehicle licences
Rates
1980
1980
1995
1997
Total taxes paid (1996/7 only)
1996
1997
Other amounts paid for taxes and levies (1996/7 only)
1996
1997
Subsidies received (1996/7 only)
1996
1997
1980
1980
1980
1980
1992
1992
1992
1992
285
286
Other work done & sales of own production (drpd 93)
Work done as main or direct contractor on GB sites (drpd 93)
Work done as sub-contractor on GB sites (drpd 93)
Work done in GB on contracts for erections of process plants
overseas (drpd93)
Water supplied
Repairs and maintenance and industrial services rendered
1980
1980
1990
1997
287
Sales of other goods of own production
1980
1997
288
289
Work done and industrial services rendered (drpd 91)
Sales of goods produced(drpd 93 equiv 261)** (nb. 'Work
done for related businesses overseas' in 1985)
Work done & maintenance (derivn ch 93) **
Sales, work done & construction work **
Sales of film processing, work done and industrial services
rendered (drpd 93)
Sales of waste products and residues (drpd 93)
Amounts payable for work given out (drpd 93)
Amounts payable for repairs and maintenance (drpd 93)
Rent & rates (excluding water rates) paid for industrial and
commercial buildings (drpd 93)
Wages and salaries ATC's (% is code 930)
1980
1985
1990
1985
1993
1993
1984
1995
1995
1992
1984
1984
1984
1984
1992
1992
1992
1992
1980
1995
281
282
283
284
290
291
293
294
295
296
297
301
112
1996 (308)
1997 (308)
1996 (311)
1997 (311)
1996 (310)
1997 (310)
1996 (306)
1997 (306)
1996 (307)
1997 (307)
1996 (869)
1997 (869)
1996 (830)
1997 (830)
1996 (412)
1997 (412)
1996 (400)
1997 (400)
1996 (413)
1997 (413)
1996 (414)
1997 (414)
1996 (309)
1997 (309)
1996 (312)
1997 (312)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
302
304
305
306
310
311
312
Wages and salaries OP's and ATC's in the bakehouse
Wages and salaries OP's (% is code 929)
Wages and salaries - jobbers
Salaries per ATC's (PH)
Wages per operatives (PH)
Jobbers wages per head **
Total wages per head **
1980
1980
1980
1980
1980
1993
1993
1988
1995
1989
1995
1995
1995
1997
314
315
Labour costs per head: point in time (1997 only)
Outworkers remuneration
Employers national insurance contributions
1997
1980
1980
1997
1995
1997
360
Costs: Other Expenditure (dummy code for 1996/7)
1993
1997
374
Gross output per head (PH)
1980
1997
375
Gross output per head: point in time (1997 only)
Net output per head (PH)
1997
1980
1997
1997
376
Net output per head: point in time (1997 only)
Gross value added at factor cost per head (PH)
1997
1980
1997
1997
1997
1997
400
Gross value added at factor cost per head: point in time (1997
only)
Purchases (dummy code for 1996/7)
1993
1997
401
Materials, stores and fuel B/Y (calc'd by ratio from 92)
1980
1997
402
Materials, stores and fuel E/Y (calc'd by ratio from 92)
1980
1997
403
Work in progress B/Y (calcd by ratio from 92)
1980
1997
404
Work in progress E/Y (calc'd by ratio from 92 equiv 190 in 92)
1980
1997
409
410
411
412
413
Work in progress - new vessels B/Y (drpd 92)
Work in progress - new vessels E/Y (drpd 92)
Other work in progress B/Y (drpd 92)
Other work in progress E/Y (drpd 92)
Goods on hand for sale B/Y (calc'd by ratio from 92)
1980
1980
1980
1980
1980
1991
1991
1991
1991
1997
414
Goods on hand for sale E/Y (calc'd by ratio from 92)
1980
1997
450
451
452
Total stocks on 1 January
Total stocks on 31 December
Related to other than dwellings : Total B/Y (nb. used for
UNKNOWN prior to 1980)
Related to other than dwellings : Total E/Y (nb. used for
UNKNOWN prior to 1980)
Related to other than dwellings : Materials, stores and fuel (nb.
used for UNKNOWN prior to 1980)
Related to other than dwellings : Materials, stores and fuel (nb.
used for UNKNOWN prior to 1980)
Related to other than dwellings : Other stock B/Y (nb. used for
UNKNOWN prior to 1980)
Related to other than dwellings : Other stock E/Y (nb. used for
UNKNOWN prior to 1980)
Related to dwellings : Total B/Y (nb. used for UNKNOWN prior
to 1980)
1980
1980
1995
1991
1991
1997
1995
1997
1995
1997
1995
1997
1995
1997
1995
1997
1995
1997
453
454
455
456
457
458
1996 (860)
1997 (860)
1997 (941)
1996 (204)
1997 (204)
1996 (871)
1997 (871)
1996 (838)
1997 (838)
1997 (943)
1996 (839)
1997 (839)
1997 (944)
1996 (865)
1997 (865)
1997 (945)
1996 (870)
1997 (870)
1996 (29)
1997 (29)
1996 (30)
1997 (30)
1996 (31)
1997 (31)
1996 (32)
1997 (32)
1996 (33)
1997 (33)
1996 (34)
1997 (34)
1996 (35)
1997 (35)
1996 (36)
1997 (36)
1996 (37)
1997 (37)
1996 (38)
1997 (38)
1996 (39)
1997 (39)
1996 (40)
1997 (40)
1996 (43)
1997 (43)
113
ANNUAL RESPONDENTS DATABASE: USER GUIDE
459
1995
1997
1995
1997
1995
1997
1995
1997
1995
1997
1995
1995
1995
1995
1995
1995
1980
1995
1995
1997
Cost of land and existing buildings purchased (calc'd by ratio
from 92)
Proceeds from land and existing buildings disposed of (calc'd
by ratio from 92)
Cost of new and second-hand vehicles purchased (calc'd by
ratio from 92)
Proceeds from vehicles disposed of (calcd by ratio from 92)
1980
1997
1980
1997
1981
1997
1981
1997
Computer equipment purchased
New and second-hand motor cars - aquisitions
Motor cars - disposals
Cost of other new and second-hand vehicles purchased
Proceeds from other vehicles disposed of
Plant, machinery, office equipment and other capital
equipment purchased (calc'd by ratio from 93)
Proceeds from equipment disposed of (calc'd by ratio from 93)
1986
1980
1980
1980
1980
1980
1988
1980
1980
1980
1980
1997
1980
1997
1980
1980
1980
1980
1980
1988
1991
1991
1991
1991
1991
1997
1996
1997
1988
1997
531
Total net capital expenditure for calendar year
Cost of Nuclear Fuel (drpd 92)
Proceeds from sale of nuclear fuel (drpd 92)
Mains and services - acquisitions (drpd 92)
Mains and services - disposals (drpd 92)
Total capital expenditure - acquisitions (split by ratios from 9295)
Capital expenditure - assets under finance leasing
arrangements (1996/7 only)
Total capital expenditure - disposals (split by ratios from 9295)
Cost of new building work - units NYIP
1980
1997
532
Cost of land and existing buildings purchased - units NYIP
1980
1997
533
Proceeds from land and buildings disposed of - units NYIP
1980
1997
534
Cost of new and second-hand vehicles purchased - units NYIP
1981
1997
535
Proceeds from vehicles disposed of - units NYIP
1981
1997
537
Cost of new and second-hand equipment purchased - units
NYIP
1980
1997
460
461
462
463
464
465
466
467
501
502
503
504
505
511
513
514
515
516
517
518
519
521
522
523
524
528
529
114
Related to dwellings : Total E/Y (nb. used for UNKNOWN prior
to 1980)
Related to dwellings : Materials, stores and fuels B/Y (nb.
used for UNKNOWN prior to 1980)
Related to dwellings : Materials ,stores and fuels E/Y (nb.
used for UNKNOWN prior to 1980)
Related to dwellings : Other stock B/Y (nb. used for
UNKNOWN prior to 1980)
Related to dwellings : Other stock E/Y (nb. used for
UNKNOWN prior to 1980)
Number of dwellings under construction B/Y (nb. used for
UNKNOWN prior to 1980)
Number of dwellings under construction E/Y (nb. used for
UNKNOWN prior to 1980)
Number of completed dwellings on hand for sale B/Y
Number of completed dwellings on hand for sale E/Y
Cost of new building work (calc'd by ratio from 93)
1996 (44)
1997 (44)
1996 (45)
1997 (45)
1996 (46)
1997 (46)
1996 (47)
1997 (47)
1996 (48)
1997 (48)
1996 (841)
1997 (841)
1996 (842)
1997 (842)
1996 (843)
1997 (843)
1996 (844)
1997 (844)
1996 (845)
1997 (845)
1996 (846)
1997 (846)
1996 (847)
1997 (847)
1996 (600)
1997 (600)
1996 (601)
1997 (601)
1996 (699)
1997 (699)
1996 (69)
1997 (69)
1996 (70)
1997 (70)
1996 (71)
1997 (71)
1996 (72)
1997 (72)
1996 (73)
1997 (73)
1996 (74)
1997 (74)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
538
Proceeds from equipment disposed of - units NYIP
1980
1997
539
541
543
544
545
1980
1986
1980
1980
1980
1991
1988
1980
1980
1980
1980
1980
1981
1981
1981
1991
1991
1991
1981
1991
558
Total net capital expenditure for calendar year - units NYIP
Computer equipment purchased (NYIP)
unknown
unknown
Cost of other new and secon-hand vehicles purchased - units
NYIP (Code not shown on form)
Proceeds from other vehicles disposed of - units NYIP (Code
not shown on from)
Cost of nuclear fuel purchased - units NYIP (drpd 92)
Proceeds from nuclear fuel disposed of - units NYIP (drpd 92)
Mains and services - acquisitions (laid or re-laid during the
year) - units NYIP (drpd 92)
Mains and services - disposals and replacements - units NYIP
(drpd 92)
Total capital expenditure acquisitions - units NYIP
1988
1997
559
Total Capex Disposals - NYIP
1988
1997
561
1988
1991
564
567
571
Amounts included in new building work for assets leased (drpd
92)
Amounts included in 504 for assets leased (drpd 92)
Amounts included in 517 for assets leased (drpd 92)
New building work (equiv 901)**
1988
1988
1993
1991
1991
1997
572
Acquisitions of land & existing buildings (equiv 902)**
1993
1997
573
Disposals of land and buildings **
1993
1997
574
Acquisitions of vehicles **
1993
1997
575
Disposals of vehicles **
1993
1997
577
Acquisitions of plant and machinery (equiv 897 from 92)**
1993
1997
578
Disposals of plant & machinery (equiv 898 from 92)**
1993
1997
581
582
583
584
591
594
597
611
619
Acquisitions of nuclear fuel (drpd 92)**
Disposals of nuclear fuel (drpd 92)**
Acquisitions of mains & services (drpd 92) **
Disposals of mains & services (drpd 92)**
Amounts included in 531 for assets leased (drpd 92)
Amounts included in 534 for assets leased (drpd 92)
Amounts included in 537 for assets leased (drpd 92)
Advertising and publicity (drpd 93)
Cost of inward transport, by road, of materials and fuel
purchased
Cost of inward transport, by other means, of materials and fuel
purchased
Cost of outward transport, by road of goods sold
Cost of outward transport, by other means of goods sold
Amounts payable for repairs, maintenance & work given out
1988
1988
1988
1984
1984
1991
1991
1991
1992
1989
1984
1989
1984
1984
1980
1989
1989
1997
1993
1995
1980
1997
546
547
548
549
550
620
621
622
623
624
625
Amounts included for work given
(augments 623 or 273)
Commercial insurance premiums paid
out
(subcontracted)
1996 (75)
1997 (75)
1996 (76)
1997 (76)
1996 (77)
1997 (77)
1996 (848)
1997 (848)
1996 (849)
1997 (849)
1996 (850)
1997 (850)
1996 (851)
1997 (851)
1996 (852)
1997 (852)
1996 (853)
1997 (853)
1996 (854)
1997 (854)
1996 (404)
1997 (404)
1996 (406)
1997 (406)
115
ANNUAL RESPONDENTS DATABASE: USER GUIDE
626
627
628
630
631
632
635
640
641
642
643
644
645
646
647
655
Bank charges
Inward transport of materials and fuel (drpd 92)
Outward transport of goods sold (drpd 92)
Other services received (equiv 190 from 93)
Other services received (Mining & Quarrying) (drpd 92)
Cost of outward transport by own staff (drpd 92)
Cost of computing services
Meat inspection charges
1984 ONLY + 1989 ONLY
1984 ONLY + 1989 ONLY (646 in 1989)
Other services received (drpd 93)
1984 ONLY + 1989 ONLY (647 in 1989)
Postal, telecommunications and transport costs (drpd 93)
Postal costs
Telecommunications costs
Amounts payable for hire of plant and machinery
1980
1980
1980
1980
1980
1980
1988
1984
1984
1984
1984
1984
1985
1989
1989
1980
1995
1991
1991
1995
1991
1991
1988
1988
1989
1984
1992
1989
1992
1989
1989
1997
656
657
702
731
Rent of industrial and commercial buildings
Cost of leasing computer equipment
Cost of materials and fuel used (drpd 92)
Cost of materials and fuel (including textiles for finishing and
sale)
Cost of textiles for finishing and sale
Cost
of
goods
purchased
for
resale
without
processing(Merchanted goods)
Cost of materials and fuel purchased (equiv 145 from 93)
1980
1986
1980
1987
1995
1988
1992
1990
1987
1980
1990
1997
1980
1997
Purchases of energy and water products (1996/7 only)
1996
1997
Purchases of materials (1996/7 only)
1996
1997
Purchases of road transport services (1996/7 only)
1996
1997
Purchases of telecommunications services (1996/7 only)
1996
1997
Purchases of computer and related services (1996/7 only)
1996
1997
Purchases of advertising and marketing services (1996/7)
1996
1997
Other services purchased (1996/7 only)
1996
1997
Purchases of electricity for resale and distribution - exc own
consumption(1996/7 only)
Purchases of gas for resale and distribution - exc own
consumption (1996/7 only)
Other services purchased - short form type (1996/7 only)
1996
1997
1996
1997
1996
1997
Payments to sub contractors - construction industry (1996/7
only)
Other services purchased - construction industry (1996/7 only)
1996
1997
1996
1997
Purchases of materials (inc stationery and goods for resale) construction (1997 only)
Purchases of water from other undertakings - exc own
consumption (1997 only)
Total purchases of goods and services (1996/7 only)
1997
1997
1996 (403)
1997 (403)
1996 (858)
1997 (858)
1996 (401)
1997 (401)
1996 (402)
1997 (402)
1996 (407)
1997 (407)
1996 (408)
1997 (408)
1996 (409)
1997 (409)
1996 (410)
1997 (410)
1996 (411)
1997 (411)
1996 (417)
1997 (417)
1996 (418)
1997 (418)
1996 (420)
1997 (420)
1996 (421)
1997 (421)
1996 (422)
1997 (422)
1997 (423)
1997
1997
1997 (424)
1996
1997
Purchases of materials (inc stationery & merchanting)
1997
1997
1996 (499)
1997 (499)
1997 (823)
732
733
734
116
1996 (405)
1997 (405)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
payments to subcontractors (1997 only)
Purchases of electricity and other energy products (1996/7
only)
Purchases of gas and other energy products (1996/7 only)
1996
1997
1996
1997
1980
1980
1990
1983
1980
1980
1986
1986
1980
1980
1986
1986
1980
1980
1980
1980
1980
1980
1980
1980
1980
1980
1980
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
808
809
810
811
812
813
814
830
Cost of crude petroleum etc. purchased for blending (drpd91)
Total quantity of crude petroleum for blending on commission
(tonnes)
Coal, coke (including breeze) and manufactured fuel - cost of
Coal , coke (including breeze) and manufactured fuel quantity (tonnes)
DERV fuel and motor spirit for use in road vehicles - cost of
DERV fuel and motor spirit for use in road vehicles - quantity
(litres)
Fuel Oil - Cost of
Fuel Oil - quantity (litres)
Gas Oil - Cost of
Gas Oil - quantity (litres)
Liquefied petroleum gases - cost of
Liquefied petroleum gases - quantity (tonnes)
Gas - cost of
Gas - quantity (therms)
Electricity - cost of
Electricity - quantity (thous) Kwh)
All other fuels
Capital expenditure - goods for letting out on hire or leasing *
Capital expenditure - assets for leasing out on a finance lease
*
Overseas royalties - total receipts from overseas exceeded
£10,000 *
Overseas royalties - toal payments to overseas exceeded
£10,000 *
Overseas services - total receipts from overseas exceeded
£10,000 *
Overseas services - total payments to overseas exceeded
£10,000 *
Pollution abatement expenditure incurred
Pollution abatement - capital expenditure
Pollution abatement - current costs
Pollution control capex post-production
Pollution control capex - production process
Pollution control costs - direct
Pollution control costs - payments to others
Excise duty payable
1993
1990
1990
1991
1991
1991
1991
1980
1995
1990
1990
1995
1995
1995
1995
1997
831
Customs and Excise drawback
1980
1997
832
833
835
Special levies, allowances etc. payable - amounts repaid
Special levies, allowances etc. receivable - amounts received
Net capex on land and existing buildings
1980
1980
1980
1995
1995
1997
836
Other net capex
1980
1997
837
1980
1997
838
Gross value added at factor cost (% is code 921) (derivn ch
93)
All non-industrial services received
1980
1997
839
Cost of purchases (drpd from derivn 93)
1980
1997
735
736
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
801
802
803
804
805
806
1996 (872)
1997 (872)
1996 (873)
1997 (873)
1996 (415)
1997 (415)
1996 (416)
1997 (416)
1996 (900)
1997 (900)
1996 (901)
1997 (901)
1996 (857)
1997 (857)
1996 (903)
1997 (903)
1996 (904)
117
ANNUAL RESPONDENTS DATABASE: USER GUIDE
840
Total sales and work done (% is code 931)(derivn ch 93)
1980
1997
1984
1985
841
Sales, work done and industrial services (excludes codes 310
and 311) (1996/7 only)
Gross output (% is code 932)(derivn ch 93)
1980
1997
842
Net output (% is code 922)(derivn ch 93)
1980
1997
843
844
No text found
Net capex (% is code 923)
1980
1980
1982
1997
845
Total employment (% is code 924)
1980
1997
846
Ratio of gross output to stocks **
1993
1997
847
Net capex as a % of GVA **
1993
1997
848
Construction Table 4. Total employees ****
1993
1997
850
1993
1995
851
Purchases and net taxes(734+830+832-831-833)(derivn ch
92)**
Net excise duty(830-831)**
1993
1997
852
853
854
855
856
Net taxes(830+832-831-833)**
Net levies(823-833)**
Total tax payable(830-832)**
Total tax rebate & allowances(831-833)**
Net land & buildings(571+572-573)**
1993
1993
1993
1993
1993
1995
1995
1995
1995
1997
857
Net plant (577-578)**
1993
1997
858
Net vehicles(574-575) **
1993
1997
859
860
861
862
863
864
865
Net nuclear fuel (drpd 92)**
Net mains & services (drpd 92) **
Backup Table 2. Includes net NYIP capex (not used) **
Increase during year - WIP - new vessels (410-409) **
Increase during year - other WIP (412 - 411) (drpd 92) **
Sales of goods produced (261 + 281) **
Work done & maintenance (262+286+282+283+284)
1993=282+283+284)**
No text found but exists on Current_Question1993
New Building Work (501 +531) (as 571) **
Aquisition of Land and Existing Buildings (502+532)**
Disposals of Land and Buildings (503-533)**
Aquisitions of Plant and Machinery(517+537)**
Disposals of Plant and machinery(518+538)**
Aquisitions of Vehicles(504+534)**
Disposals of Vehicles(505+535)**
Aquisitions of nuclear fuels(522+547) (drpd 92)**
Disposals of nuclear fuels(522+548) (drpd 92)**
Disposals of mains & services(524+550) (drpd 92)**
Aquisitions of mains & services(523+549) (drpd 92)**
Increase during year- WIP (196-197)(drpd 92)**
Ratio of gross output to stocks(841/177)**
Ratio of operatives to ATC's(205/202)**
Net capex as % of GVA((844*100)/837)**
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
118
(
1997 (904)
1996 (399)
1997 (399)
1984 (862)
1985 (862)
1996 (905)
1997 (905)
1996 (906)
1997 (906)
1996 (866)
1997 (866)
1996 (859)
1997 (859)
1996 (908)
1997 (908)
1996 (909)
1997 (909)
1996 (856)
1997 (856)
1996 (911)
1997 (911)
1996 (916)
1997 (916)
1996 (917)
1997 (917)
1996 (918)
1997 (918)
ANNUAL RESPONDENTS DATABASE: USER GUIDE
882
883
884
885
886
887
888
889
890
891
895
Total enterprises **
Total establishments **
Total tax rebate and allowance**
Net Land & Buildings(501+531+502+532-(503+555))**
Net Plant((517+537)-(518+538))**
Net Vehicles((504+534)-(505+535))**
Net nuclear fuel((521+547)-(522+548))**
Net mains & services((523+549)-(524+550))**
Net NYIP capex**
Sales of goods produced, work done and services rendered **
(dummy code for1996/7)
Purchases including net levies(derivn ch 92) ** (dummy code
for 1996/7)
Cost of industrial services received (drpd 93 equiv 623)**
Cost of non-industrial services (derivn ch 93)** (dummy code
for 1996/7)
Rent and hire of plant ** (dummy code for 1996/7)
896
897
898
899
900
901
902
903
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
Index of specialisation **
Plant and machinery - acquisitions (derivn ch 92 equiv 577)**
Plant and machinery - disposals(derivn ch 92 equiv 578) **
Number of returned forms (% is code 925) **
Number of returned employment (% is code 926) **
New Building Work (derivn ch 92 equiv 571) **
Disposals land & buildings (derivn ch 92 equiv 572) **
Finance leasing (footnote)(drpd 92) **
OPS/ATC proportion CV% **
Gross output per head CV% **
Capex per head CV% **
Stocks per head CV% **
Wages per head CV% **
Salaries per head CV% **
GVA per head CV%**
Group 130 - Sales of petroleum and other goods**
Group 130 - Exploration and other work **
Group 130 - Increase during year - stocks **
Group 130 - Purchases of materials etc.**
Group 130 - Wages and salaries less Nat. Ins**
Group 130 - Vehicles, ships etc. - Acquired **
Group 130 - Vehicles, ships etc. - Disposals **
Group 130 - Total gross capital formation **
% Total wages (140) **
% Gross value added (837) **
% Net output (842) **
% Net capex (844) **
% Total employment (845) **
% Forms returned (899) **
% Returned employment (900) **
% Employment - operatives (205) **
% Employment - ATCs (202) **
% Wages - operatives (304) **
% Wages - ATCs (301) **
% Total sales and work done (840) **
% Gross output (841) **
% Total Stocks and WIP - increase (152) **
% Total stocks and WIP - end of year (177) **
Net capex on land and existing buildings (local units)
892
893
894
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1997
1997
1993
1993
1993
1993
1993
1993
1993
1997
1993
1997
1993
1997
1993
1997
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1982
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1993
1995
1995
1995
1995
1995
1995
1995
1995
1995
1995
1993
1996 (919)
1997 (919)
1996 (920)
1997 (920)
1996 (921)
1997 (921)
1996 (922)
1997 (922)
119
ANNUAL RESPONDENTS DATABASE: USER GUIDE
936
944
945
950
951
952
953
954
955
956
961
962
963
964
965
966
967
971
972
973
974
975
976
981
982
983
984
985
986
987
990
991
992
993
994
995
996
997
998
999
Other net capex (local units)
Net capex (local units)
Total employment (local units)
Levies payable - net
Excise payments - net
Increase during year - materials, stores and fuel
Increase during year - work in progress
Increase during year - goods on hand for sale
Increase during year - work inprogress - new vessels
Increase during year - other work in progress
estab ratios
estab ratios
estab ratios
estab ratios
estab ratios
estab ratios
estab ratios
industry ratios
industry ratios
industry ratios
industry ratios
industry ratios
industry ratios
Industry ratios
Industry ratios
Industry ratios
Industry ratios
Industry ratios
Industry ratios
Industry ratios
Utility flags
Utility flags
Utility flags
Utility flags
Utility flags
Utility flags
Utility flags
Utility flags
Utility flags
Utility flags
Equivalent list for 1998, 1999 and 200 in production
120
1982
1992
1982
1982
1982
1982
1982
1982
1982
1982
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1993
1993
1993
1985
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
1992
ANNUAL RESPONDENTS DATABASE: USER GUIDE
References
Barnes, M., and Haskel, J., (2000a), “Did Lack of Downsizing Slow Down 1990s
UK Manufacturing Productivity Growth?”, Queen Mary, University of London
Draft Paper. available at <www.qmw.ac.uk/~ugte153>.
Barnes, M., and Haskel, J., (2000b), “Productivity Growth in the 1990s: Evidence
from British Plants”, Queen Mary, University of London Draft Paper, available at
<www.qmw.ac.uk/~ugte193>.
Barnes, M., and Haskel, J., (2000c), “Job Creation, Job Destruction and Small
Firms: Evidence for the UK”, Queen Mary, University of London Draft Paper,
available at <www.qmw.ac.uk/~ugte193>.
Barnes, M., and Haskel, J., (2001), “Productivity, Competition and Downsizing”,
in Summary Volume of Treasury Growth Seminar, presented at No.11 Downing
Street, October 2000, HM Treasury, London, March 2001, available at
<www.qmw.ac.uk/~ugte193> and <www.hm-treasury.gov.uk>.
Barnes, M., Haskel, J., and Maliranta, M., (2001), “The Sources of Productivity
Growth: Micro-level Evidence for the OECD”, Queen Mary, University of London
Draft Paper.
Barnes, M., Haskel, J., and Ross, A., (2001), “Understanding productivity: new
insights from the ONS business data bank”, Paper presented to Office for
National Statistics Special Session on Evidence-Based Policy, Royal Economic
Society Annual Conference, Durham.
Barnes, M. and Martin, R., (2002), “Business Data Linking: An Introduction.”
Economic Trends, 581, April; 34-41.
Bartelsman, E., and Doms, M., (2000), “Understanding Productivity: Lessons
from Longitudinal Microdata”, Journal of Economic Literature, 38, 3, 569-594.
Disney, R., Haskel, J., and Heden, Y. (1999), “Exit, entry and establishment
survival in UK manufacturing, Centre for Research on Globalisation and Labour
Markets”, Research Paper 99/9, University of Nottingham.
Disney, R., Haskel, J., and Heden, Y. (2000), “Restructuring and Productivity
Growth in UK Manufacturing”, Queen Mary, University of London Research
Paper, available at <www.qmw.ac.uk/~ugte153>.
Duranton, G., and Overman, H.G, (2001), “Localisation in UK Manufacturing
Industries: Assessing Non-randomness Using Micro-Geographic Data”, London
School of Economics Draft Paper (work in progress, March 2001)
Foster, L., Haltiwanger, J., and Krizan, C., (1998), “Aggregate productivity
growth: Lessons from Microeconomic Evidence”, NBER Working Paper 6803.
Griffith, R., (1999a), “Using the ARD establishment Level Data to look at Foreign
ownership and Productivity in the UK”, Economic Journal, 109, 456, F416-F442.
Griffith, R., (1999b), “Productivity and foreign ownership in the UK car industry”,
Institute for Fiscal Studies Working Paper 99/11.
121
Griffith, R., and Simpson, S., (2001) “Foreign firms and UK manufacturing”,
Presentation to HM Treasury, January 2001.
GSS, (1995), Report of the Task Force on Disclosure, Government Statistical
Service Methodology Series no. 4, London, December
Harris, R. I. D., (2000), “Profitability, Efficiency and Market Share in UK
Manufacturing Plants 1978-1995: Estimates for a Cross-Section of Six
Industries”, mimeo, University of Portsmouth.
Harris, R. I. D., and Drinkwater, S., (2000), “UK Plant and Machinery Capital
Stocks and Plant Closures”, Oxford Bulletin of Economics and Statistics, 62, 2,
243-265.
Haskel, J., (2000), “What raises productivity? The microeconomics of UK
productivity growth”, Queen Mary, University of London Research Paper draft
paper, available at <www.qmw.ac.uk/~ugte153>.
Haskel, J., and Heden, Y., (1999), “Computers and the demand for labour:
Industry and establishment-level evidence for the UK”, Economic Journal, 109,
454, C68-C79.
Heden, Y, (1999), “Notes on the ARD”, Draft note
Hildreth, A., and Pudney, S., (1998), “Understanding the labour market: Matching
workers and employers using respondent-level data”, Labour Market Trends,
106, December, 643-656.
Hildreth, A., and Pudney, S., (1999), “Econometric Issues in the Analysis of
Linked Cross-Section Employer-Worker Surveys”, in Haltiwanger, J. Lane, J. I.,
Spletzer, J. R., Theeuwes, J. J. M., and Troske, K. R., (eds.), The Creation and
Analysis of Employer-Employee Matched Data, North Holland, Netherlands, 461488.
HM Treasury, (2000a), Productivity in the UK: The Evidence and the
Government’s Approach, London, November 2000, available at <www.hmtreasury.gov.uk/pbr2000/index.html>.
HM Treasury, (2000b), Pre-Budget Report: Building long-term prosperity for all,
London,
November
2000,
available
at
<www.hmtreasury.gov.uk/pbr2000/index.html>.
HM Treasury, (2001a), Productivity in the UK: Progress towards a productive
economy,
London,
March
2001,
available
at
<www.hmtreasury.gov.uk/budget2001/index.html>.
HM Treasury, (2001b), Budget 2001: Investing for the Long Term: Building
Opportunity and Prosperity for All, London, March 2001, available at <www.hmtreasury.gov.uk/budget2001/index.html>.
Martin, R. (2002), “ARD Basic: notes on cleaning and linking the ARD”, draft
note.
ONS, (1999), Production and Construction Inquiries – Summary Volume 1997,
Office for National Statistics, London.
122
ANNUAL RESPONDENTS DATABASE: USER GUIDE
Oulton, N., (1997), “The ABI Respondents Database: A new resource for
industrial economics research”, Economic Trends, 528, November, 46-57.
Oulton, N., (1998), “Investment, capital and foreign ownership in UK
manufacturing”, NIESR Discussion Paper 141
Oulton, N., (2000), “A tale of two cycles: closure downsizing and productivity
growth in manufacturing, 1973-89”, National Institute Economic Review, 173,
July 2000.
123
Annual Business Survey
Background Information
ABS Publications & Special Analysis Section
Room 2.301
Office for National Statistics
Cardiff Road
Newport
NP10 8XG
By telephone on: +44 (0) 1633 456592
Email: ABS publication and analysis
CONTACT INFORMATION
If you require further help or information about the
Annual Business Survey please contact us
By email at: ABS publication and analysis
By telephone on:
+44 (0) 1633 456592 Julian Dowsell
By post at:
ABS Publications & Special Analysis Section
Room 2.301
Office for National Statistics
Cardiff Road
Newport
NP10 8XG
Statistician responsible
The statistician responsible for the Annual Business
Survey is David Knight who can be contacted by
email at the above address or by telephone on +44
(0)1633 456605.
Office for National Statistics
i
Contents
Page
Coverage..................................................................................... 1
Statistical and reporting units ...................................................... 1
Forms .......................................................................................... 1
Response .................................................................................... 1
Rounding of figures ..................................................................... 2
Methods Used ............................................................................. 2
The register ................................................................................. 2
Sample Design ............................................................................ 2
Estimation.................................................................................... 3
Regional Estimation .................................................................... 3
Regional Capital Expenditure ...................................................... 4
Sampling and other errors ........................................................... 4
Expansion of information on the forms ........................................ 5
Symbols and abbreviations ......................................................... 5
Variable Information .................................................................... 5
Regional Information ................................................................... 8
Quality Measures ...................................................................... 10
ABS Analyses............................................................................ 11
Related Articles ......................................................................... 11
General Outline
Coverage
The Annual Business Survey (ABS), formerly the Annual Business Inquiry (ABI),
estimates cover all UK businesses registered for Value Added Tax (VAT) and/or Pay
As you Earn (PAYE), classified to the UK Standard Industrial Classification 2007
headings listed in the tables. The ABS obtains details on these businesses from the
ONS Inter Departmental Business Register (IDBR), (for more details see the heading
on Register below).
Statistical and reporting units
The survey was addressed to the individual legal unit, i.e. the individual company,
partnership, sole proprietorship, etc. In a few cases, combined returns were accepted
covering two or more legal units which were not able to provide separate returns.
Forms
As with all its statistical inquiries, the ONS is concerned to minimise the form-filling
burden of the ABS. The forms are designed to ensure the burden on the individual
contributor has been kept as low as possible.
A number of different form-types are used in the survey:




Long form-types were sent to all businesses with an employment of 250 or
more and also to a proportion of selected businesses with lower employment.
Short form-types were sent to the remaining selected businesses.
From 2002 Northern Ireland data is collected by Department of Enterprise,
Trade and Investment in Northern Ireland (DETI) using their own forms
The forms differ in that long form-types ask for a detailed breakdown of purchases;
employment costs; taxes, duties and levies, whereas short form-types just ask for the
totals of these variables.
Response
Details of the response rates for all years can be found on the quality measures
spreadsheet. The Statistics of Trade Act 1947 prohibits the disclosure of any
information relating to an individual undertaking without the consent of the person
carrying on that undertaking. Rigorous checks have been made to ensure that
information relating to individual businesses has not been disclosed, either directly or
by deduction, in any of the figures released. Similar precautions will be taken in
safeguarding the confidentiality of traders' returns in the preparation of any special
analyses and in all other statistics which are made available at any time.
Rounding of figures
Where necessary, figures have been rounded to the nearest final digit shown.
Consequently, there may be slight discrepancies within a table between totals and
the sums of constituent items.
Methods Used
The development of the ABI is described in an article published in the November
2000 edition of Economic Trends, entitled The Development of the annual business
inquiry. This article must be read in conjunction with the methodology information
given on this page.
The register
The register used for the ABS is the IDBR, which consists of companies,
partnerships, sole proprietorships, public authorities, central government
departments, local authorities and non-profit making bodies. The main administrative
sources of the IDBR are HM Revenue and Customs (HMRC) for VAT and PAYE
details.
The following can be passed to the ONS in respect of each VAT registration and for
members of group registrations for use in creating a statistical register or conducting
a statistical survey: - VAT number, name, address, trading style, legal status,
standard industrial classification and turnover.
Each legal unit is classified to a single activity, whether it is wholly or mainly engaged
in that activity. The nature of a business can change with time, possibly because
another business has been absorbed in a take-over. Some businesses have a very
significant secondary activity, perhaps completely different from their main activity,
and a small change of direction can lead to a new main activity and to the
reclassification of the business. Other changes may arise from improvements to
classification data held on the IDBR as a result of new information received about
individual businesses. The reclassification of businesses for reasons such as these
contributes to the year on year changes in the figures published.
Sample Design
The sample was designed as a stratified random sample of about 66,971 businesses
(2008 survey) from the register of legal units. The survey population is stratified by
SIC(2007), employment, and country using the information from the register. The
sampling scheme is designed to give best estimates of the population totals for a
given sample size and involves selecting all the largest businesses with a
progressively reducing fraction of smaller businesses. This method ensures the
sample size is kept to a minimum.
The survey results are grossed up to the register population, so that they relate to all
active UK businesses on the IDBR for the sectors covered. For 2002 data published
on 29th June 2004 and 2003 data published on 16 December 2004 Northern Ireland
was collected and processed by DETINI. However 2002 and 2003 revised data
published on 23 June 2005 was collected by DETINI but processed by ONS.
Estimation
Factors were produced to enable estimates for all businesses classified to each
SIC(2007) to be compiled from data provided by responding businesses. These
factors were calculated for each employment size-band within each SIC(2007) and
are equivalent to the ratio of responding businesses to the total number of
businesses. Northern Ireland and Scotland are sampled and estimated for
separately, England and Wales are sampled separately but are combined for the
estimation procedure.
Returns for the few large non-responders were estimated for individually. This
estimation was normally based on summary data received from the business, or on
the business's return to the survey in the previous year adjusted to take account of
the likely change in the value of trading over the period.
Regional Estimation
Important Information about Regional Data - The total regional estimate for a variable
summed across all regions within any industry will not necessarily equal the UK
industry estimate of the variable. This is due to the method of calculating estimated
regional data. The regional estimates are constrained to the UK data at the allindustry level.
The ABS regional methodology apportions returned data to the local units (individual
sites) of the reporting unit (main site from which data is collected) before the
estimation process. The estimation process is then performed at local unit level. For
the UK industry level data the estimation process is performed at the reporting unit
level.
The local unit estimation process calculates an estimation factor for non-response
from the apportioned local unit data. The estimation factor is then applied to the
apportioned local unit values to give a total estimate for the local unit population. The
estimation factor is calculated using the IDBR employment, classification and region
of the local unit. This estimation factor will cause the regional data to differ from the
estimated national data and so the data is constrained to the UK all industry level.
The reporting unit estimation process calculates the estimation factor for nonresponse from the returned reporting unit data. The estimation factor is then applied
to the returned reporting unit data to give a total estimate for the reporting unit
population. The estimation factor is calculated using the IDBR employment,
registered turnover and classification of the reporting unit.
The local unit classification may differ from the classification of the reporting unit and
the apportioned value for the local unit will be added to the industry of the local unit,
not the industry of the reporting unit as would be the case for UK data.
The sub-national estimates are constrained to the UK data at the all-industry level
and so the total regional estimate, for a variable summed across all regions, and all
industries, will equal the UK estimate of the variable at the A-S level in the national
results.
Regional Capital Expenditure
Figures on regional capital expenditure are no longer made available on the ABS
website pages. Accurate estimation of regional capital expenditure is dependent on
there being a strong correlation between the variable on which any estimation is
based, in this case local employment, and the variable for which we are attempting to
produce estimates, here regional capital expenditure. It has been established that the
relationship between regional employment and regional capital expenditure is, in fact,
unreliable.
It will still be possible to obtain regional capital expenditure figures by requesting
either standard extracts, for which there is no charge, or special analyses, for which a
charge will be made. ONS does not recommend the use of these regional capital
expenditure figures but recognises that some users will, nevertheless, wish to have
them.
For more information about either of these services, please either email
[email protected] or telephone +44(0) 1633 456592 for standard extracts or +44(0)
1633 456601 for special analyses.
Further information on the regional methodology is contained in an article published
in the November 2000 edition of Economic Trends. Refinements to the methodology
have taken place since this article was written; methodogical information given on
these pages supersedes the Economic Trends article.
Estimates of Gross Value Added shown for regional data will not be exactly the same
as those published in the United Kingdom Regional Accounts. They differ in three
key areas; adjustments for coverage; adjustments needed to move the accounts onto
an ESA 95 basis; and adjustments for balancing purposes. More information can be
obtained through the variable information section.
Sampling and other errors
Estimates derived from samples invariably produce results which differ from those
that would have been obtained from a complete survey. If a number of different
samples were selected then each would produce a different result. Sampling errors
measure the extent to which these estimates can be expected to differ from the 'true'
value.
These sampling errors are small for the aggregates of the main ABS variables and
indeed the sample is specifically designed to achieve this. Quality measures are
available for total turnover, approximate gross value added, total purchases and
capital expenditure.
However, in addition to sampling errors there is the potential for non-statistical errors
which cannot be easily quantified. Examples where these errors may occur are
deficiencies in the survey register and errors made by respondents in completing the
survey forms.
Expansion of information on the forms
Short form-types ask only for total purchases; total employment costs; total taxes,
duties and levies rather than the detailed breakdown asked for on long form-types.
Proportionate breakdowns of the full purchases; employment costs; taxes, duties and
levies provided by the smallest businesses completing the long form-types were
calculated for each SIC(2007). Factors were then applied to the data against the
summary headings on the small form-types to derive the breakdown for the above
variables.
Symbols and abbreviations
The following symbols and abbreviations are used within the ABS data:
..
not available
-
nil or less than half the level of rounding
*
Information suppressed to avoid disclosure
Variable Information
Number of enterprises
An enterprise is defined as the smallest combination of legal units, which have a
certain degree of autonomy within an enterprise group. An enterprise group is simply
a number of enterprises under common ownership.
Total turnover
Turnover is defined as Total sales and work done. This is calculated by adding to the
value of Sales of goods produced, Goods purchased and resold without further
processing, Work done and industrial services rendered and Non industrial services
rendered.
Approximate gross value added at basic prices
Gross value added (GVA) represents the amount that individual businesses,
industries or sectors contribute to the economy. Broadly, this is measured by the
income generated by the business, industry or sector less their intermediate
consumption of goods and services used up in order to produce their output. GVA
consists of labour costs (e.g. wages and salaries) and an operating surplus (or loss).
The latter is a good approximation to profits. The cost of capital investment, financial
charges and dividends to shareholders are met from the operating surplus.
Data collected and published through the ONS Annual Business Survey (ABS) are
used to produce an approximate estimate of GVA at basic prices. This measure is
approximate because it does not allow fully for certain types of National Accounts
concepts/issues such as taxes or subsidies or income earned-in-kind.
The ABS forms a major data input in the production of Input-Output Annual Supply
and Use Tables used to set the annual level of UK Gross Domestic Product. The
input-output tables also show industry estimates of GVA at basic prices but are
different from those shown in the ABS. In producing the Input-Output based
estimates of GVA at basic prices fully consistent with the European System of
Accounts 1995, there are essentially four key adjustments required to the survey
based data: coverage adjustments; conceptual and valuation adjustments; quality
adjustments; and coherence adjustments. Details are available in the Annual
coherence adjustment in the national accounts article; the notes to the United
Kingdom Input-Output Analyses.
Estimates of Gross Value Added shown here for Regional data will also not be
exactly the same as those published in the Regional Accounts for the same reasons.
Further information is available on the Regional accounts product page.
The gross value added estimate for Group 41.1 (Section F) – Development of
building projects, is not considered as National Statistics quality but is included to
meet user needs. A gross value added estimate is now published for Real estate
activities, Section L.
Total purchases of goods materials and services
This represents the value of all goods and services purchased during the year.
Total employment
This is the point in time estimate of full and part time employees on the payroll plus
the number of working proprietors employed on a set day in September.
Total average employment
This is the point in time estimate of full and part time employees on a set day in
September, adjusted to give a year average value, plus the number of working
proprietors employed on the same day.
Total employment costs
This represents amounts paid during the year to employees. This includes all
overtime payments, bonuses, commissions, payments in kind, benefits in kind,
holiday pay, employer's national insurance contributions, payments into pension
funds by employers and redundancy payments less any amounts reimbursed for this
purpose from government sources. No deduction is made for income tax or
employee's national insurance contributions etc. Payment to working proprietors,
travelling expenses, lodgings allowances, etc are excluded.
Total net capital expenditure
This is calculated by adding to the value of new building work, acquisitions less
disposals of land and existing buildings, vehicles and plant and machinery.
Regional capital expenditure data for years later than 2004 will not be available on
the ABS website pages. It will be available on request. See the regional capital
expenditure section for additional information on regional estimates of capital
expenditure
Total net capital expenditure - acquisitions
This is calculated by adding to the value of new building work, acquisitions of land
and existing buildings, vehicles and plant and machinery.
Total net capital expenditure – disposals
This is calculated by adding the value of disposals of land and existing buildings,
vehicles and plant and machinery.
Total stocks and work in progress - value at end of year
This represents the value at the end of the year for materials, stores and fuel and
goods on hand for sale. Amounts for materials which have been partially processed
but which are not usually sold without further processing are also included.
Total stocks and work in progress - value at beginning of year
This represents the value at the beginning of the year for materials, stores and fuel
and goods on hand for sale. Amounts for materials which have been partially
processed but which are not usually sold without further processing are also
included.
Total stocks and work in progress - increase during year
This represents the increase during the year for materials, stores and fuel and goods
on hand for sale. Amounts for materials which have been partially processed but
which are not usually sold without further processing are also included.
Retail Outlets (Only applies to Division 47)
The total number of outlets consists of stores, mail order outlets, market stalls and
road side pitches, owned and operated by businesses classified to the retail sector in
the UK. It is a point in time estimate made at the end of the calendar year.
There are some methodological differences in the production of the retail outlets data
for 2008. Most significantly, the source data for 2008 was sampled on Standard
Industry Classification (SIC) 2007. The change of SIC has meant that some activities
are now outside the scope of the 2008 retail sample (e.g. household repair) while
others are now within scope (e.g. Retail sale of petrol).
Retail Commodities (Only applies to Division 47)
This is a breakdown of the total retail turnover within the retail sector into groupings
of like items based upon the European COICOP (Classification of Individual
Consumption by Purpose) classification.
The commodity details from ABS are not the same as those for household
expenditure in the National Accounts Blue Book and the Input-Output Analyses for
two reasons.

Household final consumption expenditure includes expenditure on
commodities sold by all types of seller, not just by businesses classified as
retailers in the ABS. However, household final consumption expenditure does
not include sales by retailers to businesses.

The COICOP detail from ABS for retailers and similar information for retail
sales by non-retail businesses has not been fully incorporated in the Blue
Book because long-run revisions to the National Accounts, necessary to
reconcile results from the 2000 ABS and onwards with earlier years' retail
data, have not been made.
For this year's National Accounts, only the revised total retail business
information from the ABS has been used to provide a benchmark in deriving
estimates of total household final consumption expenditure on goods sold by
retailers and non-retailers. This has been supplemented by growth-rate
movements of many of the commodities.
Regional Information
ONS has produced ABS estimates for English Government Office Regions, Scotland,
Wales and Northern Ireland. The methodology used to calculate these estimates is
consistent across the UK and the resulting estimates are constrained to the UK
national estimates at the all-industry level. This regional methodology produces
estimates suitable for comparisons to be made between regions of the UK.
Regional capital expenditure data is not available on the ABS website pages. It will
be available on request. See regional capital expenditure for additional information
on regional estimates of capital expenditure.
The Scottish Executive publishes additional analyses for Scotland based on ABS
estimates. These are available on the Scottish Executive Website.
Scottish Executive Data - More detailed results for Scotland can be obtained:
Susan Duncanson
Tel: 0300 244 6802
email: [email protected]
Scottish Executive
Business and Enterprise Statistics
3rd Floor
5 Atlantic Quay
150 Broomielaw
Glasgow
G2 8LU
Department of Enterprise, Trade and Investment Northern Ireland (DETI(NI))
estimates for Northern Ireland are available on DETI(NI) Website. These estimates
will differ from ONS Northern Ireland estimates due to differences in the
methodology. The DETI(NI) methodology, while broadly comparable with that of the
UK, is designed to give the best Northern Ireland estimate rather than an estimate
which is consistent with the rest of the UK. Users who want to look at trends in
results for Northern Ireland are advised to look at the Northern Ireland website.
Department of Enterprise, Trade and Investment, Northern Ireland Data - More
detailed results, special analyses and further information on DETI methodology can
be obtained:
Stephanie Harcourt
Tel: 028 9052 9403
email: [email protected]
Brian Spence
Tel: 028 9052 9424
email: [email protected]
Statistics Research
Netherleigh
Massey Avenue
Belfast
BT4 2JP
National Assembly for Wales Data - Further information can be obtained:
John Morris
Tel: 029 2080 1401
email: [email protected]
Second Floor
Economic and Labour Staistics
Cathays Park
Cardiff
CF10 3NQ
Quality Measures
The ABS is a stratified random sample, using SIC(2007), employment and country as
stratifying variables.
As in all samples the estimates from the survey are subject to various sources of
error. The total error in a survey estimate is the difference between the estimate
derived from the data collected and the true (unknown) value for the population. The
total error consists of two main elements; the sampling error and the non-sampling
error.
The ABS was designed to minimise both these errors.
Sampling error



The sampling error is the error that arises because the estimate is based on a
survey rather than a census of the population. The results obtained for any
single sample may, by chance, vary from the true values for the population
but the variation would be expected to average to zero over a number of
repeats of the survey.
The standard error is the estimated value of the sampling error. Our estimate
for a variable, plus and minus the standard error for the variable, gives a
range in which the true unknown value for the population should lie. The
closer the standard error to 0, the more reliable the estimate.
The coefficient of variation is the standard error of a variable divided by the
survey estimate, and it is used to compare the relative precision across
surveys or variables. The closer the coefficient of variation is to 0, the more
reliable the estimate.
Sampling errors for ABS surveys are available at 3 digit SIC(2007) group level for the
following variables:




Total Turnover
Approximate Gross Value Added at Basic Prices
Total Purchases of Goods and Services
Total Net Capital Expenditure
Non-sampling error
Non-sampling errors are not easy to quantify and include inadequate coverage,
measurement, processing and non-response.
The response rate gives an indication of the likely impact of non-response error on
the survey estimates.
A Summary Quality Report for the ABI 2 inquiry is available. This gives qualitative
information on the various dimensions of ABI 2 quality.
ABS Analyses
Standard Extracts
The following range of standard extracts is available on request;
An expanded range of variables for Sections A-O down to 5 Digit Subclass level of
SIC(92)/ SIC(2003) and, Sections A-S of SIC2007 where available.
Retail commodity data down to 5 digit subclass level of SIC(92)/ SIC(2003) (Division
52 only) and SIC(2007) (Division 47 only).
Regional information for 5 main variables at 3 Digit Group level for Sections A-O of
SIC(92)/ SIC(2003) and Sections A-S of SIC(2007).
Bespoke Analyses
Data not available from the standard extract option may be available as a bespoke
analysis.
It should be noted that, even though the bespoke analysis service, it will not be
possible to produce any back-runs of pre-2008 data based on the new SIC(2007)
There is a charge for these analyses and quotations will be given on request.
For more information about either of these services please contact us at ABS
publication and analysis, or telephone +44 (0)1633 456592 for Standard Extracts or
+44 (0)1633 456601 for Bespoke Analysis
Related Articles
The following articles are related to the Annual Business Inquiry financial data
Census of Production - 100 Years
Report on the Quinquennial Review of the Annual Business Inquiry pt 2 - Financial
questionnaire
Published May 2004
Labour Productivity Measures from the Annual Business Inquiry
Published 11 November 2002
Development of the Annual Business Inquiry
Article published in the November 2000 edition of Economic Trends.
Office for National Statistics 11
The Development of the Annual Business Inquiry
Gareth Jones
Office for National Statistics
E-mail: [email protected]
National Statistics customer enquiry line: +44 (0)845 601 3034
Introduction
This article describes the development of the Annual Business
Inquiry (ABI), a new integrated survey of employment and
accounting information from businesses and other establishments
in most industry sectors of the economy. The ABI replaces the
following annual survey systems:
●
●
●
Annual Employment Survey (AES)
Annual Censuses of Production and Construction
(ACOP/C), which include the Purchases Inquiry (PI)
The six annual Distribution and Services (DSI) inquiries,
viz:
Annual Wholesale Inquiry
Annual Retail Inquiry
Annual Motor Trades Inquiry
Annual Catering Inquiry
Annual Property Inquiry
Annual Service Trades Inquiry
●
In addition to the restructuring and integration of inquiries, major
improvements in methodology have been implemented in a
standardised way across the ABI. The ABI is expected to provide
more coherent and consistent annual industrial statistics covering a
range of variables for the whole economy. Various outputs from the
ABI will help to improve the quality of the national accounts. In
particular, it will contribute to the current price Input-Output annual
Supply and Use Tables and the benchmarking of household final
consumption and gross fixed capital formation, all of which improve
the measurement of Gross Domestic Product. Also, the quality of
“per head” type statistics, for example output per head will be
improved considerably.
Background
●
●
●
●
●
The ABI was first conducted in respect of the year 1998 and results
for that year started to become available in February 2000.
The project has been managed by the Office for National Statistics
(ONS) in close consultation with other government departments,
including the Department for Education and Employment (DfEE),
the Department of Trade and Industry (DTI), H.M. Treasury, the
Scottish Executive, the National Assembly for Wales and the
Northern Ireland Office. There has also been consultation with
representatives of Local Authorities. A Steering Group and several
lower level working groups were set up to ensure good user
consultation and effective project management.
In 1996, the Office for National Statistics (ONS) began work on a
project to integrate a range of its annual inquiries into a single system
known as the Annual Business Inquiry. The objectives of this project
are described below, but possibly the most important was a desire
for greater consistency and simplicity to be achieved by replacing a
wide range of inquiries by a single integrated system.
The development of the ABI took place against the background of
two significant developments in business statistics:
‘Economic Trends’ No. 564 November 2000 © Crown copyright 2000
●
●
The successful implementation of the Inter-departmental
Business Register (IDBR) during 1994 and 1995. This laid
the foundation for a more integrated “whole economy” inquiry.
The introduction by Eurostat (the Statistical Office of the
European Communities (EC)) of a new EC Regulation on
Structural Business Statistics (SBS) which extended the data
requirements of earlier EC Directives in a number of ways.
49
The systems replaced by the ABI are listed in the introduction above.
The Annual Employment Survey collected information on number
of employees (with a four way split between male/female, full/part
time) and industry classification according to the Standard Industrial
Classification, SIC (92). These data were collected separately for
each site used by businesses in the survey sample. The AES (which
began in 1995) was used primarily for the annual estimates of
employee jobs made in September each year and itself replaced
earlier censuses of employment which had been carried out less
frequently in earlier years. The data obtained from the AES also
provided important information for the updating of the business
register (IDBR), particularly with regard to the internal structure of
enterprises, the so-called “local units” or individual sites at which
enterprises operate. Such information is important in producing
regional or other geographical analyses of business data.
The Annual Censuses of Production and Construction and the Annual
Distribution and Services Inquiries were sample surveys collecting
annual accounting information from businesses, usually in more detail
than is obtainable from company accounts. The Annual Census of
Production also collected some data on employment, though on a
slightly different basis to the AES. The accounting data obtained
from these surveys is used for a variety of purposes. One of the
most important is the estimation of Gross Value Added which is
used in the National Accounts for the estimation of Gross Domestic
Product (GDP). The Purchases Inquiry was a subsample of the
ACOP sample from which was collected more detailed information
on the breakdown of purchases of goods and services by businesses.
Approximately one fifth of industries in the Production sector were
surveyed each year on a rotating five-yearly basis.
Objectives of the ABI project
The objectives of the ABI project can be categorised under four broad
headings:
By combining the collection of employment and accounting data into
the same survey the validity of productivity measures is improved
since value added and employment are estimated on a consistent
basis. The lack of comparability between employment and value
added had been a serious problem with the previous systems. The
integration of industry sectors into one survey avoids the problems
of omissions and double counting which were previously possible.
A common methodology is also applied across all industry sectors,
including standardisation of definitions. Also, the ABI has UK
coverage as required for the National Accounts and Eurostat whereas
the AES did not cover Northern Ireland.
The EC SBS Regulation set additional requirements which are largely
being met by adding questions to the 1999 ABI. The Regulation
also requires the use of standard definitions for international
comparability. A set of standard analyses is also specified and has
been adopted for the ABI. These analyses include regional and size
breakdowns as well as the more traditional industry analyses. Many
of the EC requirements as well as other analyses available from the
ABI system cover all industry sectors, including services. In particular,
regional analyses of value added in the service industries will become
available. In addition to the SBS regulation, ABI questions and
definitions have been put on a basis consistent with the requirements
of the EC regulation on the European System of National Accounts
(ESA95).
As will be explained below, a second inquiry known as the Annual
Register Inquiry (ARI) has been developed in parallel with the ABI.
The ARI serves the purposes of annually measuring various aspects
of register quality and also improving the updating of the register
and consequently its quality. It is designed in such a way as to meet
the requirements for register updating laid down in the EC Business
Registers Regulation and provides information on size (measured
by employment), industry classification and location of both complete
enterprises and the individual sites at which they operate (local units).
Structure of the ABI system
Figure 1 below shows the relationships between the previous
systems and the ABI and ARI.
AES
ACOP/C
↓
↓
ARI
●
●
●
●
50
Consistency, standardisation and improved methodology
Meeting international requirements
New analyses, especially in the Services sectors
Quality assessment and improved maintenance of the IDBR.
DSI
>
↓
ABI/1
ABI/2
↓
In particular the SBS regulation covered the Services sectors of the
economy for the first time. The regulation contains requirements for
data from 1995 onwards. However, there is also provision for a
transitional period during which Member States could request
derogations from the regulation. This transitional period covered the
years 1995 to 1998 inclusive so that the full requirements of the
regulation effectively come into force for the 1999 survey year.
NEW EC
REQUIREMENTS
Employment data are now collected along with accounting data in
ABI and the sectoral inquiries ACOP/C and DSI are merged. The
ABI form is in two parts, one dealing with employment data (ABI/1)
and the other with accounting data (ABI/2). The employment data
relate to a December reference date and both parts of the form are
despatched around the end of the year. The main reason for splitting
the form into two parts is that the employment data are available
from businesses much earlier than their accounting information. Data
collection for ABI/1 is closed down at the end of March of the following
year whereas the closedown for ABI/2 is about six months later in
order to allow businesses to provide information from their own
annual accounts (which can be up to the year ending fifth April). The
early return of the ABI/1 part of the form allows publication of
employment estimates to follow a timetable similar to that which has
existed for the AES. Another reason for using a two-part form is that
employment information and accounting information are often
provided by different parts of an organisation and the two sections
of the form can be addressed to different individuals.
The sample size of ABI/1 in 1998 was approximately 78,500
enterprises. ABI/2 is a subsample of ABI/1 in which some of the industry
sectors are not covered and has a sample size of approximately 75,000.
ABI/1 covers divisions 2–93 inclusive of SIC(92) - i.e. all divisions
except Agriculture, Private Households with employed persons and
Extra-territorial organisations. In 1998 ABI/2 covered the above
industries except for divisions 2 (Forestry), 5 (Fishing), 65–67 (Financial
Services), 75 (Public Administration) and the public sector in Education
(division 80) and in Health and Social Work (division 85). Doctors,
dentists and charitable organisations in division 85 were also omitted.
It is expected that the industry coverage of ABI/2 will be expanded in
future years (e.g. SIC divisions 2 (Forestry) and 5 (Fishing) are being
covered in the 2000 ABI/2).
The Purchases Inquiry has been retained as a sub-sample of ABI/2,
the additional questions on purchases detail being added to the
ABI/2 forms for the relevant industry sectors. A phased expansion of
the PI is taking place. Current plans are to expand the industry
coverage each year until all ABI/2 industries are covered every year
from 2001 onwards. This is a major development being undertaken
as part of a package of improvements to economic statistics agreed
with H.M. Treasury and Bank of England. The industry coverage of
the PI in the transitional years is planned to be as follows:
●
●
●
1999 - All production SICs plus 44 distribution and services
classes/subclasses
2000 - All production, and over two thirds of distribution and
services SICs
2001 & after - All production, construction, distribution and
services SICs.
As mentioned above, the introduction of the ABI has also been
accompanied by the development of another inquiry, the Annual
Register Inquiry (ARI) which began operation in July 1999. The ARI
subsumes previous register proving activity and also replaces the
AES in updating the local unit structures of enterprises and is
designed for this latter purpose rather than for the direct measurement
of employment. Large enterprises are covered every year and
medium sized enterprises (20-99 employment) every four years.
Below the employment threshold of 20 there is no systematic proving
of enterprises due to the very large numbers of such businesses.
However, part of the ARI sample is allocated to quality assessment
of various register characteristics (especially industry classification).
Another part of the sample is available for targeting, on an ad hoc
basis, areas of the register where quality problems are known to
exist. The ARI is thus a very flexible tool for updating the IDBR and
for assessing and improving its quality. Approximately 400,000 local
units in 68,000 enterprises are covered. The ARI has to date been
treated as an annual inquiry. However, in future, it is planned to
spread the workload more evenly by designing the sample so that
four quarterly despatches of forms are made.
Outline Project Timetable
The phased introduction of the ABI system is represented by the
outline timetable below. The dates shown here are the survey years
to which the data relate and not the dates when results become
available.
●
1996
AES, ACOP/C, DSI
●
1997
AES, modified ACOP/C and DSI for parallel run
●
1998
AES, ABI
●
1999
ARI, ABI (with full EC requirements)
The first changes were made for the 1997 inquiries. The ACOP/C
and DSI inquiries were modified to collect employment information
on a basis which would be comparable with the AES for 1997. In the
case of ACOP/C this meant an extension of existing employment
questions and a change in definition from a year average figure to
one at a given reference date. The same questions were introduced
into DSI which had not previously contained questions on
employment. The purpose of these changes was to meet the
requirements of customers for a parallel run of the ABI methodology
against that of the previous systems. The ABI results and analysis
software was written for first use on the 1997 data even though the
data were collected via the previous survey systems. The parallel
run was intended both to evaluate the new methodology and also to
51
estimate any discontinuities introduced in time series by making the
methodological change.
Other changes made in 1997 included a move in DSI from the use
of turnover to the use of employment for size stratification of the
sample design, putting DSI on the same basis as ACOP/C and the
filling of some minor gaps in the industry coverage of ACOP/C and
DSI. Some work on the quality of the IDBR in these industries was
undertaken to support this extension of coverage. Additionally, the
data collection software for ACOP/C was brought into line with DSI
which already used standard data collection software.
A further change made in 1997 was the introduction of a question
on retail turnover in all industry sectors. The purpose of this is to
capture retail activity outside of the retail sector, this being required
for consumers’ expenditure estimation within the national accounts.
The previous method of measuring such activity was to maintain on
the register so-called “retail carry-in” units. These were parts of nonretail enterprises which by one means or another had been identified
as having retail activity. Such units were sent a shortened version of
the DSI retail form even though the enterprises to which they
belonged would also be sent other forms appropriate to their industry
sector. This practice was continued for 1997 alongside the use of
the new retail turnover question, again for parallel run purposes.
The use of the retail carry-in units ceased in 1998 as it was felt that
the new arrangements gave a more complete picture of retail activity.
The ABI was introduced for the 1998 inquiry year and completed
the integration of the ACOP/C and DSI systems. This involved a full
review of sample design, form types, questions and definitions with
the perspective of a single system rather than a set of separate
sectoral inquiries. The AES was retained for 1998 alongside ABI to
provide a more complete parallel run for the employment questions.
ABI/1 has the same industry sector coverage as AES, whereas
ACOP/C and DSI had the narrower sectoral coverage associated
with ABI/2 and so could not provide a full parallel run.
The transition to the new ABI and ARI systems was completed for
the 1999 survey year with the launch of the ARI in July 1999 and
the addition of extra questions to the 1999 ABI to meet the full
requirements of the SBS Regulation. The expansion of the
Purchases Inquiry will not however be complete until 2001 as
described above.
The adoption of estimates of employee jobs from ABI/1 in preference
to those obtained from AES was complicated by the fact that 1998
ABI/1 estimates were significantly higher than those obtained from
the 1998 AES. The difference was of the order of 500,000 – 750,000
and an extensive research programme was needed to establish the
52
reasons behind this difference before the ABI/1 data could be adopted
as the official estimates of employee jobs. The 1998 and 1999
ABI/1 employee jobs estimates will be published in April 2001. An
article in the September 2000 edition of Labour Market Trends
described the reasons behind the discrepancy and gave more details
of the schedule for launching ABI/1 employee jobs estimates and
revised back data.
Form Types, Questions and Definitions
In the run up to the launch of the ABI at the end of 1998 a thorough
review was undertaken of all form types to be used, the questions
they contained and the definitions of those questions. This was
undertaken in consultation with national accounts customers to
ensure consistency with ESA95 requirements.
An attempt was made to minimise the number of form types while
also maintaining form content which would appear relevant to the
businesses which received the forms. The form types are customised
for industry sectors and sub-sectors and in 1998 there were 3 basic
form types for ABI/1 and 21 for ABI/2. In addition, for each basic
form type there is usually a short form equivalent which collects
information on the main totals but not the more detailed breakdowns
which appear on the basic form types. Short forms are used to reduce
the burden of form filling on businesses. In ABI/2 a proportion of
businesses in the sample receive the short form, this proportion
increasing as the size of businesses becomes smaller. The
businesses receiving the short forms are subsampled randomly from
the total sample in each stratum. In ABI/1 a different approach is
taken, the short forms being used for businesses which have also
been sampled for the ARI or the fourth quarter short-term employment
survey and also for all Northern Ireland businesses. (Special
arrangements, described later, have been agreed with the Northern
Ireland Office for the production of Northern Ireland employment
estimates from ABI). Imputation methods are used to expand the
information on short forms to the detailed breakdowns required for
analysis.
The main questions asked in ABI/1 are the number of employees
with a four-way split between male/female, full/part time (as in the
AES) and also the number of working proprietors/partners and the
number of other unpaid workers (e.g. family workers) as well as
total employment which includes all of the above. Also the definitions
used are consistent with the AES definition of employees. The
questions relate to a reference date in mid-December and for
comparison with AES (which uses a September reference date) and
for estimation of the year average, data from the monthly and
quarterly employment and turnover inquiries are used to convert
from one basis to another.
The ABI/2 forms in general contain many more questions than
ABI/1 and the range of questions is more variable across industry
sectors. However there is a core set of questions covering turnover
(i.e. sales of goods and services), employment costs, purchases of
goods and services, taxes and subsidies, inventories and capital
investment which occur on nearly all form types. The basic form
types also tend to contain more detailed breakdowns of these
aggregates. Definitions of these quantities have been reviewed and
standardised. In general the definitions used are now consistent with
the requirements of the SBS Regulation which in turn are usually
consistent with the European System of Accounts (ESA).
Additionally, ABI/2 contains a number of “filter questions” which are
used to identify businesses with particular types of activity so as to
improve the sampling frame for other more detailed inquiries. Filter
questions for Research and Development activity and International
Trade in Services have been included and questions on ECommerce are planned for the ABI 2000 forms.
A number of important variables can be derived from the data
collected, notably gross value added and total output, both at basic
prices, operating surplus, labour productivity and unit wage costs.
Both the collected and derived variables can be analysed by a
number of other variables mostly taken from the IDBR. In addition to
analyses by SIC industry and size, businesses can be classified
and analysed by country of ownership and by legal status (i.e.
whether they are companies, partnerships, single proprietor
businesses, public corporations, non-profit making bodies, central
or local government). Geographical analyses are also possible
according to the EC NUTS hierarchical classification system. NUTS
has five levels in the hierarchy which approximately follow the
organisation of local government:
NUTS1
NUTS2
NUTS3
NUTS4
NUTS5
Region
Group of Counties or Unitary Authorities
County or Unitary Authority
District or Unitary Authority
Ward
Analyses Available
The table below summarises the range of analyses potentially
available from ABI, subject to considerations of accuracy and
confidentiality.
Collected
Derived
Analysed by
Employment
Gross Value Added
SIC Industry
(GVA)
Turnover
Total Output
Geography
Employment Costs
Operating Surplus
Size (Employment
or Turnover)
Purchases
GVA/Turnover
Legal Status
Taxes/Subsidies
Labour Productivity
Country of
Ownership
Inventories
Unit Wage Costs
Time (i.e.time
series)
Investment
Inventory Ratios
Investment Ratios
For most of the main variables collected, breakdowns are also
available. For example, employment costs are subdivided into wages
and salaries, pension contributions, social security contributions and
redundancy payments. The breakdowns available for turnover vary
by industry sector and in the retail sector there is a detailed
breakdown by product. In addition, the Purchases Inquiry provides
considerable purchases detail for the industries it covers.
In addition to the above it is also of course possible to compare
different years and build up time series as required.
Dissemination of ABI data
The first preliminary results of the 1998 ABI/2 were published in a
News Release in February 2000. This publication timetable will be
accelerated in future years. For the 1998 ABI/2 other publications
follow the same form as in earlier years (i.e. sector reviews for the
distribution and services sectors and a summary volume for the
production and construction sectors). More detailed results will be
available on the Internet (via ONS’ StatBase®). In addition it is also
possible to obtain customised analyses on payment terms via ONS’
Data Analysis Service.
For the 1999 inquiry a review of publications is taking place. In
addition to new printed publications, the CD-ROM product PacStat
may be extended to cover all ABI/2 industry sectors. In the slightly
longer term the Annual Respondent Database (ARD) may also be
extended to all ABI/2 sectors. The ARD provides longitudinal data
for individual businesses and is available for use by external
researchers subject to a confidentiality undertaking.
ABI/2 results will be reflected in the national accounts according to
the following timetable. The 2000 Blue Book and Input-Output Annual
Supply and Use Tables published in August 2000 incorporated the
growth rates from the 1998 ABI/2 data and the 1997 ABI/2 parallel
run data as input to the balancing and setting of annual current price
GDP for 1998. The 2001 Blue Book and Input-Output Annual Supply
53
and Use Tables to be published in September/October 2001 will
incorporate the correct levels from the 1997 ABI/2 parallel run and
the 1998 and 1999 ABI/2 data. The pre-1997 data will be linked onto
1997 using appropriate link factors and included as part of a larger
package of revisions to the national accounts.
Plans for the publication of ABI/1 data have been set out in an article
in the September 2000 edition of Labour Market Trends. ABI/1 results
together with derived statistics on productivity and the claimant count
unemployment rate will be made available in April 2001 on the day
of the Labour Market Statistics First Release. Both 1998 and 1999
ABI/1 results will be published along with revised back data for earlier
years. Subsequently ABI/1 results will appear within the quarterly
workforce jobs series in the Labour Market Statistics First Release,
and, in more detail, in Labour Market Trends. ABI/1 data (and derived
results using ABI/1 and ABI/2) for 1998 will not be published before
April 2001.
Future Developments
In addition to the continued development of the Purchases Inquiry
section of ABI/2, three further areas of work are planned:
●
Further refinements in form types and questionnaire design
with a view to reducing the compliance burden, especially
for small businesses.
54
●
Investigation of whether the accuracy of employment
estimates could be further improved by using Pay-as-youearn (PAYE) data as auxiliary information in estimation.
●
Research into a better method of outlier treatment (see
methodological annex), especially with a view to using a twosided rather than one-sided method.
Contacts
ABI/1
James Partington
ONS, Room 249, East Lane House, East Lane,
Runcorn, Cheshire. WA7 2DN.
Tel: 01928 792545
E-mail: [email protected]
ABI/2
Andrew Walton
ONS, Room 1.459, Government Buildings,
Cardiff Road, Newport. NP10 8XG.
Tel: 01633 812945
E-mail: [email protected]
Annex
Methodological Aspects of the ABI
Sampling and Data Editing
The ABI is sampled from the population of reporting units on the
IDBR. A reporting unit (RU) is usually a single complete business
but in some cases arrangements have been made with businesses
to collect data either for several businesses combined or for parts of
businesses of heterogeneous nature. Each RU on the IDBR also
consists of one or more local units (i.e. individual sites) at which it
operates. Local units are particularly important when considering
any kind of geographical analysis.
The sample size of ABI/1 in 1998 was approximately 78,500. The
ABI/2 sample size is slightly lower at about 75,000 because of the
industry sectors not covered. Sampling is done for ABI/1 and the
ABI/2 sample then automatically results by excluding the appropriate
industry sectors. The sample design is a stratified random one with
three stratification dimensions. Strata are defined in terms of:
●
●
●
six employment sizebands (1–9, 10–19, 20–49, 50–99, 100–
249, 250+).
region (viz: England & Wales combined/Scotland/Northern
Ireland).
SIC industry.
Within England and Wales industry stratification is at 4 digit SIC
level. Within Northern Ireland it is at 2 digit SIC level and within
Scotland at a hybrid 2/3/4 digit level. All stratification variables are
taken from the IDBR. Special arrangements have been agreed with
the Scottish and Northern Ireland Offices to allocate larger than
proportional sample sizes to these two regions (viz: 9,000 and 3,000
respectively).
Subject to the sample size constraints above, the sample has been
allocated to strata using the Neyman optimum allocation method
which minimises the expected variance of total turnover over all
strata. This results in the strata corresponding to the largest
businesses being completely enumerated in all industries. In most
cases some of the strata in sizebands below this are also completely
enumerated. In addition, cases with high register turnover and low
register employment (defined as £50m or greater turnover and less
than 10 employment) are also completely enumerated. This provides
a limited form of stratification by the turnover variable and significantly
reduces the expected variance of estimates derived from the sample.
In 1999 this procedure was extended to provide complete
enumeration of all businesses with high register turnover regardless
of their register employment.
Below the threshold for complete enumeration, the sample is rotated
as follows. Businesses with less than 10 employment are completely
replaced each year. Businesses with employment from 10 up to the
threshold for complete enumeration are given a rotation rate of 50
per cent (i.e. half are replaced each year). This system of rotation is
designed to spread the form filling burden on businesses while
retaining a reasonable degree of matching of the sample between
consecutive years which improves the accuracy of estimates of
change between years.
A variety of credibility checks is applied to data received from
businesses and an attempt is made to correct or confirm data failing
these checks. In addition checks are carried out against data received
from the same businesses by other ONS inquiries (in particular, the
monthly and quarterly turnover inquiries and PRODCOM). At present
this form of checking is limited by resource constraints to businesses
with 250 or greater employment.
National Analyses
Data obtained from the sample are grossed up to the population of
the IDBR using the combined ratio estimator, which is similar to the
standard ratio estimator except that strata corresponding to different
sizebands may be combined if the sample size would otherwise be
too small. This estimator makes use of register auxiliary variables to
improve the accuracy of estimation. The auxiliary variable used in
grossing the employment data in ABI/1 and the employment costs
data in ABI/2 is the IDBR employment. For all other data in ABI/2
the auxiliary variable used is register turnover. The choice of auxiliary
variable was determined by examination of the size of expected
sampling errors using different auxiliary variables.
In order to improve the robustness of the above estimator a procedure
for dealing with atypical observations (outliers) has been adopted.
This consists of creating additional post-strata into which the outliers
55
are placed as though they had been completely enumerated at the
sampling stage (they are also removed from their “natural” strata).
The criteria for marking businesses as outliers have been determined
empirically so as to provide robust estimates. In ABI/1 a business is
treated as an outlier if the ratio of its reported employment to its
register employment exceeds 20. In ABI/2 a business is treated as
an outlier if either it is an outlier for ABI/1 or if the ratio of its reported
turnover to its register turnover exceeds 50. If a business is marked
as an outlier it is treated as such for all variables collected in the
survey.
Another aspect of the grossing methodology is the way in which
births and deaths are treated. The issue to be addressed is one of
time lags from the birth or death of a business and this event being
recorded on the IDBR. This is usually several months. Some births
which exist in reality will not therefore be part of the register population
at the time of inquiry selection. Also some of the units selected will
turn out to be dead. Inferences could be drawn about the number of
unrecorded deaths in the population from sample information but
the same is not true of unrecorded births.
the ratio R =X divided by register employment for each variable of
interest X. The model is in two parts. Firstly, the probability of a nonzero value is assumed to be given by a logistic linear model with
independent variables register employment sizeband, register SIC
and register region code. Secondly, given a non-zero value, R is
assumed to follow a log-linear model with the same set of
independent variables as for the logistic model. The parameters
estimated for these models are then used to derive proportions for
each local unit (adding to unity across all local units in a reporting
unit) which determine the apportionment of the reporting unit total
across the local units.
The models are fitted to the reporting unit data and then applied to
local units to conduct the apportionment (the independent variables
being available on the IDBR for both reporting units and local units).
In order to make this switch from reporting unit to local unit as reliable
as possible, the larger reporting units were excluded from
consideration in fitting the model (in fact, only reporting units with
less than 100 employment and less than 3 local units were used to
fit the model).
The ABI uses the method applied in nearly all ONS inquiries which
gross to the IDBR population. This consists of assuming that the
smaller unrecorded deaths are counter-balanced by an equal
The modelling process also required decisions concerning the most
appropriate degree of aggregation of sizebands, SIC codes and
region codes for use as independent variables. It was decided to
number of unrecorded births and that the larger unrecorded deaths
are not compensated for at all by unrecorded births. In the case of
the ABI a threshold of 50 (in all industries) for the IDBR employment
has been set to distinguish large from small deaths. The above
assumptions are known to be imperfect because of differences in
the balance of births and deaths between industries and also over
the economic cycle. However there is little information available at
present to improve on the above and no clear evidence of systematic
bias.
work at as high a degree of aggregation as possible without sacrificing
a significant degree of “goodness of fit” of the model. This has led to
the following groupings which have been used for all survey variables:
The extent of the sampling error which can be expected with the
above sample size/design and associated estimator is as follows.
The coefficient of variation of the estimate of total employment across
all industries is about 0.4 per cent and the coefficient of variation of
total value added across all industries is about 1.5 per cent. This
means that the 95 per cent confidence interval for total employment
is +0.8 per cent and the 95 per cent confidence interval for total
value added is +3 per cent.
For the questions on employees in ABI/1 (i.e. total employees and
the 4-way split) there are other data sources which may be used to
apportion from RU to LU level, viz: the ARI in Great Britain and the
2-yearly Censuses of Employment in Northern Ireland. The
information on local unit distribution of employees from these sources
is potentially better than that obtained from the above modelling
process and so is used in preference to modelling where possible.
The use of this approach is however limited to the 5 variables referred
to above and to ABI reporting units which are also present in ARI or
the NI censuses.
Local unit apportionment
Since the data collected in ABI is for reporting units (usually complete
enterprises), the first stage in any analyses at sub-national level is
to apportion reporting unit data across the local units (LUs) within
the reporting unit. This apportionment is carried out by modelling
56
●
●
●
Employment Sizebands: 1–2, 3–4, 5–9, 10–19, 20–49, 50–
99, 100–249, 250+.
Industries: 2 digit SIC for the logistic model and 3 digit SIC
for the log linear model
Regions: NUTS2 level
Sub-national Analyses
Sub-national analyses of ABI data are obtained by a combination of
conventional and synthetic estimation methods. At high levels of
aggregation conventional estimation is used, while at lower
aggregation levels conventional methods are supplemented by
synthetic estimation using IDBR data. These procedures are
described below.
A post-stratification of all local units in the RU sample and population
is used. The post-stratification variables are the LU region, the parent
RU sizeband and the parent RU SIC industry of the LU. The parent
RU sizeband is the same as is used for stratification in national
analyses. Otherwise the aggregation levels are LU NUTS1 region
and 2 digit parent RU SIC, except that some miscellaneous
aggregation of post-strata (both in terms of region and SIC) is
necessary to provide adequate sample sizes. The estimator used is
the standard (uncombined) ratio estimator with register LU
employment as the auxiliary variable.
The use of a different post-stratification for sub-national analyses to
that used for national analyses gives rise to a problem of consistency
between national and sub-national results in so far as the total of all
regional results for the UK may not equal the UK value obtained
from the national estimation system. To overcome this problem
scaling factors have been introduced for each survey variable. These
are applied to the weights arising from the sub-national estimation
system and are set so as to ensure that additivity as described above
is achieved. Furthermore, it has been agreed with the Northern
Ireland Office to constrain ABI results for the 5 employee variables
for Northern Ireland to the values obtained from the Northern Ireland
December Quarterly Employment Survey which has a high coverage.
The constraining is conducted at the level at which the NI results
are published (viz. 2 digit SIC), and requires the introduction of
additional scaling factors for these 5 variables. Finally it should be
noted that totals and other forms of derived variables are calculated
after grossing up the sample data and are not grossed separately in
the sub-national system as this would give rise to another form of
non-additivity or incoherence.
In principle the above methodology will provide estimates for any
required sub-population (or domain, in the terminology used below),
including very small areas such as local authority wards. However,
in practice, the sampling errors for very small domains are extremely
large. Also some domains will contain no sampled units so that the
resulting estimate for them will be zero. Although a zero estimate
may be within sampling error expectations, it is presentationally
troublesome since it may be known in advance that the correct
domain estimate should be greater than zero.
It is not therefore realistic to use this estimation methodology for
very small domains and a set of “minimum domains” for which it
may be used has been imposed. These minimum domains are twodimensional in the sense that they are defined in terms of a given
level of SIC disaggregation and a given level of geographical
disaggregation, each of which may be different for any particular
minimum domain. The minimum domains have been defined taking
into account both the sampling error associated with them and their
sample size in terms of the number of local units they contain.
Below the level of minimum domains a form of synthetic estimation
is used to obtain the required estimates. Essentially, if a domain is a
subset of a minimum domain, estimates for it are obtained by prorating on the basis of the total register employment of the domain of
interest and the total register employment of the minimum domain.
Sampled units are removed before prorating and added back
afterwards. If a domain straddles two or more minimum domains it
is split into smaller domains which do not cross minimum domain
boundaries. These are then estimated separately and the results
added. The use of synthetic estimation for very small domains means
that estimates of error are not available. Sampling error does not
increase beyond minimum domain level, but other forms of error
may be introduced and these are not quantifiable.
57
Virtual Micro Data Laboratory Data Brief: Winter 2008
Estimating capital stock at the firm level
Robert Gilhooly
Estimates of the stock of capital at the level of the individual firm are an essential
component of productivity analyses using microdata. It enables the user to estimate a
production function and then investigate the factors which drive the most productive
firms e.g. technology use, international reach, industry and geographical affects. The
Office for National Statistics (ONS) does not ask firms what their capital stocks are;
hence, it is necessary to construct capital stock series and then link these to the other
confidential data sets held within the Virtual Microdata Laboratory (VML).
The Microdata Analysis & User Support (MAUS) team has recently made a new
version of the Capital Stock data set available within the VML. It spans the years
1979-2005, and creates a firm-level estimate of capital stock for approximately
42,000 enterprises. This data set contains methodological improvements compared to
previous versions, and, specifically, addresses the negative capital stock series which
occurred in previous constructions of the data set.
The major development in the latest version of the capital stock data set is the ability
for users to tailor the data set for their own purposes. The syntax file available within
the VML has been designed to be very flexible; users can now easily create their own
capital stock estimates by selecting from a range of key variables & methods, and rerunning the file in their own work area. For example, by selecting a higher tolerance
ratio of imputed to real investment figures, a user can significantly expand the number
of enterprises for which a capital stock series are calculated.
This Data Brief is intended to serve as a description of the techniques used to
constructed the capital stock data set. It covers:
• The Perpetual Inventory Model (PIM)
• The data sets used
• Allocation of the aggregate capital stock series to firms’ first appearance
• Dealing with missing values and imputation techniques
• Addressing negative capital stock series and injecting capital
• Comparing the micro-Capital stocks to the aggregate published series
• Using the capital stock in productivity estimation
1
The capital stock data set is unlike the majority of the other data sets held in the VML
because it is a “constructed” data set. Therefore there are many decisions needed to
overcome the issues outlined above, and there is not a definitive approach to take. The
user of this data set is advised to read this document and then decide whether to use
the file as it stands or adjust it using the options outlined. The “Technical Guide”
provides a step-by-step guide to the construction of the capital stock, and this can
used as an aid to understanding the main syntax file.
1.
Perpetual Inventory Model
The stages leading up to running the PIM require several decisions to be made on the
methodology to allocate capital and deal with missing variables (see section 3);
however, running the PIM is relatively uncontentious. In summary, the PIM
depreciates the previous year’s capital stock and then adds on the current investment,
and this process continues for each year of the series. At this stage all that is left for
the user to do is to select the depreciation rates they would like to use. The MAUS
team follow the depreciation rates set out by ONS: Plant & Machinery 6%, Buildings
2%, Vehicles 20%. However, alternatives such as those based on BEA estimates
include: Plant & Machinery 13%, Buildings 2.5%, Vehicles 25%.1
The PIM uses geometric depreciation as follows:
Firm Capital Stockasset,yeari = (Allocation firm Capitalasset * Firm Investment Shareasset ) + rncapex yeari
Firm Capital Stock asset,yeari +1 = [Firm Capital Stock asset,yeari *( 1 − asset depreciation) ] + rncapex yeari +1
Firm Total Capital Stock year = ∑ Firm capital asset
We choose to use geometric depreciation, because the asset price depreciation and
physical deterioration (age-efficiency profile) of the assets coincide.2
2. Data sets
The capital stock data set is constructed from the Annual Respondents Database Panel
Data (ARDPD) set, which is itself constructed from the Annual Respondents
Database (ARD)3. The ARD holds firms’ responses to the Annual Business Inquiry
(ABI), while the MAUS team does some work on the raw ABI files to ensure that
there is longitudinal consistency in variables and firm reference numbers.
The ABI is the most comprehensive business survey conducted by ONS and covers
over 100 key economic variables, such as: turnover; production and operation costs;
employment: industry classification, and investment. The ABI has surveyed all
sectors since 1997, while firms in the manufacturing and construction sectors returned
questionnaires from 1973 and 1994 respectively. It is a census of large businesses,
and a stratified sample of small and medium sized enterprises. The stratified sampling
1
See Fraumeni (1997)
See ONS Productivity Handbook (2007)
3
Detailed information about the data sets held in the VML can be found at:
http://www.ons.gov.uk/about/who-we-are/our-services/vml/about-the-vml/datasets-available/datasetdownloads/index.html
2
2
framework means that smaller firms move in and out of the survey, and to create
capital stock estimates it is necessary to have a panel which accounts for such
“missing years”. The ARDPD solves this problem by creating entries for firms when
they do not appear in the ARD, but are still trading e.g. if a firm is in the ARD in
1997 and 2000, the ARDPD ensures it is present for 1998 and 1999 also. This process
creates a lot of missing values which must be imputed – in the previous example we
would not have any investment figures for 1998 and 1999 to feed into the PIM – and
section 2 outlines the approach we have taken to address this issue.
To create a PIM it is necessary to allocate shares of the aggregate capital stock series
to firms for the first year that they appear in the survey (as shown previously, the PIM
depreciates this value, adds the net investment the following year and repeats these
steps for all years). The aggregate capital stock series used is the Volume in Capital
Services (VICS)4. Capital services are a flow measure which reflects the input of
capital into production; thus, they are more suitable for analysing productivity than
the National Accounts wealth estimates of capital stock. The main difference between
the National Accounts measure of capital stock and the capital services estimate is
how the net stock estimates are weighted. Capital stocks are weighted using asset
purchase prices while capital services are weighted using asset rental prices. The
rental price is sometimes referred to as the user cost of capital, and it captures three
components: nominal rate of return, depreciation of the asset, and the change in the
purchase price of the asset. The nominal rate is not observed, and is often interpreted
as the risk-free interest rate plus a risk premium. If a firm is to rent an asset, the lease
price must cover the loss in asset value due to depreciation; the more an asset
depreciates the greater its rental price. Similarly, a declining purchase price must be
covered by a higher rental price to compensate the owner for the capital loss. The later
two key components can be illustrated by ICT assets. Prices of ICT goods continue to
fall sharply, and a capital stock measure would result in such goods being given less
weight; however, this implies a rising rental price and a capital services measure
would give such assets a higher weight. As noted by Wallis & Dey-Chowdhury
(2007), there is a divergence between the volume of capital service and volume of
capital stock series after 1980, which becomes increasingly stark after 1990. This
divergence is caused by the move towards short-lived and more productive assets,
such as computers, for which the flow of capital services is high. The standard capital
stock measure does not capture this shift and so understates growth in the productive
input of capital in the UK economy.
A real firm level capital stock series is constructed by MAUS using deflators to adjust
for price changes in asset values over time. Deflators are constructed by SIC letter and
asset type for each year, using 1995 as the base. The deflators are then applied to the
investment figures (net capital expenditure) for each asset before they are used in the
PIM. Real input and output series are required to remove price affects and reveal
changes to firms’ production efficiency.
4
More information about VICS can be found at:
http://www.statistics.gov.uk/StatBase/Product.asp?vlnk=14205&Pos=&ColRank=2&Rank=1000
3
3. Allocating capital to firms
The first stage in the production of a capital stock series is to calculate the share of
total industry capital which should be allocated to the firms in your sample. We do
this by assuming that the ratio of firms’ investment (i.e. sum of individual firms’
investment) to industry investment corresponds to their share of industry capital. The
calculation of the initial capital allocation is not precise, and allocating firms too little
capital is one reason for the implausible generation of negative capital stock series
(see section 4). It seems plausible that the stratified sampling framework of the
ABI/ARD results in too low an allocation of industry capital. The largest firms –
presumably with most capital – represent a disproportionate number of the firms in
the microdata. The formulae for the first stage of capital stock allocation is given by:
Allocation firm Capital(asset ) = Industry Capital(asset ) * Firm Investment Share(asset )
Firm Investment Share(asset ) =
∑ (rncapex)
year , siclett , asset
Industry Investment year ,siclett ,asset
Now that we have decided how much capital should be allocated to the firms we are
interested in, the second step is to devise a method for splitting this capital amongst
the firms. At this junction we want to choose a variable which is plausibly related to
capital stock and is available for all the years. The main options are “total purchases”
and “materials & fuels” – both could be seen as proxies for the capital stock and firms
are asked these questions in all years. Alternatives, which are only available from
1997 onwards, include: “number of local units”, “Spending on Insurance” and
“Spending on Road Transport”. As noted above, the new flexible .do file allows users
to specify which variable they would like to use, and within this .do file there is
another .do file which compares the alternative options against the number of capital
stock series which are created and are negative. The capital stock data set created by
the MAUS team indicates that the best results (i.e. fewest negative series), for all
years, are obtained by specifying the key variable to be “total purchases”. A basic
allocation of firm level capital can be created using:
⎡
⎤
m0 _Share = ⎢key var
⎥
key
var
∑
siclett, year ⎦
⎣
We shall refer to the basic formulae above as Method 0 (m0) in the syntax file.
Method 0 is only able to split capital at the level of SIC letters, which is another
reason why capital stock series may become negative. Consider sectors in the same
SIC letter but different SIC sub-divisions (e.g. a1=Civil engineering, a2=plumbing):
Method
m0
m0
m0
m0
Firm #
1
2
3
4
Year
1998
1998
1998
1998
SIC
letter
a
a
a
a
SIC
code
a1
a1
a2
a2
total
purchases
20
20
50
100
Share of
capital
0.11
0.11
0.26
0.53
We can make an attempt to (partially) remedy this situation by using a more advanced
method of allocating capital to firms. We can weight the allocation of capital by both
4
Total Purchases and Employment (or an alternative variable). The formulae below
outline two alternative methods for creating a share measure:
⎤
⎡
⎤ ⎡∑ employmentsic 3d , year
m1_Share = ⎢totpurch
∗⎢
⎥
∑ totpurchsic3d, year ⎦ ⎣
∑ employmentsiclett , year ⎥⎦
⎣
⎡
⎤ ⎡∑ totpurchsic 3d , year
m2 _Share = ⎢employment
∗
∑ employment sic3d, year ⎥⎦ ⎢⎣
⎣
⎤
∑ totpurchsiclett , year ⎥⎦
The tables below illustrate how the 2 different methods create different shares
compared to the original methodology.
Method
m1
m1
m1
m1
Firm
#
1
2
3
4
SIC
letter
a
a
a
a
SIC
code
a1
a1
a2
a2
Total
purchases
20
20
50
100
Employment
50
25
200
500
Method
m2
m2
m2
m2
Firm
#
1
2
3
4
SIC
letter
a
a
a
a
SIC
code
a1
a1
a2
a2
Total
purchases
20
20
50
100
Employment
50
25
200
500
SIC
code:
TP
0.50
0.50
0.33
0.67
SIC
code:
Employ.
0.67
0.33
0.29
0.71
SIC code by
letter:
Employment
0.10
0.10
0.90
0.90
SIC code by
letter: Total
purchases
0.21
0.21
0.79
0.79
Share
of
capital
0.05
0.05
0.30
0.60
Share
of
capital
0.14
0.07
0.23
0.56
The flexible set-up of the capital stock syntax allows a user to specify which method
(m0, m1 or m2) they would like to use in their capital stock. Investigation by the
MAUS team indicates that Method 1 produces the best results. See table below:
Counts of the number of negative capital stock series created under different methods
method
All assets
Plant & Machinery
Vehicles
Buildings
m0_totpurch
229
173
1719
207
m0_matfuel
622
444
4888
943
m1_totpurch
216
164
1652
189
m1_matfuel
613
446
4719
853
m2_totpurch
293
220
2596
333
m2_matfuel
393
282
4525
689
This table illustrates one of the major problems in constructing the capital stock data
set. The volatility in the net investment figures results in over 10% of the capital stock
series for vehicles becoming negative. Obviously, it is impossible for a firm to hold a
negative stock of vehicles, and this points to a combination of the main problems: not
allocating enough initial capital; an inability to split capital at a detailed enough level;
imputation techniques failing to create a large enough investment to assets (e.g.
vehicles) when there are missing cell values; and, possibly, using too high a
depreciation rate.
5
In the current methodology, all firms are allocated a share of the aggregate capital
stock when they first enter the data set. The PIM which we specify also adds on the
firm’s investment which it undertakes in its first year. It could be argued that if an
investment figure is available we should just use that as the firm’s first year capital
stock; however, it seems highly probable that the firm was in existence before it is
picked up by the ABI survey, thus allocating a share of the aggregate capital stock
figure in its first year provides a proxy for a firm’s “pre-ABI” capital stock.
4. Missing values and imputation techniques
Some of the most important decisions to make in the construction of the capital stock
data set are with regards to the treatment of missing cell values, and the cleaning of
the data set before a share of capital is allocated and then the PIM is run. There are
many different approaches that we could take to this problem, and this section
outlines how MAUS have chosen to impute cell values.
Employment is used as the spine for many of the imputations, plus the allocation of
the aggregate capital share, thus the first stage is to ensure that we have a complete
employment series. Generally, we can be confident using employment as this
information is available from the Inter-Departmental Business Register (IDBR);
hence, even if a firm does not respond to the questions in the ABI survey we are likely
to have an employment figure. When a cell value is missing between years we simply
interpolate the value. Missing (usually coded 0) cell values are more common in the
first years of a company’s existence, and these are filled in rolling back a three-year
average of the employment figures available. This is illustrated in the table below:
Year
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
Original cells
Employment
0
0
0
100
120
120
114
150
170
165
Final cells
Adjusted Employment
108
111
113
100
120
120
114
132
150
170
165
We require a complete series of asset investment figures to compute the PIM and
must also impute these missing values. The investment series in the ARD are net
capital expenditure (capex) by asset. It is worth repeating that these figures are net i.e.
acquisitions minus disposals. While we do ultimately require the net investment figure
for PIM, the volatile nature of firm investment and missing values provide a challenge
for imputation – we are put further into the dark as acquisition and disposals series,
which might provide more information on underlying investment patterns, are not
available. The following graphs illustrate the problems we face:
6
35000
30000
25000
imputed 1
20000
imputed 2
15000
imputed 3
10000
5000
0
-
-
1985
1986
10000 15000
1987
-
1988 1989
20000 25000
1990
1991
1992
The graph above shows 3 methods of imputing missing values (labeled “-“; while the
figures on the x-axis represent real values e.g. 15,000 for 1988) for a firm which has
four positive net capex cells from 1985 to 1992. Imputation 1 interpolates the 1989
value and provides a constant extrapolation at the edges (1985, 1986 & 1992)5.
Imputation 2 also interpolates, but provides momentum extrapolation at the edges.
Imputation 3 creates an average capex figure over the real values and then uses this
number to fill-in the missing cells. At this stage, the three capex series already paint
different pictures, which will impact on the estimated capital stock series produced by
the PIM. It is critical to stress the importance of the net position i.e. negative values
are likely.6 The graph below shows what happens to the imputed series when net
capex becomes negative in some years:
40000
30000
20000
imputed 1
imputed 2
10000
imputed 3
-20000
1985 1986 1987
1988 1989
-
-6000
-17000
-
13000
20000
-10000
-
-
0
1990 1991 1992
The capex values given above are typical of the values we see in the microdata, and
reflect the volatile nature of firm investment patterns. The firm shown above is a net
investor (+10k) over the four years for which we have data; it is not clear which
imputation technique (if any) could be deemed to be “best” given the volatile nature
of firm investment. Methods 1 and 2 give strong investment figures, but the true
values could equally be negative. It should be noted that (almost) all imputation
5
This technique is used in Martin (2003).
The use of net investment partially counters criticism of constructing capital stock estimates at the
Reporting Unit level i.e. firms which close local units should report a negative investment. However, as
illustrated by Harris (2005), firms will not be allocated additional capital when they acquire new local
units. This is another reason why the capital stock series may underestimate the true figures.
6
7
techniques will smooth the estimated investment patterns below the volatile true
series which we would see if we had the real investment figures for the missing years.
The MAUS team decided to use the employment series to help with the imputation of
capex. It is hoped that this gives the estimated capex figures more real information
compared to just using one of the techniques outlined above. The latest capital stock
data set thus creates a capex series which inputes using an average of net capital
expenditure per employee for the real data, and then applies this to the missing values.
This is similar to imputation method 3 above, and while it produces low investment
figures for missing years it does ensure some consistency in the net position i.e. net
(positive) investors continue to invest over the years. This technique is increasingly
valuable in highly volatile series, where the other methods often impute large negative
figures. The table and graph below illustrate the imputation for a firm which is a net
investor for the four years which we have real data.
year
capex
1985
1986
1987
1988
1989
1990
1991
1992
Averages
employment
100
145
200
190
250
200
190
230
195
Average capex
per employee
20000
13000
-17000
-6000
2500
Imputed
1282
1859
20000
13000
3205
-17000
-6000
2949
n/a
13
25000
300
20000
250
15000
10000
200
5000
150
-15000
-20000
employment
-
-6000
-17000
-
13000
-10000
20000
-5000
-
-
0
Imputed
1985 1986 1987 1988 1989 1990 1991 1992
100
50
0
While it is possible to impute for any number of missing values the quality of the
imputation will suffer when we have fewer real values to base our calculations on (i.e.
we are likely to impute disposals of capital for all years for some firms). In practice, it
is prudent to set a tolerance level. Currently the maximum ratio of imputed to real
values is set at a maximum of 1:1. Lowering our intolerance to imputed values will
create estimates of firm capital stock for many more firms, and is likely to pick-up
more SMEs due to the stratified sampling framework within the ABI. The table below
outlines the number of firms for which we calculate capital stock given different
8
tolerance levels. Users can easily alter the “calc_ratio” in the syntax file and re-run to
suit their needs.
Calc_ratio
Tolerance level (max number of
imputed to real values)
0.33
0.66
1
1.5
2
5.
Number of firms with
capital stock series
1:3
2:3
1:1
3:2
2:1
13131
21809
41782
50622
74356
Dealing with the remaining negative capital stock series
After missing values have been imputed, the initial share of capital is allocated to
firms, and the PIM is run, we encounter many capital stock series being produced
which have negative values. Given that a negative capital stock series is impossible in
reality, we can address this problem by injecting additional capital at a firm level.
This can be done by stipulating that a firm’s capital stock must not go below zero,
increasing the investment figures in the proceeding years and then re-running the
PIM. It is advisable to inject the capital close to the year in which the capital stock
turns negative; consider the huge capital injection which would be required if a 20
year vehicles series turned negative in the last year i.e. a -£1000 figure would need
£87,000 (1000*[1/0.8]^20) injection in the first year to ensure that an entirely positive
series is created. For this reason we localize the injection to at most three years
previously:
{
}
Asset Investment Injection = ( negative capital stock )* ⎡ 1
^n ⎤
⎢⎣ (1 − depreciation ) ⎥⎦
The table below illustrates how a firm’s vehicle stock may turn negative – assuming
20% depreciation – and how a localized injection of capital can ensure that stocks do
not fall below zero:
year
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
net
capex_vehicles
10,000
12,000
-5,000
-2,000
20,000
40,000
-20,000
-50,000
20,000
5,000
5,000
Capital
allocation
50,000
0
0
0
0
0
0
0
0
0
0
capital stock
60,000
60,000
43,000
32,400
45,920
76,736
41,389
-16,889
6,489
10,191
13,153
adjusted
capex
10,000
12,000
-5,000
-2,000
52,986
40,000
-20,000
-50,000
20,000
5,000
5,000
adjusted capital
stock
60,000
60,000
43,000
32,400
78,906
103,125
62,500
0
20,000
21,000
21,800
In practice, the calculated capital stock series often becomes negative for many
observations within a series; hence, it is often necessary to run the above injection
several times. The current syntax file calls in the “localised_boost.do” file several
times in a row to deal with this problem.
9
6.
Comparing the firm-level Capital to the published series
The firm level capital stock series can be compared to the aggregated published series
to see if they maintain any consistency with the economic cycle. The graph below
shows the main series for whole economy and production industries VICS, and as can
be seen there are definitive dips in the growth of the series associated with the ’92
recession and the bursting of the dot-com bubble in 2001.
0
2
4
6
8
Capital Services
1980
1985
1990
1995
2000
2005
y ear
market_s ec tor
all_production_indus tries
A comparison of the unweighted average firm-level growth in capital stock with the
published series by industry grouping also shows some consistency with the economic
picture. The following graphs provide a selected comparison within manufacturing by
SIC letter groupings.
Capital Stock Growth - SIC letter: DA
-10
-10
-5
-5
0
0
5
5
10
10
Capital Stock Growth - SIC letter: DBC
1980
1985
1990
1995
2000
2005
1980
1985
1990
year
1995
2000
2005
year
grow th_avCS_all95
DA
grow th_avCS_all95
Capital Stock Growth - SIC letter: DG
DBC
-20
-10
-10
-5
0
0
10
5
20
10
Capital Stock Growth - SIC letter: DK
1980
1985
1990
1995
2000
year
grow th_avCS_all95
2005
1980
1985
1990
1995
2000
2005
year
DG
grow th_avCS_all95
DK
10
Capital Stock Growth - SIC letter: DM
-20
-50
0
0
50
20
100
40
150
Capital Stock Growth - SIC letter: DL
1980
1985
1990
1995
2000
year
grow th_avCS_all95
2005
1980
1985
1990
1995
2000
2005
year
DL
grow th_avCS_all95
DM
It is not surprising that the unweighted microdata series show higher volatility than
the published series, and the fact that we do not calculate capital stock for all firms
also adds to the erratic nature of the series. We can gain confidence from the above
graphs showing similar patterns of growth. The large volatility within SIC letter DL
(“Manufacture of electrical & optical equipment”) is likely to represent the dotcom
and telecoms bubble, and illustrates the benefit of examining the microdata to gauge
the scale of the events within the industry – the published series appears to mask the
magnitude of the bubble and the changes taking place within firms.
7.
Using the capital stock in productivity analysis
One of the key advantages of microdata, rather than macro-level statistics, is that they
permit the examination of a great deal of variability that occurs at firm-level - industry
statistics often hide this important information. In terms of productivity, it is possible
to isolate the effects of factors such as region, firm size and foreign ownership on the
productivity of individual businesses. There have been many studies conducted on
productivity within the VML - a selection of references are included in the appendix.
Microdata research also has implications for our understanding of the aggregate
figures. Haltiwanger (2000) reports that within US manufacturing roughly half of the
multi-factor productivity growth can be accounted for by the re-allocation of outputs
and inputs away from the less productive to more productive firms. This adds to the
understanding of aggregate productivity growth; productivity growth occurs with
entry and exit i.e. creative destruction can be viewed as driving industry-level
productivity.
The remainder of this section outlines the basic steps which are often used in
productivity analysis on microdata, and then presents some research on sensitivity
analysis to different capital stock specifications.
7.1 Specifying a productivity measure and a production function
Productivity is a derived figure (output/inputs); hence, we require our input and
output measures to be consistent with each other. To remove price effects, deflators
are used to produce real measures. Productivity growth occurs because of improved
efficiency, such as using fewer inputs to produce the same outputs, or through inputs
being used more effectively to produce outputs of greater value – nominal series
would distort this information.
11
The primary inputs into the production process are Labour (L) and Capital (K), “other
inputs” e.g. energy and materials can be grouped under intermediate consumption. By
using Gross Value Added (GVA) we can take account of the changing composition of
intermediate consumption, and thus GVA is the preferred measure of output rather
than total output. A firm which purchases components rather than raw materials may
increase total output; however, this would not represent productivity growth.
Multi-factor productivity (MFP) is defined as the residual output growth of a firm,
industry or economy, after calculating the contribution from all inputs. Put another
way, it is the output growth which cannot be explained by increasing volume of inputs
and is assumed to reflect increases in the efficiency of use of these inputs. Typically it
is estimated indirectly as the residual after estimating the effect of the change in the
volume of inputs. A common form of MFP estimation is Labour-Capital value added
productivity. Ideally, this uses GVA as the output measure, while a Quality-Adjusted
Labour Input (QALI) and a VICS-based capital estimate are used as the inputs. QALI
attempts to adjust to the characteristics of the labour force i.e. a highly skilled
workforce would be treated as a more productive and valuable asset.
There are many different types of production functions which could be specified e.g.
Cobb-Douglas, Constant Elasticity of Substitution, Transcendental logarithmic etc. To
illustrate the sensitivity of coefficient estimates to changes in the capital stock
estimation we shall use the basic Cobb-Douglas model as outlined below:
Production Function : Q = F(K,L)
Cobb - Douglas : Q = AK α Lβ
Taking logarithms and estimating by OLS for each industrial sector gives us the
following basic formulae:
∧
∧
∧
∧
∧
log Qit = log Ait + α log K it + β log Lit + ε it
Running this specification by 3 digit SIC allows the estimated coefficients to reflect
industry specific factors of production i.e. we assume that firms in the same industry
have access to the same technology and labour.
Altering the depreciation rates used will feed-through into the construction of the
capital stock series. While the depreciation rates affect firms in the same manner (i.e.
buildings do not depreciate at different rates between firms), the interaction of
investment spending and depreciation with the PIM methodology has the potential to
alter the picture of capital holdings across firms. The following table provides
coefficient estimates for a selection of different industries, and shows how these
estimates change when different capital stock series are produced.
12
Depreciation rates: asset
Industry
SIC
code
Description
11
Crude Extraction
11
Crude Extraction
11
Crude Extraction
11
Crude Extraction
11
29
Crude Extraction
Manufacturing of Machinery
& Equipment
Manufacturing of Machinery
& Equipment
Manufacturing of Machinery
& Equipment
Manufacturing of Machinery
& Equipment
Manufacturing of Machinery
& Equipment
52
Retail
52
Retail
52
Retail
52
Retail
52
Retail
72
Computer & related
72
Computer & related
72
Computer & related
72
Computer & related
72
Computer & related
73
Research & development
73
Research & development
73
Research & development
73
Research & development
73
Research & development
29
29
29
29
B
3%
2.50%
2%
1.50%
1%
3%
V
30%
25%
20%
15%
10%
30%
PM
10%
8%
6%
4%
2%
10%
2.50%
25%
8%
2%
20%
6%
1.50%
15%
4%
1%
10%
2%
3%
2.50%
2%
1.50%
1%
3%
2.50%
2%
1.50%
1%
3%
2.50%
2%
1.50%
1%
30%
25%
20%
15%
10%
30%
25%
20%
15%
10%
30%
25%
20%
15%
10%
10%
8%
6%
4%
2%
10%
8%
6%
4%
2%
10%
8%
6%
4%
2%
Coefficient estimates:
***significant at 0.1%
α
β
0.589***
0.355***
0.591***
0.355***
0.593***
0.355***
0.594***
0.355***
0.595***
0.355***
0.271***
0.673***
0.287***
0.658***
0.297***
0.648***
0.302***
0.643***
0.303***
0.642***
0.329***
0.699***
0.329***
0.701***
0.328***
0.703***
0.327***
0.706***
0.325***
0.709***
0.191***
0.821***
0.195***
0.817***
0.187***
0.831***
0.191***
0.826***
0.192***
0.828***
0.0494
1.011***
0.0526
1.007***
0.0560
1.003***
0.0594*
0.999***
0.0631*
0.994***
As can be seen, changing the depreciation rates used to calculate the capital stock has
almost no impact on the estimated coefficients of the industry production functions
(this result holds for all industry breakdowns, not just those presented above). This
seems intuitive as the capital asset series are affected in the same way by the PIM
operation. Running the production functions by two digit SIC code illustrates the
importance of controlling for different technologies e.g. the relative importance of
capital in Crude Extraction, compared to the importance of labour in Computer &
Related activities.
In a similar vein, we can re-run the production functions for the capital stock series
which are created after altering our tolerance to the use of imputed values (the
“calc_ratio”). The table below shows the affects on the estimated coefficients.
13
Tolerance Specifications
Industry
SIC
code
Description
Calc_ratio
Tolerance
level
Coefficient estimates:
***significant at 0.1%
α
β
11
Crude Extraction
0.33
1:3
0.790***
0.271***
11
Crude Extraction
0.66
2:3
0.587***
0.335***
11
Crude Extraction
1
1:1
0.593***
0.355***
11
Crude Extraction
1.5
3:2
0.596***
0.377***
11
2
2:1
0.597***
0.384***
0.33
1:3
0.300***
0.704***
0.66
2:3
0.291***
0.660***
1
1:1
0.297***
0.648***
1.5
3:2
0.302***
0.646***
29
Crude Extraction
Manuf Machinery &
Equipment
Manuf Machinery &
Equipment
Manuf Machinery &
Equipment
Manuf Machinery &
Equipment
Manuf Machinery &
Equipment
2
2:1
0.304***
0.645***
52
Retail
0.33
1:3
0.430***
0.588***
52
Retail
0.66
2:3
0.390***
0.618***
52
Retail
1
1:1
0.328***
0.703***
52
Retail
1.5
3:2
0.306***
0.723***
52
Retail
2
2:1
0.289***
0.743***
72
Computer & related
0.33
1:3
0.253***
0.761***
72
Computer & related
0.66
2:3
0.161***
0.881***
72
Computer & related
1
1:1
0.187***
0.831***
72
Computer & related
1.5
3:2
0.211***
0.797***
72
Computer & related
2
2:1
0.220***
0.786***
73
Research & development
0.33
1:3
0.331***
0.693***
73
Research & development
0.66
2:3
0.0720**
0.948***
73
Research & development
1
1:1
0.0560
1.003***
73
Research & development
1.5
3:2
0.0727**
0.978***
73
Research & development
2
2:1
0.0895***
0.943***
29
29
29
29
Altering the tolerance to the use of imputed values has a more noticeable affect on the
estimated coefficients, while the change in the estimates is fairly industry specific –
some industrial groups show little change when more imputed values are used. The
largest changes tend to be in the difference between the 0.33 and 0.66 ratios, and this
is likely to reflect the increasing number of firms used in the estimation techniques
(the 0.33 firm estimations are likely to pick-up a large firm effect too).
7.2 Quality adjusting the stock of labour at the firm-level
The sensitivity analysis presented above regressed real GVA on: real Capital, Labour
and various firm characteristics. It should be noted that the measure of the stock of
Labour is still “nominal”, and we would like to adjust this to provide a “real” measure
of the labour stock. This can be done by quality adjusting firms’ labour to take
14
account of changes to the mix of job types (i.e. part-time vs full-time) and skill levels
within the firm.
The changes within the manufacturing industry provide a good example to
conceptualise the real labour stock. Developed countries have seen a large shift in the
occupational composition of employment away from unskilled low technology
activities towards more highly skilled manufacturing, resulting in a higher spend per
employee. Economic theory says that workers should be paid to represent their
productivity, so it is natural to view a skilled workforce as a more valuable asset than
its unskilled counterpart. As with the capital stock, a firm can engage in human capital
deepening either by increasing investment (i.e. training the workforce) or by changing
the mix of employees (i.e. hiring greater numbers of highly skilled workers). These
two forms of investment are essentially equivalent – the latter represents buying
labour with the skills pre-packaged, and the former buying the additional skills. In the
same manner, examining the firm’s spend per employee also enables us to adjust for
the mix of job-types. For example, a firm may choose to hire 3 part-time workers and
- ignoring price and investment affects - we can easily adjust our labour stock to show
that the real stock has increased by 1.5 units.
The creation of the quality adjusted labour stock can be done by adjusting the number
of employees in each firm to take into account of the change in per employee costs.
Firstly, we adjust the total labour cost figures into real terms, by deflating using the
Annual Earnings Index (AIE) by SIC letter. Secondly, we produce an index of real
labour costs per worker (base year 1995) for each firm. Finally, we use this index to
adjust the stock of firm labour, which should now capture changes to the real labour
stock. This process is illustrated in the table below.
Year
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
Employment
80
85
90
105
110
100
110
150
155
140
130
Total
Labour
Spend
800
800
995
1000
1100
1000
1200
2000
2100
2000
2000
Real
Labour
Spend
931.6
903.7
1090.2
1062.8
1134.0
1000.0
1176.5
1922.3
1978.9
1847.7
1811.5
AEI
85.9
88.5
91.3
94.1
97.0
100.0
102.0
104.0
106.1
108.2
110.4
Real
spend
per
employee
11.65
10.63
12.11
10.12
10.31
10.00
10.70
12.82
12.77
13.20
13.93
QALI
116.5
106.3
121.1
101.2
103.1
100.0
107.0
128.2
127.7
132.0
139.3
Real
Employment
93.2
90.4
109.0
106.3
113.4
100.0
117.6
192.2
197.9
184.8
181.1
The production function now includes “Quality Adjusted Labour” (QAL), and is then
re-estimated using the standard ONS depreciation rates, and a 1:1 tolerance ratio:
∧
∧
∧
∧
∧
log Qit = log Ait + α log K it + β log QALit + ε it
As can be seen from the results below, using a QAL measure significantly affects the
estimated coefficient for labour and capital across industries. This implies that a key
factor in productivity analysis is the construction of the QAL and the selection of the
tolerance to imputed values; depreciation rates chosen are comparatively insignificant.
15
Industry
SIC
code
Coefficient
estimates:
unadjusted labour
α
β
Coefficient
estimates: QAL
measure
α
β
11
0.593***
0.355***
0.299***
0.580***
14
0.208***
0.825***
0.320***
0.553***
15
0.490***
0.526***
0.599***
0.336***
16
0.423***
0.660***
0.666***
0.346***
17
0.339***
0.641***
0.461***
0.424***
18
0.261***
0.767***
0.444***
0.477***
19
0.338***
0.669***
0.493***
0.431***
20
0.333***
0.646***
0.494***
0.307***
21
0.413***
0.583***
0.513***
0.349***
22
0.387***
0.662***
0.548***
0.359***
23
0.498***
0.386***
0.533***
0.297***
24
0.433***
0.576***
0.558***
0.338***
25
0.384***
0.636***
0.499***
0.386***
26
0.422***
0.593***
0.538***
0.407***
27
0.303***
0.630***
0.455***
0.351***
28
0.220***
0.762***
0.363***
0.454***
29
0.297***
0.648***
0.424***
0.429***
30
0.242***
0.771***
0.314***
0.587***
31
0.247***
0.755***
0.353***
0.509***
32
0.385***
0.595***
0.547***
0.336***
33
0.365***
0.631***
0.528***
0.366***
34
0.322***
0.624***
0.425***
0.412***
35
0.225***
0.760***
0.381***
0.518***
36
0.355***
0.690***
0.502***
0.415***
37
40
41
45
50
51
52
55
60
61
62
63
64
70
71
72
73
74
0.420***
0.393***
0.561***
0.284***
0.297***
0.386***
0.328***
0.330***
0.304***
0.294***
0.283***
0.294***
0.342***
0.385***
0.384***
0.187***
0.056
0.209***
0.566***
0.435***
0.527***
0.699***
0.770***
0.587***
0.703***
0.746***
0.745***
0.756***
0.800***
0.730***
0.721***
0.570***
0.594***
0.831***
1.003***
0.698***
0.574***
0.355***
0.562***
0.370***
0.247***
0.310***
0.371***
0.424***
0.223***
0.554***
0.0321
0.336***
0.384***
0.325***
0.351***
0.220***
0.208***
0.161***
0.209***
0.396***
0.473***
0.493***
0.602***
0.502***
0.555***
0.550***
0.702***
-0.237*
1.123***
0.615***
0.602***
0.484***
0.450***
0.640***
0.730***
0.606***
16
7.3 Endogenous determination of the capital stock
The use of Ordinary Least Squares in productivity analysis may not be appropriate
because it treats labour and capital as exogenous variables. Griliches and Mairesse
(1995) originally argued that inputs should be considered endogenous since they are
chosen by the firm based on their own known (or estimated) production capabilities.
Ignoring this endogeneity is likely to bias the estimated coefficients. This section
follows closely the work of Olley and Pakes (1996) as outlined by Javorcik (2003),
which uses a semi-parametric estimation procedure to allow firm-specific productivity
to change over time, and thus reflect that capital stocks are a choice by the firm.7
In the Olley-Pakes methodology firms choose their variable factors of production (e.g.
labour) and a level of investment at the start of each period, these combine with the
current capital stock (i.e. last period’s capital stock depreciated) to form the firm’s
productive capacity. We assume a Cobb-Douglas production function as before:
Cobb - Douglas : Q = AK α Lβ
∧
∧
∧
∧
∧
∧
qit = αk it + βl it + ait + ω it + η it
where qit , k it , lit represent the logarithm of Gross Value Added, capital and quality
adjusted labour, as previously. ωit denotes productivity, and ηit denotes either
measurement error or a shock to productivity – both of these terms are unobservable
to the econometrician. ωit is a state variable which affects the firms investment
decision, while ηit is not a state variable and captures the error term. Capital is a fixed
factor and is affected by the knowledge of productivity in the previous period ωit −1 .
Simultaneity bias arises because the firm’s input choices are determined in part by the
firms knowledge and belief about the distribution of ωit . OLS fails to take the
unobserved productivity differences into account, and the positive correlation between
ωit and inputs in period t produces an upward bias on the coefficient for labour and a
downward bias on the coefficient for capital.
The insight of Olley-Pakes method is that the observable characteristics of the firm
can be modeled as a monotonic function of the productivity of the firm. Inverting
such a function allows one to model the unobserved component of productivity as a
function of the observed variables, namely investment.
iit = iit (ω it , k it )
Inverting this equation gives us unobserved productivity as a function of observable
investment and capital, and thus we can control for it in estimation:
ωit = hit (iit , kit )
∧
∧
∧
∧
∧
∧
qit = αk it + βl it + a it + hit (iit , k it )+ η it
7
A special thanks goes to Alessandra Tucci who provided guidance on my productivity work,
highlighted the Olley and Pakes (1996) and Javorcik (2003) papers, and also allowed me to copy her
Stata code to run the Olley-Pakes method. Any mistakes or errors are my own.
17
The functional form of hit (iit , k it ) is not known; hence, the α coefficient on capital
cannot be estimated at this stage. To obtain an estimate of hit (iit , k it ) we create a
partially linear model including a third order polynomial expansion of capital and
investment. The third order polynomial of iit , kit is:
ψ it = a + αk it + hit (iit , k it )
which can be redefined as:
hit (iit , k it ) = ψ it − αk it
The second stage of the estimation considers the expectation of GVA and labour,
given the planned capital:
E [ y it − βl it | k it +1 ] = a + αk it +1 + E [ω it +1 | ω it ] ≡ αk it +1 + g (ω it )
This assumes that ωit +1 can be written as a function of ωit , letting ξ it represent the
difference, we can thus re-write the above equation as:
y it − βl it = αk it +1 + g (ω it ) + ξ it +1 + η it +1
then:
yit = αk it +1 + β lit + g (ψ it − αk it ) + ξ it +1 + η it +1
where g(.) is the third order polynomial of ψ it − αkit . Only in this stage do we realise
the consistent estimates for the capital coefficient α, and then the above equation is
estimated using non-linear least squares by SIC.
The table below illustrates the comparison between the OLS exogenous coefficient
estimates and the Olley-Pakes endogenous coefficients. As is expected, we see that
generally the coefficients decline for labour and increase for capital using the OlleyPakes methodology compared to the OLS estimates. The table below shows
coefficients when the capital stock is calculated using the standard ONS depreciation
rates, and a 1:1 tolerance ratio.
It should be noted that altering the tolerance ratios when calculating coefficients using
the Olley-Pakes methodology has a more noticeable impact on the consistency of the
estimates. As before, there is often a major change when moving from a “Calc_ratio”
of 0.33 to 0.66; however, increasing this ratio further does not always result in a
convergence of the estimates. The degree of volatility/consistency varies considerably
between SIC classifications, and this suggests that researchers must use more caution
when choosing the level of imputation in an Olley-Pakes framework, as this will have
a corresponding impact on MFP calculations. 8
8
∧
∧
Firm level MFP can be estimated using: mfp it = y it − β * l it − α * k it
18
Industry
SIC
code
11
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
40
41
45
50
51
52
55
60
61
62
63
64
65
66
67
70
71
72
73
74
Labour
Capital
β _O&P β_OLS α_O&P α_OLS
0.463
0.407
0.280
0.196
0.388
0.388
0.402
0.300
0.328
0.342
0.201
0.270
0.352
0.375
0.303
0.385
0.387
0.501
0.440
0.285
0.311
0.355
0.445
0.362
0.171
-0.005
0.325
0.432
0.471
0.397
0.436
0.516
0.637
0.509
0.506
0.687
0.508
0.566
0.479
0.580
0.553
0.336
0.346
0.424
0.477
0.431
0.307
0.349
0.359
0.297
0.338
0.386
0.407
0.351
0.454
0.429
0.587
0.509
0.336
0.366
0.412
0.518
0.415
0.209
0.396
0.473
0.493
0.602
0.502
0.555
0.550
0.702
0.615
0.602
0.484
0.640
0.730
0.606
0.666
0.016
0.984
0.131
0.991
1.455
0.693
0.473
0.622
0.577
0.645
0.651
0.542
0.620
0.584
0.866
0.481
0.412
0.442
0.460
0.605
0.535
0.516
0.578
0.494
0.819
0.716
0.434
0.417
0.546
0.547
0.453
0.237
0.199
0.461
0.183
1.056
0.143
0.425
0.299
0.320
0.599
0.666
0.461
0.444
0.493
0.494
0.513
0.548
0.533
0.558
0.499
0.538
0.455
0.363
0.424
0.314
0.353
0.547
0.528
0.425
0.381
0.502
0.574
0.355
0.562
0.370
0.247
0.310
0.371
0.424
0.223
0.336
0.384
0.325
0.220
0.208
0.161
19
References
Bloom N, Van Reenen J and Sadun R, (2005) “It Ain’t What You Do, it’s the Way
that You do IT”, Centre for Economic Performance, London School of Economics.
Criscuolo C and Martin R (2003) “Multinationals and US Productivity Leadership,
Evidence from Britain”, CeRiBA.
Farooqui S (2005) ‘IT use by Firms and Employees, Productivity Evidence across
Industries’, Economic Trends 625,
pp 65–74, available at: www.statistics.gov.uk/CCI/article.asp?ID=1233
Foster L, Haltiwanger J and Krizan C J (1998) “Aggregate Productivity Growth:
Lessons From Microeconomic Evidence”, Working Papers 98–12, Center for
Economic Studies, U.S. Census Bureau: Washington.
Fraumeni, B.M. (1997) “Measurement of Depreciation in the U.S. National Income
and Wealth Accounts”, Survey of Current Business, 74, 7-23.
Harris, R. (2005) “Deriving Measures of Plant-level Capital Stock in UK
Manufacturing”. Report to DTI.
Harris R, Li Q (2007) “Firm Level Empirical Study of the Contribution of Exporting
to UK Productivity Growth”. Report to UKTI.
Harris R and Li Q (2005) “Establishment Level Empirical Study of Links between
Exporting, Innovation and Productivity”, Report to UKTI.
Javorcik, B (2003) “Does Foreign Direct Investment Increase the Productivity of
Domestic Firms? In search of spillovers through backward linkages.” CEPR.
Martin, R (2003) “Building the capital stock”, CeRiBA
Olley S and Pakes A (1996) “The dynamics of productivity in the
Telecommunications Equipment Industry.” Econometrica, Vol 64, No. 6.
Wallis & Dey-Chowdhury (2007), “Volume of capital services: estimates for 1950 to
2006”. Economic & Labour Market Review, Vol 1, no. 12 . Available at:
http://www.statistics.gov.uk/cci/article.asp?ID=1905
OECD manual: Measuring Capital: measurement of capital stocks, consumption of
fixed capital and capital services (2001). Available
at:http://www.oecd.org/document/13/0,3343,en_2649_34253_2350157_1_1_1_1,00.h
tml
The ONS Productivity Handbook: a statistical overview and guide (2007). Available
at:http://www.statistics.gov.uk/StatBase/Product.asp?vlnk=14900&Pos=3&ColRank=
1&Rank=208
20
APPENDIX
I. LIST OF VARIABLES
Q Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
100
101
102
103
104
105
106
107
108
109
110
111
112
Employment
emp - total number of employees - point in time (pit)
emp - male full time employees -pit
emp - male part time* employees -pit
emp - feale full time employees - pit
emp - female part time employees - pit
emp - wrkng props not rcvng paymnt - pit
emp - other unpaid wrkers not vols - pit
emp - tot no of people in your emp - pit
emp - total employment** - pit
emp - total number of employees - avg
emp - male full time employees -avg
emp - female full time employees - avg
emp - male part time employees - avg
emp - female part time employees - avg
emp - year average total employment
gross wages/salaries/redundancy
soc security costs (ni codes/pensions)
tot emp costs ph (excludes unpaid)
soc sec costs % of tot emp costs
Gross wages and salaries (% is code 920)
Administrative, technical and clerical employees (% is code 202)
OP's and ATC's employeed in the bakehouse
Computer employees
Total employment (% is code 924)
Construction Table 4. Total employees ****
Construction Table 4. Total employees ****
Number of returned employment (% is code 926) **
Total employment (local units)
Gross wages and salaries (including outworkers remuneration)
Wages and salaries ATC's (% is code 930)
Wages and salaries OP's and ATC's in the bakehouse
Wages and salaries OP's (% is code 929)
Wages and salaries - jobbers
Wages per operatives (PH)
Turnover and Sales
sales turnover motors veh parts & acces
all other motor trades receipts
turnover from other sales of goods
amts rec from other countries fsr (exp)
amts rec from other countries fsr (imp)
turnover from service activs (car hire)
turnover from serv acts - sales of f&d
retail turnover of total turnover
sales of goods of own production
sales of electricity
sales of gas
val of vessels etc cmpltd for sale yor
water supply charges etc
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
sale of petroleum products
charge for commission refining
value of work done on customers maters
value of ind services provided by you
value of non-ind servs provided by you
sales of goods bought & resold no prcss
other sales of goods of own production
other work done on customers maters
value of new construction work
value of insurance claims received
value of grants donations
rent income earned on rented property
selling price of prop bought and sold
amount of vat included in q346
selling price of property devd by you
fees/commission earned by estate agents
additional income
value of any other operating income
value of sales of other goods & servs
commission and fees
cost of sales
sales of own account
sub postmaster sal/inc from po counters
fees from conveyancing
turnover arising from rental
turnover arising from sales
all other turnover
total retail turnover
total amount recvd sale of goods & servs
total non-retail turnover
% of turnover that is payments to you
as 349 - alternative - % entered
turnover from packaged software
trnovr from provis of prof comp servs
trnovr from tot provision of clients it
trnovr from provision of comp proc servs
trnovr from network and database servs
turnover from repair
turnover from direct sales
turnover from any other services
comments
captive turnover
sales to foreigner clients
sales of prosed, froz & preservd fish
retail - sales of demonstration cars
sales of demo cars to other mo traders
retail - sales of new cars
sales of other new vehicles
sales to other motor traders: new cars
sales to other motor traders: other vehs
gross sales of used vehicles and cycles
turnover from sales of petrol diesel etc
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
turnover from sales of other motor activ
retail sales food/drink not included
all other retail turnover not included
all other retail turnover not included
all other turnover not included
sales of other goods see notes 3.1 ab&c
total turnover
retail t/o
retail t/o % total t/o
production value
gva at market prices
gva at factor cost
gva at basic prices
total output at basic prices
gross operating surplus
gva % total output
gross operating rate
gva per unit turnover
labour prodctvty - gva ph bsc prcs
total output per head
unit wage costs
income received from sub-contracting
total fees and commissions receivable
total fees and commissions payable
all other turnover - inc rent exc other
other non retail turnover
amts rec from abroard fsr ind (ex)
amts rec from abroard fsr non-ind (ex)
amts paid to abroard fsr ind (im)
amts paid to abroard fsr non-ind (im)
Gross value added at market prices (excluding VAT)(derivn chgd 93)
Gross value added as a percentage of net output
Wages and salaries as a percentage of gross value added
Gross value added as a percentage of Total Sales and work done
Ratio of operatives to ATC's **
Sales of goods produced (derivn ch 93 equiv 261, used for top 10 prt) ***
Value of vessels and floating equipment completed for sale during the year
Work done on textiles on commission
Sales of waste products, work done and industrial services rendered
Work done & industrial services rendered
Amounts included for work done (augments 262 or 286)
Other services rendered
Work done on own textiles
Sales for waste products, residues, work done & industrial services rendered (drpd 91)
Rent & rates (excluding water rates) paid for industrial and commercial buildings (drpd 93)
Outworkers remuneration
Total sales and work done (% is code 931)(derivn ch 93)
Gross output (% is code 932)(derivn ch 93)
Net output (% is code 922)(derivn ch 93)
Sales of goods produced, work done and services rendered **
Total sales (derivn ch 93)
Gross value added excluding all taxes (derivn chgd 93)
217
218
219
220
Gross value added as a percentage of gross output
Sales of goods produced **
Sales of goods of own production (equiv 289 from 93)
Net output per head (PH)
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
Purchases
total taxes duties levies
purchases of energy and water products
purchases of goods and mats for own cons
purchases of gfr
value of industrial services purchased
payments for hiring /leasing/renting
commercial insurance premiums paid
purchase of road transport services
purchase of telecommunication servs
purchases of computer services
purchases of advertising services
purchase of other services
amounts paid in business rates
other amounts paid for taxes and levies
subsidies recieved from uk gov and ec
customs and excise duty payable
total amount of excise drawback
purchase of elec from others for resale
purchases of gas from others for resale
other amounts paid for taxes and levies
other purchases of goods & services
payment to sub-contractors
other goods for resale without process
purchases of materials (merchanting)
purchase of water from others
amounts payable to winning customers
purch of other goods bought for resale
purch of energy for own consumption
purchases of water
payments to employment agencies
vehicle excise duty paid
subsidies recvd under welfare to work
purch of servs for resale no processing
cost of waste disposal & sewerage
gross wages and salaries
redundancy and severance payments
employers ni contributions
contributions to pension funds
total employment costs
payments made under climate change levy
value of grants and donations paid
depreciation of capital assets
motor vehicles duties paid
purchase of used motor vehicles
purchase of parts for repair/servs
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
purchase of parts bought for resale
purchase of other goods bght for resale
total purchases of goods maters/servs
cost of acquisitions
purchases per unit
purchase of energy products for resale
Total plant and hire
Cost of non-industrial services recd, other than rent, insurance and bank charges (drpd 93 e
Cost of inward transport, by road, of materials and fuel purchased
Cost of inward transport, by other means, of materials and fuel purchased
Cost of outward transport, by road of goods sold
Cost of outward transport, by other means of goods sold
Amounts payable for repairs, maintenance & work given out
Amounts included for work given out (subcontracted) (augments 623 or 273)
Commercial insurance premiums paid
Bank charges
Inward transport of materials and fuel (drpd 92)
Outward transport of goods sold (drpd 92)
Meat inspection charges
1984 ONLY + 1989 ONLY
1984 ONLY + 1989 ONLY (646 in 1989)
Other services received (drpd 93)
1984 ONLY + 1989 ONLY (647 in 1989)
Postal, telecommunications and transport costs (drpd 93)
Postal costs
Amounts payable for hire of plant and machinery
Rent of industrial and commercial buildings
Cost of leasing computer equipment
Gas Oil - Cost of
Gas Oil - quantity (litres)
Liquefied petroleum gases - cost of
Capital expenditure - assets for leasing out on a finance lease *
Pollution abatement expenditure incurred
Pollution abatement - capital expenditure
Pollution abatement - current costs
Pollution control capex post-production
Pollution control capex - production process
Pollution control costs - direct
Pollution control costs - payments to others
Excise duty payable
Customs and Excise drawback
Special levies, allowances etc. payable - amounts repaid
Special levies, allowances etc. receivable - amounts received
All non-industrial services received
Cost of purchases (drpd from derivn 93)
Purchases including net levies(derivn ch 92) **
Cost of non-industrial services (derivn ch 93)**
Rent and hire of plant **
Purchases of raw materials and industrial services received (derivn ch 92)
Work done industrial services rendered, repairs etc. **
Work of a capital nature by own staff **
Work of a capital nature by own staff
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
Costs: Other Expenditure
Cost of other new and second-hand vehicles purchased
Proceeds from other vehicles disposed of
Plant, machinery, office equipment and other capital equipment purchased (calc'd by ratio fro
Cost of Nuclear Fuel (drpd 92)
Proceeds from sale of nuclear fuel (drpd 92)
Mains and services - acquisitions (drpd 92)
Cost of Nuclear Fuel (drpd 92)
Proceeds from sale of nuclear fuel (drpd 92)
Mains and services - acquisitions (drpd 92)
Other services received (Mining & Quarrying) (drpd 92)
Cost of outward transport by own staff (drpd 92)
Cost of computing services
Cost of materials and fuel used (drpd 92)
UNKNOWN
Cost of textiles for finishing and sale
Cost of materials and fuel purchased (equiv 145 from 93)
Cost of crude petroleum etc. purchased for blending (drpd91)
Total quantity of crude petroleum for blending on commission (tonnes)
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
Stocks, Capex and Investment
total stocks at beginning of year
total stocks at end of year
increase in total stocks
percentage increase in total stocks
stocks - material stores and fuel (by)
stocks of wip + goh (by)
stocks of msf (ey)
stocks of wip + goh (ey)
increase in stocks of msf
increase in stocks of wip and gih
const stocks - land (by) not dwellings
const stocks - land (ey) not dwellings
const stocks - msf (by) not dwellings
const stocks - msf (ey) not dwellings
const stocks - wip+goh (by) not dwell
const stocks - wip+goh (ey) not dwell
const stocks - total (by) not dwells
const stocks - total (ey) not dwells
const stocks - land (by) dwellings
const stocks - land (ey) dwellings
const stocks - msf (by) dwellings
const stocks - msf (ey) dwellings
const stocks - wip+goh (by) dwellings
const stocks - wip+goh (ey) dwellings
const stocks - total (by) dwellings
const stocks - total (ey) dwellings
total val of all stocks at bop inc wip
value of wip at bop
value of wip at eop
bys - stocks bought for resale
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
eys - stocks bought for resale
total value of all stocks at bop bys
total value of all stocks at eop bys
total value of all stocks end of yr
capex acq nyip new building work
capex acq nyip land & existing build
capex dis nyip land & existing build
capex acq nyip vehicles
capex dis nyip vehicles
capex acq nyip other
capex dis nyip other
cost of acquisitions - nyip
proceeds from disposals - nyip
nyip - total net capex
capex - total net (exc. nyips)
capex - acq - new building work
capex - acq - land and buildings
capex - acq - vehicles
capex - acq - other fixed capital
capex - dis - land and buildings
capex - dis - vehicles
capex - dis - other fixed capital
capex per unit
capex per head
acqui - new dwell and building work
acqui - other new building work
acqui - land and existing buildings
disp - land and existing buildings
acqui - vehicles
disp - vehicles
acquisitions - plant and machinery
disposals - plant and machinery
proceeds from disposals
capex acquisitions of land
capex acquisitions of existing buildings
capex proceeds from sale of land
capex proceeds from sale of buildings
new construction work
acquis of existing buildings not dwells
acquis of existing buildings
proceeds from disposal of exist buildgs
proceeds from disposal of exist dwells
new construction work related to dwells
Period covered by the return - from (STOCKS)
Period covered by the return - to (STOCKS)
Capital expenditure - date
Period covered by the return - from (CAPEX)
Period covered by the return - to (CAPEX)
Net NYIP net capex
Gross capex
Capex disposals
Increase during year - Goods on hand **
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
Increase during year - WIP - new vessels (drpd 92) **
Increase during year - other WIP (drpd 92)**
Total stocks (beginning of year) (not derived from 92)
Total stocks (end of year) (% is code 934)(not derived from 92)
Related to other than dwellings : Other stock B/Y (nb. used for UNKNOWN prior to 1980)
Related to other than dwellings : Other stock E/Y (nb. used for UNKNOWN prior to 1980)
Related to dwellings : Other stock B/Y (nb. used for UNKNOWN prior to 1980)
Related to dwellings : Other stock E/Y (nb. used for UNKNOWN prior to 1980)
Number of dwellings under construction B/Y (nb. used for UNKNOWN prior to 1980)
Number of dwellings under construction E/Y (nb. used for UNKNOWN prior to 1980)
Number of completed dwellings on hand for sale B/Y
Number of completed dwellings on hand for sale E/Y
Proceeds from vehicles disposed of (calcd by ratio from 92)
Cost of new building work - units NYIP
Cost of land and existing buildings purchased - units NYIP
Proceeds from land and buildings disposed of - units NYIP
Cost of new and second-hand equipment purchased - units NYIP
Proceeds from equipment disposed of - units NYIP
Computer equipment purchased (NYIP)
Cost of nuclear fuel purchased - units NYIP (drpd 92)
Proceeds from nuclear fuel disposed of - units NYIP (drpd 92)
Mains and services - acquisitions (laid or re-laid during the year) - units NYIP (drpd 92)
Mains and services - disposals and replacements - units NYIP (drpd 92)
Total capital expenditure acquisitions - units NYIP
Total Capex Disposals - NYIP
Amounts included in new building work for assets leased (drpd 92)
Amounts included in 504 for assets leased (drpd 92)
Amounts included in 517 for assets leased (drpd 92)
Amounts included in 531 for assets leased (drpd 92)
Amounts included in 534 for assets leased (drpd 92)
Amounts included in 537 for assets leased (drpd 92)
Net capex on land and existing buildings
Other net capex
Net capex (% is code 923)
Ratio of gross output to stocks **
Net capex as a % of GVA **
Backup Table 2. Includes net NYIP capex (not used) **
Increase during year - WIP - new vessels (410-409) **
Increase during year - other WIP (412 - 411) (drpd 92) **
Sales of goods produced (261 + 281) **
Work done & maintenance (262+286+282+283+284) ( 1993=282+283+284)**
No text found but exists on Current_Question1993
New Building Work (501 +531) (as 571) **
Aquisition of Land and Existing Buildings (502+532)**
Disposals of Land and Buildings (503-533)**
Aquisitions of Plant and Machinery(517+537)**
Disposals of Plant and machinery(518+538)**
Aquisitions of Vehicles(504+534)**
Disposals of Vehicles(505+535)**
Net capex on land and existing buildings (local units)
Other net capex (local units)
Net capex (local units)
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
Increase during year - materials, stores and fuel
Increase during year - work in progress
Increase during year - goods on hand for sale
Increase during year - work inprogress - new vessels
Increase during year - other work in progress
Work in progress - new vessels B/Y (drpd 92)
Work in progress - new vessels E/Y (drpd 92)
Other work in progress B/Y (drpd 92)
Other work in progress E/Y (drpd 92)
Goods on hand for sale B/Y (calc'd by ratio from 92)
Goods on hand for sale E/Y (calc'd by ratio from 92)
Cost of land and existing buildings purchased - units NYIP
Proceeds from land and buildings disposed of - units NYIP
Cost of other new and secon-hand vehicles purchased - units NYIP (Code not shown on form
Mains and services - disposals and replacements - units NYIP (drpd 92)
Disposals of vehicles **
Advertising and publicity (drpd 93)
Net capex (% is code 923)
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
Retail
sales of fresh fish & shellfish
fresh fruit and veg
meat
fish
bakery products and cereals
alcoholic drinks
soft drinks
tobacco
milk cheese eggs (inc yogs & cream)
canned
other food (tea, coffee)
pharmaceuticals
national health receipts
other medical prods & therapeutics
if mainly hearing aids in 713 38018
other applincs etc for personal care
articles of clothing - making clothes
household textiles
mens wear
ladies wear
adult clothing
boys girls infant wear
footwear
leather
domestic gas
soft furnishing
other household goods
furniture
domestic electrical
audio visual equipment
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
audio and video
home computers
decorators
wallpaper
lawnmowers
books
stationery
floor coverings
photographic, cinematic and optical
office and telecom
jewellery clocks and watches
souvenirs
prints and pictures
sports
antiques
other goods not elsewhere classified
retail sales from shops
retail sales from mail orders
retail sales from stalls and pitch
rs - direct selling (home(rounds))
rs - direct sell (home(no rounds))
retail sales by other means
meat fresh chilled dried froz cannd pro
meat fresh chilled dried froz cannd pro
oils and fats (inc butter and marg)
fruit fresh chilld dried froz cannd pro
veg fresh chilled dried froz cannd pro
sugar jam honey choc & confection
food not elsewhere classified
non-alcoholic beverages
tobacco (exc pipes lighters etc)
garments
decorating and diy supplies
furniture and furnishings
carpets and other floor coverings
works of art and antiques
household textiles
household and personal appliances
glassware tableware and utensils
tools and equipment for house & garden
non-durable household goods
travel goods and other pers effects
telephone and telefax equipment
information processing equipment
recordng material for pictures & sound
equip for sport camping and recreation
spare parts for all vehicles and bikes
games toys and hobbies
natural or artificial plants & flowers
pets and related products
books
newspapers and periodicals
782 stationery and drawing materials
783 Operation of retail shops by bakers
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
Misc
research and development
period covered - from date
period covered - to date
tick box if overseas receipts/payments
time taken to complete form - hrs
time taken to complete form - mins
remarks
internet/edi used to place orders
internet/edi used to receive orders
have a website accessible by any user
total taxes duties levies
intermediate consumption
net excise duty
total net taxes
amount in wq600 of assets leased
amount in q600 cap work by own staff
gross invsts in conses, licenses etc
number of shops
number of mail order outlets
number of market stalls and pitches
number of establishments
number of letting bedrooms
number of bedplaces
computer software dev by own staff
film admissions (actual number)
gross box office takings
amount paid for hire of films
tick box (y/n)(for 772-774)
total number of cinema sites
total number of screens
purchas of goods bought for resale wfp
investmnt in purchasd computer software
Showroom activity (nb. usedfor 'New LA code' in 1975
Enterprise number
Activity heading (SIC 80)
NACE Number
Directory of manufacturing business marker * ('nb. used for 'Confidence Code' prior to 1986)
Generating stations
Working proprietors (for CCL)
Operatives (% is code 928)
Jobbers - printers only (drpd 90)
Purchases and net taxes(734+830+832-831-833)(derivn ch 92)**
Net excise duty(830-831)**
Net taxes(830+832-831-833)**
Net levies(823-833)**
Total tax payable(830-832)**
Total tax rebate & allowances(831-833)**
Net land & buildings(571+572-573)**
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
Net plant (577-578)**
Net vehicles(574-575) **
Net nuclear fuel (drpd 92)**
Net mains & services (drpd 92) **
Levies payable - net
Excise payments - net
Work done on textiles on commission (drpd 91)
Other sales of goods for own production, work done & industrial services rendered
Commission refining, work done & services rendered **
Road vehicle licences
Rates
Other work done & sales of own production (drpd 93)
Work done as main or direct contractor on GB sites (drpd 93)
Work done as sub-contractor on GB sites (drpd 93)
Work done in GB on contracts for erections of process plants overseas (drpd93)
Water supplied
Repairs and maintenance and industrial services rendered
Sales of other goods of own production
Work done and industrial services rendered (drpd 91)
Sales of goods produced(drpd 93 equiv 261)** (nb. 'Work done for related businesses overs
Work done & maintenance (derivn ch 93) **
Sales, work done & construction work **
Sales of film processing, work done and industrial services rendered (drpd 93)
Sales of waste products and residues (drpd 93)
Amounts payable for work given out (drpd 93)
Amounts payable for repairs and maintenance (drpd 93)
Rent & rates (excluding water rates) paid for industrial and commercial buildings (drpd 93)
ARD2
19971
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
100
101
pre-1996
49
49
49
49
49
49
49
49
845
845
845
845
845
845
845
140
140
312
312
140
202
203
207
845
848
849
900
945
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
110
102
120
102
50
51
52
53
54
55
58
59
224
234
235
236
237
238
239
250
251
253
254
140
202
203
207
845
848
849
900
945
250a
301a
302a
304a
305a
310a
Pre 97 name
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
297
289
302
304
306
#N/A
#N/A
emp - total number of employees - point in time (pit)
emp - male full time employees -pit
emp - male part time* employees -pit
emp - feale full time employees - pit
emp - female part time employees - pit
emp - wrkng props not rcvng paymnt - pit
emp - other unpaid wrkers not vols - pit
emp - tot no of people in your emp - pit
emp - total employment** - pit
emp - total number of employees - avg
emp - male full time employees -avg
emp - female full time employees - avg
emp - male part time employees - avg
emp - female part time employees - avg
emp - year average total employment
gross wages/salaries/redundancy
soc security costs (ni codes/pensions)
tot emp costs ph (excludes unpaid)
soc sec costs % of tot emp costs
Gross wages and salaries (% is code 920)
Administrative, technical and clerical employees (% is code 202)
OP's and ATC's employeed in the bakehouse
Computer employees
Total employment (% is code 924)
Construction Table 4. Total employees ****
Construction Table 4. Total employees ****
Number of returned employment (% is code 926) **
Total employment (local units)
Gross wages and salaries (including outworkers remuneration)
Wages and salaries ATC's (% is code 930)
Wages and salaries OP's and ATC's in the bakehouse
Wages and salaries OP's (% is code 929)
Wages and salaries - jobbers
Wages per operatives (PH)
sales turnover motors veh parts & acces
all other motor trades receipts
2
3
5
6
8
9
14
15
17
18
19
20
21
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
69
70
2
3
5
6
8
211
14
15
17
18
19
20
21
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
69
70
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
158
163
164
169
179
300
301
302
303
304
305
306
307
308
309
310
311
312
315
316
317
318
319
320
321
322
323
324
325
327
337
338
339
340
341
342
343
157
160
160
160
177
299
248
256
256
256
256
271
272
272
272
272
266
259
314
260
260
260
260
260
260
260
260
260
260
260
260
130
130
130
130
130
130
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
turnover from other sales of goods
amts rec from other countries fsr (exp)
amts rec from other countries fsr (imp)
turnover from service activs (car hire)
turnover from serv acts - sales of f&d
retail turnover of total turnover
sales of goods of own production
sales of electricity
sales of gas
val of vessels etc cmpltd for sale yor
water supply charges etc
sale of petroleum products
charge for commission refining
value of work done on customers maters
value of ind services provided by you
value of non-ind servs provided by you
sales of goods bought & resold no prcss
other sales of goods of own production
other work done on customers maters
value of new construction work
value of insurance claims received
value of grants donations
rent income earned on rented property
selling price of prop bought and sold
amount of vat included in q346
selling price of property devd by you
fees/commission earned by estate agents
additional income
value of any other operating income
value of sales of other goods & servs
commission and fees
cost of sales
sales of own account
sub postmaster sal/inc from po counters
fees from conveyancing
turnover arising from rental
turnover arising from sales
71
72
73
74
75
76
77
101
102
139
140
144
145
148
150
154
155
157
159
160
176
177
181
183
184
190
190
201
202
203
204
205
206
207
208
224
248
71
72
73
74
75
76
77
101
102
139
140
22
23
148
150
154
155
157
159
160
176
177
181
183
184
190
630
201
202
203
204
205
206
207
208
845
248
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
344
345
346
347
349
350
351
352
353
354
355
356
357
358
359
360
361
370
371
372
373
374
375
376
377
378
379
380
381
382
383
398
399
551
552
610
611
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
156
550
550
597
148
all other turnover
total retail turnover
total amount recvd sale of goods & servs
total non-retail turnover
% of turnover that is payments to you
as 349 - alternative - % entered
turnover from packaged software
trnovr from provis of prof comp servs
trnovr from tot provision of clients it
trnovr from provision of comp proc servs
trnovr from network and database servs
turnover from repair
turnover from direct sales
turnover from any other services
comments
captive turnover
sales to foreigner clients
sales of prosed, froz & preservd fish
retail - sales of demonstration cars
sales of demo cars to other mo traders
retail - sales of new cars
sales of other new vehicles
sales to other motor traders: new cars
sales to other motor traders: other vehs
gross sales of used vehicles and cycles
turnover from sales of petrol diesel etc
turnover from sales of other motor activ
retail sales food/drink not included
all other retail turnover not included
all other retail turnover not included
all other turnover not included
sales of other goods see notes 3.1 ab&c
total turnover
retail t/o
retail t/o % total t/o
production value
gva at market prices
250
253
255
257
258
262
264
267
268
269
270
270
271
271
272
274
276
278
279
280
281
282
283
286
287
288
289
290
291
294
295
299
301
302
306
307
311
140
312
255
257
258
262
264
267
268
269
176
450
177
451
152
401
402
153
151
452
453
454
455
178
179
458
459
460
461
185
186
299
248
256
271
272
266
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
612
613
614
615
630
631
632
633
634
635
679
777
778
779
787
1033
1034
1035
1036
148
181
183
184
208
248
255
257
258
262
264
267
268
269
299
314
840
841
837
837
837
837
181
181
181
376
374
374
657
748
748
748
733
996
996
996
996
148
181
183
184
208
248
255
257
258
262
264
267
268
269
299
314
840
841
gva at factor cost
gva at basic prices
total output at basic prices
gross operating surplus
gva % total output
gross operating rate
gva per unit turnover
labour prodctvty - gva ph bsc prcs
total output per head
unit wage costs
income received from sub-contracting
total fees and commissions receivable
total fees and commissions payable
all other turnover - inc rent exc other
other non retail turnover
amts rec from abroard fsr ind (ex)
amts rec from abroard fsr non-ind (ex)
amts paid to abroard fsr ind (im)
amts paid to abroard fsr non-ind (im)
Gross value added at market prices (excluding VAT)(derivn chgd 93)
Gross value added as a percentage of net output
Wages and salaries as a percentage of gross value added
Gross value added as a percentage of Total Sales and work done
Ratio of operatives to ATC's **
Sales of goods produced (derivn ch 93 equiv 261, used for top 10 prt
Value of vessels and floating equipment completed for sale during th
Work done on textiles on commission
Sales of waste products, work done and industrial services rendered
Work done & industrial services rendered
Amounts included for work done (augments 262 or 286)
Other services rendered
Work done on own textiles
Sales for waste products, residues, work done & industrial services re
Rent & rates (excluding water rates) paid for industrial and commerci
Outworkers remuneration
Total sales and work done (% is code 931)(derivn ch 93)
Gross output (% is code 932)(derivn ch 93)
312
314
316
338
399
408
409
423
448
450
456
457
462
463
464
465
466
467
501
501
502
502
505
511
513
514
515
517
518
521
521
522
524
526
529
531
532
259
314
260
130
156
647
511
145
315
142
456
457
462
463
464
465
466
467
197
403
196
404
505
137
534
535
138
528
529
139
539
519
501
504
505
531
532
213
214
215
216
217
218
219
220
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
842
891
399a
611a
630a
301a
301b
634a
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
842
891
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
156
156
156
156
156
156
156
156
647
511
511
511
511
511
511
511
511
511
511
511
511
511
511
145
145
145
145
145
145
Net output (% is code 922)(derivn ch 93)
Sales of goods produced, work done and services rendered **
Total sales (derivn ch 93)
Gross value added excluding all taxes (derivn chgd 93)
Gross value added as a percentage of gross output
Sales of goods produced **
Sales of goods of own production (equiv 289 from 93)
Net output per head (PH)
total taxes duties levies
purchases of energy and water products
purchases of goods and mats for own cons
purchases of gfr
value of industrial services purchased
payments for hiring /leasing/renting
commercial insurance premiums paid
purchase of road transport services
purchase of telecommunication servs
purchases of computer services
purchases of advertising services
purchase of other services
amounts paid in business rates
other amounts paid for taxes and levies
subsidies recieved from uk gov and ec
customs and excise duty payable
total amount of excise drawback
purchase of elec from others for resale
purchases of gas from others for resale
other amounts paid for taxes and levies
other purchases of goods & services
payment to sub-contractors
other goods for resale without process
purchases of materials (merchanting)
purchase of water from others
amounts payable to winning customers
purch of other goods bought for resale
purch of energy for own consumption
purchases of water
533
537
538
541
547
548
549
550
558
559
561
564
567
570
591
594
597
611
612
619
620
621
622
623
624
625
626
627
628
630
633
634
637
640
641
642
643
533
537
538
541
547
548
549
550
558
559
561
564
567
147
591
594
597
148
837
619
620
621
622
623
624
625
626
627
628
181
376
374
93
640
641
642
643
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
430
431
432
433
435
446
447
448
449
450
455
475
476
477
478
479
480
483
499
600
639
767
150
190
619
620
621
622
623
624
625
626
627
628
640
641
642
145
145
145
145
145
145
145
315
315
142
142
467
467
467
467
467
467
467
467
597
93
748
150
630
619
620
621
622
623
624
625
626
627
628
640
641
642
payments to employment agencies
vehicle excise duty paid
subsidies recvd under welfare to work
purch of servs for resale no processing
cost of waste disposal & sewerage
gross wages and salaries
redundancy and severance payments
employers ni contributions
contributions to pension funds
total employment costs
payments made under climate change levy
value of grants and donations paid
depreciation of capital assets
motor vehicles duties paid
purchase of used motor vehicles
purchase of parts for repair/servs
purchase of parts bought for resale
purchase of other goods bght for resale
total purchases of goods maters/servs
cost of acquisitions
purchases per unit
purchase of energy products for resale
Total plant and hire
Cost of non-industrial services recd, other than rent, insurance and b
Cost of inward transport, by road, of materials and fuel purchased
Cost of inward transport, by other means, of materials and fuel purch
Cost of outward transport, by road of goods sold
Cost of outward transport, by other means of goods sold
Amounts payable for repairs, maintenance & work given out
Amounts included for work given out (subcontracted) (augments 623
Commercial insurance premiums paid
Bank charges
Inward transport of materials and fuel (drpd 92)
Outward transport of goods sold (drpd 92)
Meat inspection charges
1984 ONLY + 1989 ONLY
1984 ONLY + 1989 ONLY (646 in 1989)
644
645
646
655
656
657
680
682
683
684
685
686
687
746
747
748
781
794
796
797
802
808
809
810
811
812
813
814
816
817
818
819
820
821
822
826
830
644
645
646
655
656
657
531
502
503
513
514
577
578
746
747
748
733
571
572
573
802
808
809
810
811
812
813
814
816
817
818
819
820
821
822
826
830
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
643
644
645
646
655
656
657
746
747
748
802
808
809
810
811
812
813
814
830
831
832
833
838
839
892
894
895
423a
250a
251a
251b
360a
526a
529a
517a
521a
522a
643
644
645
646
655
656
657
746
747
748
802
808
809
810
811
812
813
814
830
831
832
833
838
839
892
894
895
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Other services received (drpd 93)
1984 ONLY + 1989 ONLY (647 in 1989)
Postal, telecommunications and transport costs (drpd 93)
Postal costs
Amounts payable for hire of plant and machinery
Rent of industrial and commercial buildings
Cost of leasing computer equipment
Gas Oil - Cost of
Gas Oil - quantity (litres)
Liquefied petroleum gases - cost of
Capital expenditure - assets for leasing out on a finance lease *
Pollution abatement expenditure incurred
Pollution abatement - capital expenditure
Pollution abatement - current costs
Pollution control capex post-production
Pollution control capex - production process
Pollution control costs - direct
Pollution control costs - payments to others
Excise duty payable
Customs and Excise drawback
Special levies, allowances etc. payable - amounts repaid
Special levies, allowances etc. receivable - amounts received
All non-industrial services received
Cost of purchases (drpd from derivn 93)
Purchases including net levies(derivn ch 92) **
Cost of non-industrial services (derivn ch 93)**
Rent and hire of plant **
Purchases of raw materials and industrial services received (derivn c
Work done industrial services rendered, repairs etc. **
Work of a capital nature by own staff **
Work of a capital nature by own staff
Costs: Other Expenditure
Cost of other new and second-hand vehicles purchased
Proceeds from other vehicles disposed of
Plant, machinery, office equipment and other capital equipment purch
Cost of Nuclear Fuel (drpd 92)
Proceeds from sale of nuclear fuel (drpd 92)
831
832
833
835
836
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
831
832
833
835
836
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
403
404
405
406
407
408
409
410
411
412
413
414
415
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
523a
521a
522a
523a
631a
632a
409a
731a
731b
732a
731c
735a
736a
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
450
451
152
152
401
401
402
402
153
151
452
453
454
455
455
455
178
179
458
459
460
461
461
461
Mains and services - acquisitions (drpd 92)
Cost of Nuclear Fuel (drpd 92)
Proceeds from sale of nuclear fuel (drpd 92)
Mains and services - acquisitions (drpd 92)
Other services received (Mining & Quarrying) (drpd 92)
Cost of outward transport by own staff (drpd 92)
Cost of computing services
Cost of materials and fuel used (drpd 92)
UNKNOWN
Cost of textiles for finishing and sale
Cost of materials and fuel purchased (equiv 145 from 93)
Cost of crude petroleum etc. purchased for blending (drpd91)
Total quantity of crude petroleum for blending on commission (tonnes
total stocks at beginning of year
total stocks at end of year
increase in total stocks
percentage increase in total stocks
stocks - material stores and fuel (by)
stocks of wip + goh (by)
stocks of msf (ey)
stocks of wip + goh (ey)
increase in stocks of msf
increase in stocks of wip and gih
const stocks - land (by) not dwellings
const stocks - land (ey) not dwellings
const stocks - msf (by) not dwellings
const stocks - msf (ey) not dwellings
const stocks - wip+goh (by) not dwell
const stocks - wip+goh (ey) not dwell
const stocks - total (by) not dwells
const stocks - total (ey) not dwells
const stocks - land (by) dwellings
const stocks - land (ey) dwellings
const stocks - msf (by) dwellings
const stocks - msf (ey) dwellings
const stocks - wip+goh (by) dwellings
const stocks - wip+goh (ey) dwellings
870
871
872
873
891
892
894
895
900
901
903
904
905
906
908
909
911
916
917
918
935
936
944
945
950
951
952
953
954
955
956
961
962
963
964
965
966
870
871
872
873
891
892
894
895
900
901
903
904
905
906
908
909
911
916
917
918
935
936
944
945
950
951
952
953
954
955
956
961
962
963
964
965
966
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
294
295
500
501
502
503
504
507
508
599
510
511
512
513
514
515
516
517
518
521
522
524
525
526
527
528
529
530
636
637
680
681
682
683
684
685
686
185
186
467
403
404
404
404
505
505
597
505
137
137
534
535
138
138
528
529
539
519
501
501
504
504
504
505
505
374
93
531
531
502
503
513
514
577
const stocks - total (by) dwellings
const stocks - total (ey) dwellings
total val of all stocks at bop inc wip
value of wip at bop
value of wip at eop
bys - stocks bought for resale
eys - stocks bought for resale
total value of all stocks at bop bys
total value of all stocks at eop bys
total value of all stocks end of yr
capex acq nyip new building work
capex acq nyip land & existing build
capex dis nyip land & existing build
capex acq nyip vehicles
capex dis nyip vehicles
capex acq nyip other
capex dis nyip other
cost of acquisitions - nyip
proceeds from disposals - nyip
nyip - total net capex
capex - total net (exc. nyips)
capex - acq - new building work
capex - acq - land and buildings
capex - acq - vehicles
capex - acq - other fixed capital
capex - dis - land and buildings
capex - dis - vehicles
capex - dis - other fixed capital
capex per unit
capex per head
acqui - new dwell and building work
acqui - other new building work
acqui - land and existing buildings
disp - land and existing buildings
acqui - vehicles
disp - vehicles
acquisitions - plant and machinery
967
971
972
973
974
975
976
981
982
983
984
985
986
987
990
991
992
993
994
995
996
250a
250a
251a
251b
270a
274a
276a
277b
282a
283a
284a
285a
286a
288a
290a
291a
967
971
972
973
974
975
976
981
982
983
984
985
986
987
990
991
992
993
994
995
996
141
250
251
252
270
274
276
277
282
283
284
285
286
288
290
291
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
687
699
763
764
765
766
794
795
796
797
798
799
14
15
17
18
19
139
154
155
157
159
160
176
177
456
457
462
463
464
465
466
467
505
531
532
533
578
578
748
748
748
748
571
571
572
573
573
573
14
15
17
18
19
139
154
155
157
159
160
176
177
456
457
462
463
464
465
466
467
505
531
532
533
disposals - plant and machinery
293a
proceeds from disposals
294a
capex acquisitions of land
295a
capex acquisitions of existing buildings
296a
capex proceeds from sale of land
297a
capex proceeds from sale of buildings
301a
new construction work
301a
acquis of existing buildings not dwells
301b
acquis of existing buildings
301c
proceeds from disposal of exist buildgs
302a
proceeds from disposal of exist dwells
304a
new construction work related to dwells
305a
Period covered by the return - from (STOCKS)
306a
Period covered by the return - to (STOCKS)
310a
Capital expenditure - date
311a
Period covered by the return - from (CAPEX)
312a
Period covered by the return - to (CAPEX)
312b
Net NYIP net capex
321c
Gross capex
360a
Capex disposals
399a
Increase during year - Goods on hand **
400a
Increase during year - WIP - new vessels (drpd 92) **
409a
Increase during year - other WIP (drpd 92)**
409a
Total stocks (beginning of year) (not derived from 92)
410a
Total stocks (end of year) (% is code 934)(not derived from 92)
411a
Related to other than dwellings : Other stock B/Y (nb. used for UNKN412a
Related to other than dwellings : Other stock E/Y (nb. used for UNKN413a
Related to dwellings : Other stock B/Y (nb. used for UNKNOWN prior414a
Related to dwellings : Other stock E/Y (nb. used for UNKNOWN prior423a
Number of dwellings under construction B/Y (nb. used for UNKNOWN513a
Number of dwellings under construction E/Y (nb. used for UNKNOWN514a
Number of completed dwellings on hand for sale B/Y
517a
Number of completed dwellings on hand for sale E/Y
521a
Proceeds from vehicles disposed of (calcd by ratio from 92)
522a
Cost of new building work - units NYIP
522a
Cost of land and existing buildings purchased - units NYIP
523a
Proceeds from land and buildings disposed of - units NYIP
524a
293
294
295
296
297
249
301
261
289
302
304
305
306
310
311
273
281
287
360
144
400
409
635
410
411
412
413
414
146
545
546
517
521
522
844
523
524
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
537
538
541
547
548
549
550
558
559
561
564
567
591
594
597
835
836
844
846
847
861
862
863
864
865
866
867
868
869
870
871
872
873
935
936
944
952
537
538
541
547
548
549
550
558
559
561
564
567
591
594
597
835
836
844
846
847
861
862
863
864
865
866
867
868
869
870
871
872
873
935
936
944
952
Cost of new and second-hand equipment purchased - units NYIP
526a
Proceeds from equipment disposed of - units NYIP
529a
Computer equipment purchased (NYIP)
550a
Cost of nuclear fuel purchased - units NYIP (drpd 92)
581a
Proceeds from nuclear fuel disposed of - units NYIP (drpd 92)
611a
Mains and services - acquisitions (laid or re-laid during the year) - un 611a
Mains and services - disposals and replacements - units NYIP (drpd 630a
Total capital expenditure acquisitions - units NYIP
631a
Total Capex Disposals - NYIP
632a
Amounts included in new building work for assets leased (drpd 92) 634a
Amounts included in 504 for assets leased (drpd 92)
682a
Amounts included in 517 for assets leased (drpd 92)
683a
Amounts included in 531 for assets leased (drpd 92)
684a
Amounts included in 534 for assets leased (drpd 92)
685a
Amounts included in 537 for assets leased (drpd 92)
731a
Net capex on land and existing buildings
731b
Other net capex
731c
Net capex (% is code 923)
732a
Ratio of gross output to stocks **
735a
Net capex as a % of GVA **
736a
Backup Table 2. Includes net NYIP capex (not used) **
740a
Increase during year - WIP - new vessels (410-409) **
741a
Increase during year - other WIP (412 - 411) (drpd 92) **
742a
Sales of goods produced (261 + 281) **
743a
Work done & maintenance (262+286+282+283+284) ( 1993=282+28744a
No text found but exists on Current_Question1993
745a
New Building Work (501 +531) (as 571) **
746a
Aquisition of Land and Existing Buildings (502+532)**
747a
Disposals of Land and Buildings (503-533)**
748a
Aquisitions of Plant and Machinery(517+537)**
749a
Disposals of Plant and machinery(518+538)**
750a
Aquisitions of Vehicles(504+534)**
751a
Disposals of Vehicles(505+535)**
752a
Net capex on land and existing buildings (local units)
753a
Other net capex (local units)
754a
Net capex (local units)
entref
Increase during year - materials, stores and fuel
sic68
515
516
550
581
149
611
182
631
632
375
532
533
574
575
702
721
734
732
735
736
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
1
4
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
953
954
955
956
409a
410a
411a
412a
413a
414a
682a
683a
513a
550a
685a
611a
522a
336
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
953
954
955
956
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
260
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
Increase during year - work in progress
ultfoc
7
Increase during year - goods on hand for sale
Increase during year - work inprogress - new vessels
Increase during year - other work in progress
Work in progress - new vessels B/Y (drpd 92)
Work in progress - new vessels E/Y (drpd 92)
Other work in progress B/Y (drpd 92)
Other work in progress E/Y (drpd 92)
Goods on hand for sale B/Y (calc'd by ratio from 92)
Goods on hand for sale E/Y (calc'd by ratio from 92)
Cost of land and existing buildings purchased - units NYIP
Proceeds from land and buildings disposed of - units NYIP
Cost of other new and secon-hand vehicles purchased - units NYIP (Code not shown on form)
Mains and services - disposals and replacements - units NYIP (drpd 92)
Disposals of vehicles **
Advertising and publicity (drpd 93)
Net capex (% is code 923)
sales of fresh fish & shellfish
fresh fruit and veg
meat
fish
bakery products and cereals
alcoholic drinks
soft drinks
tobacco
milk cheese eggs (inc yogs & cream)
canned
other food (tea, coffee)
pharmaceuticals
national health receipts
other medical prods & therapeutics
if mainly hearing aids in 713 38018
other applincs etc for personal care
articles of clothing - making clothes
household textiles
mens wear
ladies wear
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
753
754
755
756
757
758
1363
1364
1365
1366
1367
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
578
748
748
748
748
748
748
996
996
996
996
996
adult clothing
boys girls infant wear
footwear
leather
domestic gas
soft furnishing
other household goods
furniture
domestic electrical
audio visual equipment
audio and video
home computers
decorators
wallpaper
lawnmowers
books
stationery
floor coverings
photographic, cinematic and optical
office and telecom
jewellery clocks and watches
souvenirs
prints and pictures
sports
antiques
other goods not elsewhere classified
retail sales from shops
retail sales from mail orders
retail sales from stalls and pitch
rs - direct selling (home(rounds))
rs - direct sell (home(no rounds))
retail sales by other means
meat fresh chilled dried froz cannd pro
meat fresh chilled dried froz cannd pro
oils and fats (inc butter and marg)
fruit fresh chilld dried froz cannd pro
veg fresh chilled dried froz cannd pro
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
800
801
802
803
804
805
806
807
808
809
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
20
9
11
12
143
144
145
146
166
167
168
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
996
20
211
211
211
140
22
23
23
160
160
160
sugar jam honey choc & confection
food not elsewhere classified
non-alcoholic beverages
tobacco (exc pipes lighters etc)
garments
decorating and diy supplies
furniture and furnishings
carpets and other floor coverings
works of art and antiques
household textiles
household and personal appliances
glassware tableware and utensils
tools and equipment for house & garden
non-durable household goods
travel goods and other pers effects
telephone and telefax equipment
information processing equipment
recordng material for pictures & sound
equip for sport camping and recreation
spare parts for all vehicles and bikes
games toys and hobbies
natural or artificial plants & flowers
pets and related products
books
newspapers and periodicals
stationery and drawing materials
Operation of retail shops by bakers
research and development
period covered - from date
period covered - to date
tick box if overseas receipts/payments
time taken to complete form - hrs
time taken to complete form - mins
remarks
internet/edi used to place orders
internet/edi used to receive orders
have a website accessible by any user
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
400
570
580
581
601
602
653
750
751
752
768
769
770
771
772
773
774
775
776
780
781
788
2
3
5
6
8
21
201
205
206
850
851
852
853
854
855
156
147
147
147
597
597
646
748
748
748
748
748
748
748
748
748
748
748
748
748
733
733
2
3
5
6
8
21
201
205
206
850
851
852
853
854
855
total taxes duties levies
intermediate consumption
net excise duty
total net taxes
amount in wq600 of assets leased
amount in q600 cap work by own staff
gross invsts in conses, licenses etc
number of shops
number of mail order outlets
number of market stalls and pitches
number of establishments
number of letting bedrooms
number of bedplaces
computer software dev by own staff
film admissions (actual number)
gross box office takings
amount paid for hire of films
tick box (y/n)(for 772-774)
total number of cinema sites
total number of screens
purchas of goods bought for resale wfp
investmnt in purchasd computer software
Showroom activity (nb. usedfor 'New LA code' in 1975
Enterprise number
Activity heading (SIC 80)
NACE Number
Directory of manufacturing business marker * ('nb. used for 'Confidence Code' prior to 1986)*
Generating stations
Working proprietors (for CCL)
Operatives (% is code 928)
Jobbers - printers only (drpd 90)
Purchases and net taxes(734+830+832-831-833)(derivn ch 92)**
Net excise duty(830-831)**
Net taxes(830+832-831-833)**
Net levies(823-833)**
Total tax payable(830-832)**
Total tax rebate & allowances(831-833)**
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
856
857
858
859
860
950
951
270a
312a
274a
276a
277b
312b
282a
283a
284a
285a
286a
321c
288a
301c
290a
291a
293a
294a
295a
296a
297a
684a
306a
856
857
858
859
860
950
951
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Net land & buildings(571+572-573)**
Net plant (577-578)**
Net vehicles(574-575) **
Net nuclear fuel (drpd 92)**
Net mains & services (drpd 92) **
Levies payable - net
Excise payments - net
Work done on textiles on commission (drpd 91)
Other sales of goods for own production, work done & industrial services rendered
Commission refining, work done & services rendered **
Road vehicle licences
Rates
Other work done & sales of own production (drpd 93)
Work done as main or direct contractor on GB sites (drpd 93)
Work done as sub-contractor on GB sites (drpd 93)
Work done in GB on contracts for erections of process plants overseas (drpd93)
Water supplied
Repairs and maintenance and industrial services rendered
Sales of other goods of own production
Work done and industrial services rendered (drpd 91)
Sales of goods produced(drpd 93 equiv 261)** (nb. 'Work done for related businesses overseas' i
Work done & maintenance (derivn ch 93) **
Sales, work done & construction work **
Sales of film processing, work done and industrial services rendered (drpd 93)
Sales of waste products and residues (drpd 93)
Amounts payable for work given out (drpd 93)
Amounts payable for repairs and maintenance (drpd 93)
Rent & rates (excluding water rates) paid for industrial and commercial buildings (drpd 93)
UNKNOWN
UNKNOWN
Summary quality report for Annual Business Inquiry
part 2 (ABI/2)
Introduction
This report is part of a rolling programme of quality reports being introduced
by the Office for National Statistics (ONS). The full programme of work being
carried out on Statistical Quality1 is available on the National Statistics
website.
This report relates to the Annual Business Inquiry (ABI). The Annual
Business Inquiry collects information via 2 surveys. The ABI/1 solely
concentrates on business employment details and the ABI/2 collects
information relating to the financial information of a business. This report
specifically concentrates on the ABI/2, and aims to provide users with
guidance to assess the quality and usability of the ABI/2 estimates.
The ABI/22 has a comprehensive online guide to the methods and terminology
used in compiling business and economic statistics.
Summary of quality
1 Relevance
The degree to which the statistical product meets user needs for both coverage and content.
The Annual Business Inquiry Part 2 (ABI/2) collects comprehensive financial
information from various businesses representing the majority of the UK
economy. The ONS ABI/2 survey collects business information for England,
Scotland and Wales. The Department of Enterprise Trade and Investment
(DETINI)3 of Northern Ireland collect the same ABI information (NIABI)4
independently. Both data sources are then combined to produce ABI/2
estimates on a UK basis.
In terms of business turnover the ABI/2 sample covers approximately two
thirds of the whole economy including industry areas such as Production,
Construction, Service Trades and Distribution sectors. More detail on ABI/2
industry coverage can be found under the Summary of methods section at the
end of the report or at http://www.statistics.gov.uk/abi/sic92_desc.asp The ABI/2 data and
estimates are used widely, both within and outside the government and are a
vital source of business financial information.
What it measures
Frequency
Sample Size
Periods available
Sample frame
Sample design
Created on 21/07/06 – Version 1
ABI/2
Business and financial information including Total
Turnover, Total Employment costs, Total Purchases,
Total Taxes and many other aggregates.
Annual
Approximately 74,000 (GB)
From 1997
Inter Departmental Business Register (IDBR)
Stratified random sample where the strata are defined
1
Weighting
Estimation
Imputation
Outliers
by Standard Industrial Classification (SIC), Country and
employment size of a business.
Each business represents a number of similar
businesses, based on number of employees and the
standard industrial classification (currently SIC 2003).
Weights are updated annually.
Ratio estimation is carried out for all non responders
with employment less than 250.
Automatic imputation using period on period movements
is carried out for non-responding companies with 250+
employment. Manual construction is used in exceptional
circumstances.
For example an influential key
responder would have a manual construction if it did not
respond.
Firms with extreme or atypical key variable returns for
their business size are treated as outliers. Outlier
business information is placed into a separate ‘one-inone’ stratum.
Data from the ABI/2 outputs are used by a wide range of users. The ABI/2 is
vital for a variety of internal and numerous external customers. The key
users and uses of the output include:
• National Accounts – The ABI/2 is central to the compilation of InputOutput tables, which provide a framework for understanding and
analysing the interdependence of industries in the UK.
• Index of Services – The ABI/2 provides weights for index aggregation
and turnover deflation.
• Eurostat – The ABI/2 is a source of annual structural statistics for the
Structural Business Statistics Regulation (SBSR) for policy monitoring
and formulation by the European Union.
• The Scottish Executive and the Welsh Assembly – The ABI/2
provides data on production and construction and are essential inputs
in the calculation of the Scottish and Welsh Indices of Production.
Data on all sectors are incorporated into the Scottish and Welsh InputOutput tables and may also be utilised in internal briefings.
• Department of Trade and Industry (DTI) - DTI uses the ABI/2 data to
assess the structure and performance of industries. They also have
specific interest in the purchases of energy products.
Additional users include other local and national government departments and
bodies, businesses, academics and the general public.
The national ABI First Release5 together with the ABI website outputs2
provides a comprehensive overview of the UK economy. User groups are
consulted to ensure that data remain relevant to their needs.
Created on 21/07/06 – Version 1
2
2 Accuracy
The closeness between an estimated result and the (unknown) true value.
Estimates from this survey are subject to various sources of error. Total error
consists of two elements, the sampling error and the non-sampling error.
More detail on estimates and measures of these errors can be found at
http://www.statistics.gov.uk/abi/quality_measures.asp
Sampling error
This occurs because estimates are based on a sample rather than a census.
The ABI/2 estimate this error through standard errors and coefficients of
variation by industry type.
Non-sampling error
Non-sampling errors are not easy to quantify and include errors of coverage,
measurement, processing and non-response. Response rates give an
indication of the likely impact of a random non-response error on the
estimates.
In seeking to maximise accuracy of the survey, the sample selection for ABI/2
is carried out after the major Inter Departmental Business Register (IDBR)
update processes for the year are complete. This should minimise the
selection of misclassified companies and inadequate coverage of newly
established companies and defunct reporting units.
Various procedures are in place to ensure that errors are minimised. Year on
year comparisons are made at contributor data and aggregate level.
Disparities are investigated to ensure consistent annual returns. Congruence
checks are made against other surveys to ensure that the industries are
reporting mutually consistent values from different surveys.
Another indicator of accuracy is reliability, which can be measured by
assessing the difference between the first published estimate, and the final
revised figure. ABI/2 adheres to a Revisions Policy which can be seen in the
ABI First Release5. Late returns or information received in the course of the
following year’s inquiry may make changes to source data after publication. If
the changes have a significant effect on estimates, published data will be
revised. Significantly revised data will be highlighted at the time of the
release. Further detail is available on the ABI/2 revisions history6 for each
inquiry year.
3 Timeliness and Punctuality
Timeliness refers to the lapse of 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.
The National Statistics Release Calendar7 is available on the National
Statistics website and provides twelve months advance notice of releases.
ABI/2 has consistently met the target publication deadlines. In the unlikely
event of a change to the release schedule, an announcement will be made at
least two weeks in advance, as set out in the National Statistics Code of
Practice8.
Created on 21/07/06 – Version 1
3
The following list shows the time lag between publication and the reference
period to which the ABI/2 data refer.
Provisional National Results Release:
Final National Results Release:
12 months after the reference period.
18 months after the reference period.
The current timings for the publication of the ABI/2 data have been in place
since the implementation of the ABI/2 in 1997.
4 Accessibility and Clarity
Accessibility is the ease with which users are able to access the data, also reflecting the
format(s) in which the data are available and the availability of supporting information. Clarity
refers to the quality and sufficiency of the metadata, illustrations and accompanying advice.
National ABI/2 data are published twice for every annual survey, once in
December at the provisional level and then a final release in the following
June. Each release includes tables, text and charts relating to each industry
classification.
The First Release5 is available in a paper format directly from the press office
and time series data contained within the releases are available to download
free of charge, from the National Statistics website.
The ABI/2 publication team9 can advise on additional data and alternative data
formats.
5 Comparability
The degree to which data can be compared over time and domain.
The ABI/2 covers selected parts of the Standard Industrial Classification
(SIC). Data for 1995 - 2002 have been collected under SIC(92)10 and data
from 2003 have been collected under SIC(2003)11. The full list of the ABI/2
SIC industry coverage12 gives a detailed breakdown of the industries that have
been covered by the ABI since the inquiry was first implemented.
The designation of key variables13 for the ABI/2 has remained fairly consistent
since implementation in 1997, with only minor changes being made to the
questionnaires in accordance with customer needs and requirements.
The ABI/2 has been published on a comparable basis since 1997. Whenever
methodological changes or other factors, such as changes to the SIC, impact
on the latest data, every effort is made to ensure that all previous data are
amended to make them directly equivalent. Policies on revisions to data are
covered in the accuracy section above.
Key discontinuities5 are highlighted in the First Release and information on
discontinuities dating back to earlier than 2003 can be accessed on the
National Statistics website http://www.statistics.gov.uk/abi/background_info.asp
Created on 21/07/06 – Version 1
4
6 Coherence
The degree to which data that are derived from different sources or methods, but which refer
to the same phenomenon, are similar.
The users of ABI/2 require ABI/2 aggregate data to be coherent with other
surveys. Where possible, ABI/2 data are coherent with various short term
indicators produced from the Monthly Production Inquiry (MPI) and Monthly
Inquiry into the Distribution and Services Sector (MIDSS). In addition, ABI/2
aims to be consistent with related annual and quarterly inquiries, such as the
Annual Business Inquiry Employment questionnaire (ABI/1), Capital
Expenditure (Capex), Products of the European Community (Prodcom), Retail
Sales Index (RSI) and inquiries conducted by Department of Trade and
Industry (DTI).
Each of the named ONS surveys uses the same sampling frame, the IDBR.
Thus samples for these surveys are selected from the same universe as the
ABI/2 sample. However there are also specific disparities. For example the
ABI/1 collects employment data from the same sample but at a given point in
time; conversely the ABI/2 returns may relate to different reporting periods.
The MIDSS and MPI surveys both collect turnover information. Nevertheless
their sample sizes are much smaller than the ABI/2 and as such, this will
cause variation in the returns for these short term surveys and the ABI/2. In
addition ABI/2 Construction data and the DTI Construction survey collate
similar data; however these surveys use a different sampling frame and have
some significant differences in their contributor sample selection methods.
Summary of methods used to compile the output
Coverage
The ABI/2 estimates cover UK businesses registered for Value Added Tax
(VAT) and/or Pay As You Earn (PAYE) and are classified to the 2003
Standard Industrial Classification11. The ABI/2 obtains the required details on
these businesses from the IDBR which is then used as the ABI/2 sampling
frame. The ABI/2 sample covers all major industry groups, such as
Production, Construction, Distribution, Service Trades and many more. A
complete listing of the industry groups can be seen at
http://www.statistics.gov.uk/abi/sic92_desc.asp.
The Sample and Design
The ABI/2 sample covers around 76,000 contributors across the majority of
the UK economy. The IDBR is used as the sampling frame from which a
stratified random sample is drawn. The strata are defined by SIC at industry
level, by country and by employment size, with all employment sizes of
companies being covered.
The sampling scheme is designed to give the best estimates of the population
totals for a given sample size and involves selecting all the largest businesses
with a progressively reducing fraction of smaller businesses. This method will
ensure that the sample targets those businesses with the largest contribution
Created on 21/07/06 – Version 1
5
to an industry. It also ensures that the survey sample size of the smaller
businesses minimizes the compliance burden on them.
Weighting and Estimation
As it is not possible to survey every company in the population, to obtain
estimates of the population values it is necessary to weight the data. ABI/2
uses two weights, one to represent the effect of stratified sampling and the
other to represent the effect of ratio estimation.
The implementation of weights ensures the estimates take into account the
characteristics of non-selected businesses. The National Statistics ABI/2
websites provides further detail on ABI methodology14.
Statistical Disclosure
Statistical disclosure control methodology is also applied to ABI/2 data. This
ensures that information attributable to an individual organisation is not
disclosed in any publication. The National Statistics Code of Practice8, and
specifically in the Protocol on Data Access and Confidentiality set out
principles for how we protect data from being disclosed. The Protocol
includes a guarantee to survey respondents that “no statistics will be
produced that are likely to identify an individual unless specifically agreed with
them”. For more information on ONS statistical disclosure control
methodology see: http://www.statistics.gov.uk/about/data/methodology/general_methodology/sdc.asp
Further information on the initial Implementation of the Annual Business
Inquiry can be found in a paper written by Dr Gareth Jones (2000) ‘The
Development of the Annual Business Inquiry’15.
References
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Title of Reference
Statistical Quality Programme
The Annual Business Inquiry
Department of Enterprise,
Trade and Investment (DETINI)
Northern Ireland ABI (NIABI)
The ABI First Release
Revisions History
Release Calendar
National Statistics Code of
Practice
ABI Publication team
SIC 92
SIC (2003)
SIC Industry Coverage
Key variables
ABI Background
The Development of the ABI
Created on 21/07/06 – Version 1
Website location
http://www.statistics.gov.uk/about/data/methodology/quality/default.asp
http://www.statistics.gov.uk/abi
http://www.detini.gov.uk/cgi-bin/gethome
http://www.detini.gov.uk/cgibin/get_builder_page?page=1930&site=4&parent=57&prevpage=2125
http://www.statistics.gov.uk/abi/whatsnew.asp
http://www.statistics.gov.uk/abi/revisions_history.asp
http://www.statistics.gov.uk/ReleaseCalendar/currentreleases.asp
http://www.statistics.gov.uk/about/national_statistics/cop/default.asp
http://www.statistics.gov.uk/abi/contacts.asp
http://www.statistics.gov.uk/methods_quality/sic/contents.asp
http://www.statistics.gov.uk/statbase/Product.asp?vlnk=14012
http://www.statistics.gov.uk/abi/sic92_desc.asp
http://www.statistics.gov.uk/abi/variable_info.asp
http://www.statistics.gov.uk/abi/background_info.asp
http://www.statistics.gov.uk/CCI/article.asp?ID=74
6