The Scottish Government

« Previous | Contents | Next »

Listen

Scottish Economic Statistics 2004

Definitions, Methodologies and Sources

Industrial Groupings

At the end of 2002 the UK moved from the 1992 Standard Industrial Classification (SIC 92) to SIC2003. Data that apply to 2002 and before (e.g. ABI data in chapter 3, Global Connections Survey) are coded to SIC92, while later data use SIC2003. The quarterly GDP index is all on a SIC 2003 basis. The coding changes (72 new codes to provide additional detail, 1 deletion and 74 changes) mainly affect the fourth and fifth digit of the coding system.

The first two digits of the classification (2 digit SIC), which are used in much of this report, are not affected by the change, with two exceptions: the tiny sector of coin-operated photocopying machines and the new code 7415 for head office activities. The effect on the overall business register statistics of these changes is detailed in the methodological notes below (under corporate sector).

The table below summarises the groupings used in the publication, by SIC code and Scottish input-output category. Where other groupings have been used, these are shown in individual tables. Details of the SIC 92, SIC2003 and progress on the major re-coding exercise planned for 2007 can be found at http://www.statistics.gov.uk/methods-quality/sic/contents.asp.

Table A: Standard Industrial Classifications used in Scottish Economic Statistics

Scottish quarterly GDP (chapter 1)

Gross Domestic Product (GDP) is a measure of the value of goods and services produced by residents, before allowing for depreciation or capital consumption. Net receipts from interest, profits and dividends abroad are excluded. There are two measures of GDP, market prices and basic prices. The Scottish quarterly index is measured in basic prices, which excludes taxes less subsidies on products (taxes on products include VAT and excise duties). Gross Value Added (GVA) is another term for GDP at basic prices.

GDP at market prices is the headline measure used by the UK but they also produce estimates of GVA for their industry breakdown as it is difficult to break down taxes and subsidies below whole economy level. GDP at market prices is not currently produced for Scotland due to the same difficulty of allocating taxes and subsidies below national level.

Information is compiled for over 260 industries in Scotland. For each industry an index is created representing the volume of Gross Value Added (GVA) created by that industry over time. It can be difficult to get value added figures on a quarterly basis so proxy indicators are used where value added data are not available. Examples of proxies used include: deflated turnover, deflated production, the volume of a good or service sold or produced and, for some parts of the public sector, employee numbers. These data come from a range of sources including monthly and quarterly turnover inquiries carried out by the Office for National Statistics; published data sources (e.g. on employment levels or activity levels in certain industries); and data received directly from companies and other organisations.

In February 2004 on publication of results for 2003 Q3, the Scottish GDP estimates moved to annually weighted and chained estimates of volume measures - referred to as "annual chainlinking" - as recommended in the System of National Accounts 1993. This is consistent with the UK where this approach was introduced on 30 September 2003 in respect of the 2003 Q2 results.

The main difference between chainlinking and the previous "fixed base" methodology is that the weights applied to each industry (reflecting importance in the Scottish economy) are updated on an annual basis, instead of a 5-yearly basis. The major effect of chainlinking has been to more accurately reflect the changing importance of sectors. The impact of chainlinking the Scottish GDP series to 2000 weights was to reduce the negative effect of the low/declining growth in some sectors, while simultaneously increasing the importance of those which had been performing well. Both of these changes had a positive effect on the overall level of growth estimated by the Scottish GDP series. However, updating weights does not necessarily result in a positive impact on GDP. For example, if prices in an industry fall over time, despite high growth in the volume of output (as measured by the GDP index), the industry's relative value can decrease. This would result in a fall in the weight, and a reduction in the impact of high growth in that industry. A more detailed explanation of the new methodology and its effects is contained in article A1 and further information is available on the Scottish Executive internet site www.scotland.gov.uk/gdp.

Scottish GDP estimates will generally be less reliable than the estimates for the UK, primarily because the equivalent UK figures are produced by balancing 3 independent sets of estimates (Output (GVA), Income & Expenditure-based approaches). Furthermore, the survey data tend to be based on smaller numbers of units, making figures for Scotland more likely to be subject to small random fluctuations.

Scottish Input-output tables (chapter 1)

A wide variety of data sources are used in the construction of the Scottish Input-Output tables. Wherever possible, Scottish data from ONS inquiries have been used, in addition to data from Scottish Executive surveys or other official sources. The tables are constructed in accordance with international guidance, as given in the European System of Accounts 1995. The industrial classifications are based on Standard Industrial Classifications, as shown in the table above.

Further information on the construction and use of the Input-Output tables can be found on the Input-Output website at: www.scotland.gov.uk/input-output.

Total exports (chapter 1)

Exports by industry and destination have been estimated from Scotland's Global Connections Survey. This is a sample survey, which was circulated to a representative sample of 19,795 companies who had operations in Scotland in 2002. The companies were selected using stratified random sampling from the Inter-Departmental Business Register.

Responses were received from over 5,365 companies (including nil responses). The information provided was supplemented with data from the ONS UK Monthly Production Inquiry, and was then grossed up to cover all companies on the IDBR using turnover. Adjustments were also made to the data for some industries based on other available information.

More information on the sampling and grossing methodology is available on the Scottish Executive website via www.scotland.gov.uk/exports.

Manufactured exports (chapter 1)

Estimates of Scottish manufactured exports are compiled on the standard industrial classification UKSIC(2003). Implemented on 1st January 2003, UKSIC(2003) is a minor revision to UKSIC(92), in line with the latest revision of the European Union classification system.

The estimates are derived from data on sales of goods produced by the Scottish manufacturing industry for export outwith the UK. The Office for National Statistics collects the data used to produce these figures in their Monthly Production Inquiry. The data are then deflated to 2000 prices using UK export producer price indices, which are also produced by the ONS.

Deflated export sales of the companies covered by the survey are then grossed up to represent the entire population using the Inter Departmental Business Register (IDBR).

The data are then seasonally adjusted where appropriate. The Scottish Executive has developed systems to seasonally adjust the series in line with standard National Statistics practice.

The index of Scottish manufactured exports is based on a sample of around 1,000 Scottish manufacturing companies per quarter. This sample covers all sizes of unit across the manufacturing sector.

Further information about the methodology of the Index of Manufactured Exports can be found on the website www.scotland.gov.uk/exports.

Corporate Sector (chapter 2)

The estimates given have been constructed using data from the Inter Departmental Business Register (IDBR), Labour Force Survey (LFS), the Family Resources Survey (FRS) and the Survey of Personal Incomes (SPI). The IDBR extract provides an estimate of the number of enterprises registered for VAT and/or PAYE. A modelling procedure that combines data from the IDBR with estimates derived from the LFS, FRS and SPI is used to calculate the number of unregistered enterprises. The principles of the model were developed by economic consultants working with the Department of Trade and Industry (DTI) and Eurostat.

It should be borne in mind that the number of enterprises with no employees which are not registered on the IDBR is significant. These are estimated from a combination of sample surveys which are all subject to sampling error. For example a yearly estimate of 6,000 taken from the LFS has a 95% confidence interval of +/- 2,640. It is the nature of sampling variability that the smaller the group whose size is being estimated, the proportionately less precise that estimate is. Very small estimates are subject to larger standard errors, which can result in fluctuation between years. For this reason year on year comparisons containing the smallest size band (enterprises with no employees) should be regarded with caution.

The estimates:

• Include enterprises that operate in Scotland irrespective of whether their head office is located in Scotland or elsewhere;
• Count enterprises only once (in tables 2.1 and 2.2) or once each in each of the local areas they operate in (in table 2.3), irrespective of the number of local units they maintain;
• Cover enterprises in the business sector - that is companies, sole traders, partnerships, public corporations/ nationalised bodies and not for profit organisations - and exclude central and local government;
• Include enterprises from the sources described above, as follows (in 2003): 117,775 enterprises were registered for VAT and 29,715 for PAYE only; to this 115,055 enterprises with zero employees are added, estimated on the basis of LFS, FRS and SPI figures.

Employment and turnover values were calculated for all enterprises, however turnover values for Financial Intermediation enterprises have been excluded, as these are not available on a comparable basis.

Registered Enterprises

These are estimated from Scottish extracts of the IDBR taken in November of each year. Enterprises with no UK activity or dummy enterprises created to help with clerical procedures are excluded. Enterprises which have zero employment and zero turnover are also excluded as they are holding companies whose activity is recorded elsewhere or are enterprises not contributing to the economy at the time of the estimates.

Companies with only one employee (who is also the employer) are treated as a self-employed person working in a firm with zero employees unless the enterprise is part of an enterprise group. As these companies provide no employment for others it is more consistent to classify them as enterprises with no employees.

Most tables report on the business sector, excluding central or local government. The tables are therefore affected by the decision by ONS to re-classify Primary Health Care Trusts to the government sector in 2001 (in line with international guidelines or National Accounts) and subsequently in 2003 to re-classify the remaining NHS trusts to the government sector. While this affects relatively few large enterprises, the discontinuity in the employment and turnover tables is considerable: in 2001 the re-classification removed employment of 54,000 from the business sector; in 2003 the effect on business sector employment was -95,530 with a turnover change of -4.7 billion.

These are also the first tables produced to the new SIC2003 industrial classification. At 2-digit (division) level, the change from SIC92 affects the tiny sector of coin operated photocopying machines - effect on count of enterprises and employment is less than 10 - and the new coding for head office activities (7415). As a result of the head office coding change 30 enterprises have changed coding to division 74, with combined employment of 580.

Unregistered Enterprises

The LFS provides data on self-employment without employees (as first or second job). The figure for second jobs is augmented by data from the FRS on self-employment in third jobs. This estimate of the total number of self-employed jobs (with no employees) is compared to the self-employment registered on the IDBR in enterprises with no employees. As many self-employed people are not required to pay VAT or register for PAYE, the figure from the LFS is generally higher.

The difference between the two figures, self-employed jobs in unregistered enterprises, can be used as a starting point to estimate the additional sole traders or partners with no employees. The ratio between sole traders and partners is different for each industry and can be obtained from the SPI. Scottish ratios are calculated where statistically reliable, otherwise UK ratios are used to derive the number of unregistered enterprises from the number of self-employed jobs in unregistered enterprises. Each of these unregistered enterprises is assumed to provide employment for one (if sole trader) or two (if partnership).

Turnover in unregistered businesses will generally be lower than that of registered businesses of the same size, as turnover in the former would usually be below the VAT threshold. Turnover for the unregistered enterprises was imputed from turnover per head of registered enterprises with zero employment in that industry division and then scaled down by a factor of a half. For a few 2-digit industry divisions, this still left average annual turnover per unregistered business above the VAT threshold. In these cases, the unregistered turnover total was adjusted until turnover per unregistered business was under the VAT threshold for that year.

Geographical Analyses

The geographical analyses now use a postcode index file from the General Register Office for Scotland. Unregistered enterprises are not included in the geographical analyses. Hence, the allocable total for Scottish employment and turnover in Table 2.3 does not equate to those in the tables containing full estimates for the zero employee size-band.

Size bands

Enterprises in Table 2.1 to Table 2.5 are classified by employment size-bands on the basis of their total UK employment. The rationale behind this approach is that the size of the overall enterprise determines its behaviour as an economic agent. An enterprise with a large number of employees in the UK as a whole is likely to behave like a large enterprise, irrespective of its level of Scottish employment.

An alternative approach involving the allocation of enterprises to employment size-bands based on their total Scottish employment has not been included. This type of analysis is available on request from the Scottish Executive.

In the Tables 2.4 and 2.5, the enterprises were matched between years by their enterprise reference or by some extra measures such as name and postcode and industry to minimise the number of enterprises that are classified as closures or openings as an error. Some enterprises may have changed their enterprise reference between years, this can be caused by enterprises restructuring. Such enterprises appear as closures and re-openings, that there may be some overstatement of the degree of churning (of enterprises and jobs). It is not currently possible to quantify the numbers of such enterprises.

In addition, enterprises that opened in Scotland after November 2000 but closed before November 2003 are not included in this analysis.

The section on R&D classifies expenditure by product group. Details of these can be found at http://www.statistics.gov.uk/downloads/theme-commerce/MA14-2002.pdf

Scottish Annual Business Statistics (chapter 3)

Background & Methodology

Scottish annual business statistics are sourced from the Annual Business Inquiry (ABI) conducted by the Office for National Statistics (ONS). All figures in this publication are at current prices. For information on the methodology used to compile regional ABI statistics, please visit the following ONS site: http://www.statistics.gov.uk/abi/background-info.asp

Since 1998, the Scottish Executive has paid for an enhanced Annual Business Inquiry sample in Scotland, to improve the quality of Scottish figures. Around 3,000 extra firms in Scotland are sampled annually as a result of this "boost", giving a total annual sample size in Scotland of around 9,000 firms.

The manufacturing, construction and service sectors are based on the Standard Industrial Classifications (SIC92) classifications. Although the tables refer to the service sector for simplicity, it should be noted that the figures quoted do not relate to the entire service sector, but only to those sectors covered by the Office for National Statistics' Annual Business Inquiry. The SICs that are excluded from the ABI are 1.1 - 1.3 (part of Agriculture), 65-67 (Financial Intermediation), 75 (Public Administration) and the following sub-classes of SIC 85 (Health and Social Work): 85.111, 85.12, 85.13, 85.311, 85.321.

In 2000 for the first time, the Forestry (SIC 2) and Fishing (SIC 5) sectors were surveyed as part of the ABI. Parts of Agriculture (SIC1.4 & 1.5) were surveyed for the first time in 2001. Figures for these sectors have been included in this publication. However, caution should be exercised in their use, until fuller comparison and validation is possible when several years' data become available.

For additional Tables not included in this publication, please visit the following website: http://www.scotland.gov.uk/about/ELLD/EI/00016170/Introduction.aspx

Definition of Terms

Labour Market (chapter 4)

Sources

Population data. Table 4.1 shows 2003 mid-year estimates of population. Chart 4.1 shows 2003 mid-year estimates and 2002 based population projections. Population estimates are provided by General Register Office for Scotland and are based on the 2001 census.

Estimates for economic activity, unemployment and total employment, as shown in Tables 4.2, 4.3 and 4.5, Charts 4.2, 4.3 and 4.4, are from the Labour Force Survey (LFS). Figures in Tables 4.2 and 4.5 are from the spring quarter of the survey and are not seasonally adjusted. Figures in table 4.3 are from the Annual Local Area Labour Force Survey data set. The use of different time periods and datasets accounts for the slight differences in estimates between these tables. The LFS is a survey of individuals and is subject to sampling error, with information for local areas based on relatively small samples. For example from the Spring quarter of the LFS, estimates of 10,000 have approximate 95% confidence intervals of +/-4,000.

Unemployment The Government's preferred measure of unemployment is the International Labour Organisation (ILO) definition. Unemployment calculated on this basis is shown in Tables 4.2 and 4.3. This is obtained from the Labour Force Survey. However the Labour Force Survey is not a large enough sample to give reliable estimates of unemployment at small geographical areas such as ward level and some local authorities.

Qualifications and training information shown in Tables 4.7 and 4.8 is obtained from the Spring quarter of the Labour Force Survey. Information in Chart 4.4 is obtained from an annual average dataset from the LFS.

It should be noted that all estimates produced from the Labour Force Survey have been revised over the last year. This is because the information has been re-weighted to be consistent with the 2001 census. This means some information will be inconsistent with what was published in SES 2003.

Estimates of employee jobs, as shown in Table 4.4 are from the June quarter of the quarterly employee jobs series, data are not seasonally adjusted. Estimates in Chart 4.5 are taken from the 2002 Annual Business Inquiry. Data in Table 4.4 and Chart 4.5 are shown by industry groups based on the SIC '92 classification system as outlined in Table A above. Both data sources are surveys of employers and are subject to sampling error, with information for local areas based on relatively small samples. Figures should therefore be treated with caution.

The source of data for Tables 4.9, 4.10, 4.11 and 4.12 showing unemployment figures, is the claimant count. This measures the number of people claiming unemployment-related benefits. Only information on computerised claims is available for age and duration analysis (Table 4.10 and 4.11), however these make up 99 per cent of all claims nationally. Data in Table 4.10 & 4.11 is for June in each year and is not seasonally adjusted.

Table 4.12 shows annual claimant count rates for Local Authorities. The claimant count data are used to estimate the numerator of the claimant count unemployment rate. The denominator is a measure of economic activity for working aged people (ages 16 to 59 for females and 16 to 64 for males), which is a residence-based rate. Economic activity is derived using three sources of data: The Labour Force Survey (LFS), 2001 Census of Population and General Register's Office for Scotland (GROS) mid-year population estimates. The number of economically active people of working age is obtained from combining the four quarters of the LFS for each year. This information is combined with population data at 1991 ward boundary level from the 2001 Census population and updated using GROS population data at Local Authority level for each year. Rates are then scaled so that the Scotland rate equals the rate shown in table 4.9. Table 4.9 shows annual average claimant count unemployment level and work-based rate.

Figures on earnings shown in Tables 4.14 and 4.15 are from the New Earnings Survey. This is a publication produced by the Office for National Statistics. Due to the size of sample the New Earnings Survey cannot give reliable information for all local authority areas in Scotland. The ONS published the NES for the last time in 2003. This will be replaced by the new Annual Survey of Hours and Earnings (ASHE) which is due to be published for the first time on 28 October 2004.

Vacancies data shown in Table 4.13 is not a national statistic. Data are from Jobcentre Plus vacancies notified by industry series (extracted from Nomis). Vacancies reported to Jobcentre Plus are thought to represent approximately one third of all vacancies that are available. Industry groups relate to industry groups based on the SIC '92 classification system as outlined in Table A above.

New Deal data shown in Table 4.16 and Chart 4.7 are from New Deal Evaluation database. An administrative data source compiled by Department of Work and Pensions.

NEET data shown in Table 4.17 are from the 2001 census of population.

Definition of Terms - Labour Market

Household Income and Benefits (chapter 5)

Household income

Source

The household income measures and percentages below median income are derived from the Department for Work and Pensions' Households Below Average Income (HBAI) analyses, which are derived from data from the Family Resources Survey.

Definitions

The income measure used in HBAI is weekly net equivalised household income. Income is the total income of all members of the household, including dependants. Income is adjusted for household size and composition by equivalisation, which reflects the common sense notion that a household of five adults will need a higher income than a person living alone to enjoy a comparable standard of living. The adjusted income is referred to as equivalised income.

HBAI employs two measures of income: Before Housing Costs (BHC) and After Housing Costs (AHC). The need for both measures arises from the variation in housing costs: in part this reflects variations in the quality of housing, but there are also significant cost variations that do not reflect quality variations. The growth in BHC income is likely to overstate improvements in the living standards of low-income groups, because it counts, as an income rise, higher Housing Benefit which merely offsets higher rents. Conversely, income growth AHC will tend to understate improvements in living standards where higher housing costs reflect improved housing. Because of this each measure has imperfections as a guide to differences in, and changes in, living standards, but the two are complementary.

Income Before Housing Costs (BHC) includes the following main components: - usual net earnings from employment; profit or loss from self-employment; all Social Security benefits; income from occupational and private pensions; investment income; maintenance payments (if a person receives them directly); income from educational grants and scholarships (including, for students, top up loans and parental contributions); the cash value of certain forms of income in kind.

Housing Costs include rent; water rates and community water charges; mortgage interest payments; structural insurance premiums; ground rent and service charges.

Income is net of the following: - income tax payments; National Insurance contributions; domestic rates/council tax; contributions to occupational pension schemes; all maintenance and child support payments, which are deducted from the income of the person making the payment; parental contributions to students living away from home.

Benefits

Table 5.3 concentrates on working age people who are in receipt of one or more of a range of key benefits and tax credits. The range of benefits included in this analysis are: Jobseeker's Allowance (JSA); Incapacity Benefit (IB); Severe Disablement Allowance (SDA); Disability Living Allowance (DLA); and Income Support (IS). The tax credits are: Working Families' Tax Credit (WFTC) and Disabled Person's Tax Credit (DPTC). Statistical group, this is defined as follows:

As WFTC and DPTC awards can remain in payment after employment has ended (they are paid for six months irrespective of circumstances), those people who also receive a key benefit (other than DLA) are allocated to the appropriate group based on the benefit received. For example, a person receiving WFTC who also receives IS as a lone parent has been allocated to the "Lone Parent" group. This explains why the numbers in some groups shown in this publication can differ from those shown in the separate DWP and IR benefit and tax credit publications.

Public Sector data (chapter 6)

Income and Expenditure

The data in Tables 6.1 and 6.4 and Chart 6.1 are taken from the Scottish Executive's Government Expenditure and Revenue (GERS) 2001/02 publication. More details on the estimation methods used can be found in the publication, which is available at http://www.scotland.gov.uk/library5/government/gers04-00.asp.

Charts 6.2 and 6.3 take data directly from the HM Treasury Public Expenditure Statistical Analyses (PESA) for 2002/03. This information will feed into GERS 2002/03. The entire PESA publication can be accessed via http://www.hm-treasury.gov.uk/Documents/Public-Spending-and-Services/Public-Spending-Data/pss-pss-pesaindex.cfm.

Tables 6.2, 6.3 and 6.5 are based on Scottish Local Government Financial Statistics 2003/03, which was published by the Scottish Executive in June 2004. Further information is available via the Local Government Finance Statistics website www.scotland.gov.uk/stats/lgfstats.

Public Sector employment

The information in Table 6.6 has been derived from a variety of sources and gives an approximate estimate of public sector employment based on the latest available data from each source. Because it covers different time periods, comparisons over time will not be possible based on this analysis. Work is currently underway within the Office for National Statistics and the Scottish Executive to improve the information on Public Sector employment statistics and to develop time series information.

Organisations have been classified according to the ONS definitions of sectors. The data on Central Government Departments is extracted from the Cabinet Office publication on Civil Service Statistics, which is available at http://www.civil-service.gov.uk/statistics/css.htm. The Armed Forces data refers to 2001, and is an estimate from a DASA table on location of forces which is currently suspended, and is included for completeness.

The NDPB estimate is a combination of figures from the Scottish Public Bodies Directory ( http://www.scotland.gov.uk/government/publicbodies/direct.asp), for bodies under the responsibility of the Scottish Executive, and the UK Public Bodies Directory published by the Cabinet Office for bodies based in Scotland but working to the UK government. Scottish estimates for public bodies that are based throughout the UK have been taken from the Inter-departmental business register (IDBR). NHS staffing data is from the NHS ISD and is available on their website ( www.isdscotland.org).

The local government data is from the Joint Staffing Watch September 2003, which is published by the Scottish Executive at http://www.scotland.gov.uk/library5/government/jswss03-01.asp. The estimate of employment in Public Corporations in Scotland has been derived from the IDBR November 2003.

Please note that the table published in SES 2003 (Table 6.7) contained an error which affected the estimate of Central Government employment. This should have been around 47,500. The table also contained estimates for Further and Higher Education and other bodies that are not included under the National Accounts definition used in the current table.

« Previous | Contents | Next »