Sweden (1)
Title of the survey
Wages and Employment in mining, quarrying and manufacturing
Organization responsible
Statistiska Centralbyrån (SCB) (Statistics Sweden)
Periodicity of the survey
Monthly.
Objectives of the survey
To obtain average hourly earnings of wage-earners, which serve as a
basis for the computation of the quarterly Labour Cost Index.
Main labour topics covered by the survey
Employment, earnings and hours of work.
Reference period
Employment: a normal working day in the month;
Earnings and hours of work: the month or at least a two-week period,
blown up to the full month.
Coverage of the survey
Geographical
The whole country.
Industrial
Mining and quarrying; and manufacturing.
Establishments
All types of enterprises with five or more employees in the private
sector.
Persons
Wage earners only, aged 18 to 64 years.
Occupations
Not relevant.
Concepts and definitions
Employment
Data refer to wage earners. They include piece workers,
casual, temporary, seasonal and part-time workers, as well as persons
temporarily absent from work, provided they have worked some hours
during the reference period.
They exclude supervisors, foremen, store managers, engineers,
apprentices, office clerks, shop assistants, cleaning staff, restaurant
staff, workers in agriculture and forestry, and homeworkers.
Data on the number of wage earners are collected by sex.
Earnings
Data are collected on gross earnings of wage earners. Gross
earnings include pay for normal time worked, premium pay for overtime
and shift work, additional pay for inconvenient hours of work (e.g.
night work) and piece-work compensation, as well as bonuses and premiums
regularly paid at each pay period. These components are known as
direct wages.
They exclude remuneration for time not worked (e.g. public
holiday pay, vacation pay, sick-leave pay, layoff pay), the value of
earnings in kind and irregularly paid bonuses and premiums.
Wage/salary rates
Not relevant. Basic wage rates form part of gross earnings, but they
are not separately identified.
Hours of work
Data relate to hours actually worked. They include normal
hours of work; overtime; time spent at place of work on preparation of
workplace, repairs, maintenance, preparation and cleaning of tools,
preparation of receipts, time sheets, reports; time spent at place of
work waiting or standing by due to breakdown, lack of supply or bad
weather; time spent at place of work during which no work is
done but for which payment is made under a guaranteed employment
contract; and time corresponding to short rest periods at the workplace
including tea or coffee breaks.
They exclude hours paid for but not worked (for holidays, vacation,
sick leave, time off, etc.).
Normal hours of work and overtime hours are separately identified and
measured.
International recommendations
The concept of earnings used in the survey is limited to direct wages
for time worked and regular bonuses and gratuities. It excludes
remuneration for time not worked and the value of payments in kind.
The concept of hours actually worked used in this survey is in line
with that recommended by the international guidelines.
Classifications
Industrial
Data on employment, earnings and hours of work are classified at the
5-digit level of the Swedish Standard Classification of all Economic
Activities (SNI). The 1992 version of the SNI (SNI-92) is based on the
Statistical Classification of Economic Activities of the European
Communities (NACE), Rev.1, which itself is based on the International
Standard Industrial Classification of all economic activities (ISIC),
Rev.3, 1990.
Occupational
Not relevant.
Others
The survey data are classified by employment size of establishments
(5-9, 10-19, 20-49, 50-99, 100-199, 200-499, 500 and more).
Employment data are classified by sex.
Sample size and design
Statistical unit
The reporting unit is the enterprise, which may consist of
one or more establishments. In this case, separate information is
collected with respect to each establishment.
Survey universe / sample frame
The sample frame is derived from the Central Register of Enterprises and
Establishments (CFAR) which covers all Swedish enterprises of all sizes.
This register is continuously updated from administrative sources and by
a special annual inquiry conducted by mailed questionnaires.
Sample design
The survey is based on a sample of approximately 2,400 enterprises with
about 5,200 establishments. The sample is stratified by industry and
employment size. There are altogether 20 strata. All large enterprises
and establishments (500 and more employees) are included with certainty
and a sample is drawn from the other strata. The overall sampling
fraction is 22 percent.
Rotation sampling is used, whereby one fifth of the strata rotates each
year.
Field work
Data collection
By mailed questionnaire. A permanent survey organisation is responsible
for data collection. In case of missing or inconsistent data,
respondents are contacted by telephone.
Survey questionnaire
(Available in Swedish only in national publications).
Substitution of sampling units
There is no substitution of sampling units in case if total non-response
(due to refusal, companies which have closed down or moved with an
unknown address, etc.).
Data processing and editing
Data are processed both manually and by computer, and edited through
machine edit.
Types of estimates
- Totals;
- Averages;
- Distribution by standard error.
Totals are first calculated for each stratum and then aggregated to
obtain totals of employment, earnings and hours actually hours worked.
Average hourly earnings are obtained by dividing total earnings paid
during a pay period covering at least two weeks by the number of hours
actually worked during the same period.
Part-time workers are taken into account as such (no conversion is
done). For missing data, imputations are made from values from
previous periods.
Construction of indices
Monthly series of indices of hours actually worked and employed wage
earners (men and women together), are constructed on the basis of
January 1994 = 100.
Weighting of sample results
The sample results are grossed up to the level of the universe by using
the inverse of the sampling fraction, after adjustment for
non-responses, according to the following formula:
N/(n-nr) where,
N = number in strata
n = number in sample
nr = non-response in sample
Adjustments
Non-response
In the case of non-response, the weights of the different strata are
corrected according to the response rates (see above).
Other bias
Not relevant.
Use of benchmark data
Not relevant.
Seasonal variations
No adjustments are made for seasonal variations.
Indicators of reliability of the estimates
Coverage of the sampling frame
Approximately 99 per cent.
Sampling error / sampling variance
The standard error of average hourly earnings is computed each month,
with a 95 per cent confidence interval.
Non-response rate
In 1994, the non-response rate was about 8.1 per cent in terms of
non-responding units and 3.0 per cent in terms of weighted employment.
Non-sampling errors
Not available.
Conformity with other sources
Not available.
Available series
The following series are published monthly:
- average hourly earnings with 95 per cent confidence intervals, and
changes from the preceding month,
- total number of wage earners, by sex, number of hours actually
worked, number of overtime hours and overtime hours in percent of hours
actually worked,
- index of hours actually worked (including overtime), and
- index of wage earners and changes from the preceding month.
History of the survey
The survey was introduced in 1947, and since then, it has been conducted
on a regular basis without any major changes.
The Swedish standard industrial classification of all economic
activities has been revised a number of times. The previous version
(SNI-69) was linked to ISIC, Rev.2, 1968 at the four-digit level.
Documentation
Statistiska Centralbyrån (Statistics Sweden): Statistiska
Meddelanden, Serie Am 20 SM (Statistical Reports) (monthly,
Stockholm). The data are available some eight weeks after the survey
reference month.
idem: Allman månadsstatistik (Monthly digest of Swedish
statistics) (monthly, ibid.).
Unpublished tabulations can be made available upon request, and data
can be obtained on diskette.
Confidentiality / Reliability criteria
The publication and release of data are subject to confidentiality
rules, whereby tables may not reveal any of the particulars of
individual establishments.
Other information
Data supplied to the ILO for publication
Monthly series of average number of wage earners and
average hourly earnings of wage earners in manufacturing are
published in the quarterly Bulletin of Labour Statistics.
Other sources of data
Statistics Sweden conducts a series of other surveys which provide
statistics of hourly earnings of wage earners in mining and quarrying,
manufacturing and construction (private sector), transport, wholesale
and retail trade (private sector) and agriculture.
These surveys are conducted once a year (during the second quarter) and
usually cover all enterprises affiliated to employers' organizations,
and a sample of non-affiliated enterprises.
The data on earnings include all the regular components of earnings, in
cash and in kind, for time worked or work done, as well as remuneration
for time not worked, and exclude irregularly paid bonuses and
gratuities.
The survey results are published in AM57 Statistiska Meddelanden
(annual).
The following series, derived from these surveys, are published in the
ILO Yearbook of Labour Statistics:
- average hourly earnings of wage earners in mining and quarrying,
manufacturing, construction, transport, storage and communication, and
agriculture, in Tables 17A to 21.