Belgium (1)
Title of the survey
Enquête semestrielle sur les salaires (six-monthly survey on wages)
Enquête semestrielle sur la durée du travail (six-monthly survey on
hours of work)
Organization responsible
The Institut national de statistique (INS) is responsible for organizing
and conducting the survey. The results are published under the
responsibility of the INS and the Statistical Office of the European
Communities (EUROSTAT).
Periodicity of the survey
Six-monthly, in April and October.
Objectives of the survey
To follow trends in wages and hours of work.
Main labour topics covered by the survey
Earnings and hours of work.
Reference period
For earnings in the service sector: the months of April and October,
for both wage earners and salaried employees.
For earnings in industry: one pay period in the reference month (April
and October), which does not include public holidays, days of leave nor
compensatory rest days.
For hours of work: a pay period corresponding to one or more complete
working weeks during the months of April and October.
Coverage of the survey
Geographical
Whole country.
Industrial
Industry (i.e. mining and quarrying, manufacturing, electricity, gas and
water, as well as construction) and the services (wholesale and retail
trade, financing and insurance).
Establishments
Private sector establishments and enterprises employing ten persons or
more.
Persons
Wage earners in industry, and wage earners and salaried employees
in the services.
Occupations
Occupations are not taken into account.
Concepts and definitions
Employment
A distinction is made between wage earners and salaried
employees.
Wage earners are defined as all manual workers employed in the
establishment or enterprise as a whole, having a contract of employment
and registered as such with the Office National de la Sécurité Sociale
(national social security office) (ONSS).
The following categories are included: production workers; maintenance
workers; foremen and supervisors generally engaged in manual work;
manual workers paid by the month; manual workers engaged in warehouses,
packing and dispatch; manual workers not engaged in production
(messengers, caretakers, daily cleaners, drivers); part-time, seasonal
and casual workers; shift workers and those providing continuous
service; fitters working outside the establishment; workers below the
age of legal majority; workers who, for personal reasons or because of a
reduction or stoppage of work, have worked only part of the period.
Excluded from wage earners are foremen and supervisors not engaged in
manual work; totally or partially disabled workers; home workers;
workers under contract of apprenticeship with the enterprise and young
trainees engaged under the terms of a social programme designed to
reduce unemployment; workers who, because of sickness or other reasons,
have not performed any work during the survey period. In principle,
therefore, only workers who have worked during part of the survey are
taken into account.
Salaried employees are all employees not counted as wage earners,
employed in the establishment and with a contract of employment. They
include clerical employees, managers, foremen and supervisors. They
also include administrative staff as well as technicians and persons
employed in the commercial or sales services (provided they have a
contract of employment).
Excluded from the category of salaried employees are: the president and
managing director of the enterprise and active members of the board of
directors; employees under apprenticeship contracts with the
establishment; assistants; home workers; persons working entirely or
partly on a commission basis; managers who are majority shareholders;
employees who did not work at all during the reference month and who
were not paid in full by the employer (as a result of a long illness,
recent recruitment, lay off, etc.) and employees engaged on a part-time
basis, i.e. whose contract of employment stipulates that the work may be
performed either on a half-day basis or during a few days of the week,
or during one or two weeks in the month.
Data are collected separately on the number of wage earners and salaried
employees who have freely chosen part-time employment as defined above.
These persons are classified according to the number of hours to be
worked per week under the terms of the contract binding them to the
enterprise.
Earnings
Data are collected on the gross monthly earnings of wage
earners and the gross monthly salaries of salaried employees,
before any deductions at source by the employer for tax, social security
contributions or fines payable by the worker.
Gross monthly earnings of wage earners and gross salaries of
salaried employees are defined as all payments to wage earners
and salaried employees for work carried out during the reference
period, including overtime pay and bonuses paid with each monthly
pay packet.
Gross earnings comprise the hourly, daily or weekly contractual wage;
piece rates, job rates, output rates; extra payment for shift work or
normal continuous work; wages for hours not worked as a result of a
reduction in weekly working hours; punctuality, regularity and
seniority bonuses; premiums for dangerous, dirty or unhealthy work;
premiums for overtime, night work or work on Sundays.
The gross monthly salaries of salaried employees comprise in particular
collectively agreed or non-obligatory children's allowances and
statutory, contractual or non-obligatory vacation pay.
Excluded from gross earnings and salaries are: benefits in kind,
bonuses, end-of-year gratuities and profit sharing bonuses not paid
each pay period, double vacation pay, removal indemnities, allowances
for personal assistance, employers' legal and non-statutory
contributions, legal and non-statutory family allowances, compensation
for wear and tear on work clothes and personal tools, removal and
accommodation costs, wages for hours not worked by wage earners, in
cases where the worker is absent (sickness, leave).
In the survey on the earnings of wage earners in industry,
individual data are collected on the earnings of a sample of wage
earners.
In the survey on the earnings of employees in the services, data on
earnings are collected separately for all wage earners and all salaried
employees, working full-time and part-time, and for each category by
sex.
Wage/salary rates
Not relevant.
Hours of work
The surveys collect data on hours actually worked by wage
earners and on normal hours of work per week for wage earners
and salaried employees.
Hours actually worked correspond to the hours actually worked and
include normal hours of work, overtime hours, hours on night work and on
Sundays.
In the six-monthly survey on weekly hours of work, the total
number of hours actually worked is registered separately for:
- all wage earners on the payroll on the first and last days of the
pay period selected, after deduction of the number of workers recruited
to work less than half the normal hours of work, and
- all wage earners present, i.e. after deduction of the number of
workers not having worked for a single hour during the period.
The survey on the earnings of wage earners in industry collects
data on the total number of hours actually worked (to the nearest
hour) by each worker selected for the sample.
Normal or collectively agreed hours of work refer to the normal working
hours for the majority of wage earners and salaried employees employed
on a full-time basis, as determined by collective agreements or the
internal regulations of the establishments. If the number of weekly
working hours varies from one week to another, the enterprise indicates
the average hours of work per week, expressed in hours and minutes.
International recommendations
The concept of wages and salaries used corresponds to the concept of
regular earnings contained in the international recommendations, except
that payments in kind are excluded from gross earnings.
The definition of hours actually worked corresponds to the international
definition, which includes short inactive periods at the work place
(slack periods, meal breaks, etc.). It also includes hours devoted to
trade union activities at the workplace.
Classifications
Industrial
The data relating to earnings and hours of work are classified in
accordance with the Classification of economic activities of the
European Communities (NACE), Rev.1, 1990. The NACE, Rev.1 is based on
the International Standard Industrial Classification of All
Economic Activities (ISIC), Rev.3, 1990.
Occupational
Not relevant.
Others
The data are also classified according to worker category, size of
establishment and region. The data on earnings are classified by sex.
Sample size and design
Statistical unit
The statistical unit is the establishment in industry
and the
enterprise in the services.
In the survey on the earnings of wage earners in industry,
the final sampling unit is the wage earner.
Survey universe / sample frame
The sampling frame comprises the register of establishments of the ONSS.
This register includes all establishments and enterprises in the sectors
covered by the survey, for which registration is compulsory.
Sample design
The design varies according to the survey.
In the survey on the earnings of wage earners in industry,
establishments and wage earners are selected by random sampling. This
is designed so as to cover a sufficient number of wage earners
to obtain representative figures for each industry. The establishments
are stratified by industry group, for each of which a separate
sampling fraction is used. Large establishments (200 workers or more)
are all included in the sample with a specified proportion of their wage
earners. A selection of medium-sized establishments (50 to 199 workers)
is included with a specified fraction of their wage earners. A
selection of small establishments (10 to 49 workers) is included; these
supply information on each of their wage earners. In the first two
cases, the establishment is asked to provide individual data for wage
earners whose names, in alphabetical order, follow the first name
beginning with a letter selected at random, which is different for each
establishment, until the required number has been covered.
The sample contains only eight per cent of the wage earners employed in
2,500 establishments, out of a total of about one million wage earners
employed in nearly 14,000 establishments.
In the survey on the monthly wages and salaries of salaried employees in
industry and the services, the sample design is as follows:
- In industry, all establishments taking part in the survey on wage
earners also provide information on the number of employees and the
gross monthly wage bill paid in respect of these employees during the
reference period.
- In the services, the universe is stratified by class of activity and
size of enterprise (10 to 49, 50 to 199, 200 or more). All enterprises
employing 200 workers or more are included in the sample. In the other
strata, the sampling fraction varies according to the variables
(activity and size).
Field work
Data collection
The survey is conducted by mailed questionnaire and supplemented with
telephone calls to establishments in the event of missing or incorrect
data, as well as reminder letters to establishments that have not
returned the questionnaire by the required date.
Survey questionnaire
Different questionnaires are sent out according to whether the
survey concerns an establishment or an enterprise, and according
to industry.
The questionnaire relating to wage earners in industry states the
percentage of the establishment's wage earners on whom individual data
are to be provided, as well as the initial of the surname of the first
wage earner to be registered. The data concern the person's name or
number, sex, year of birth, total number of hours actually worked during
the pay period selected (which should not include public holidays or
days of leave), gross wage, and information on the duration and reasons
for absence. The purpose of this is to distinguish between wage earners
who have not been absent more than two hours a week, those who have been
absent because of short-time work, and those who have been absent more
than two hours a week for personal reasons, such as resignation,
dismissal, recruitment, sickness, accident and half-time work schedule.
The questionnaire relating to salaried employees in industry
seeks information on the number of employees in full-time and in
part-time employment, by sex and, for each of these categories,
on the total amount of gross monthly salaries.
The questionnaire for workers in the services covers the number of
workers (wage earners and salaried employees in wholesale and retail
trade, salaried employees only in other services) and the total amount
of gross monthly remuneration. A distinction is made between persons
working full-time and part-time, and by sex. Persons working part-time
are subdivided according to the number of weekly hours of work
required of them under the terms of their contract with the enterprise.
The questionnaire for the six-monthly survey on weekly hours of work is
in two parts on the one sheet. One part concerns wage earners;
information is requested on:
- the days of the month corresponding to half and to full working
days;
- the total number of wage earners registered in the establishment (on
the payroll) on the first and last day of the reference period, after
deduction of the number of wage earners engaged to work less than half
of normal hours of work;
- the total number of wage earners not having worked even one hour
during the period;
- the total number of hours actually worked by all wage earners
during the period;
- the total number of wage earners employed under each system of work
if more than one system is applied in the establishment (five-day,
five-and-a-half-day, six-day week, etc.);
- normal hours of work per week for wage earners in the establishment.
The other part concerns salaried employees and requests
information on:
- the total number of salaried employees present during the
month in question (the number of salaried employees listed on the
payroll for the said month, after deduction of the number of
salaried employees recruited to work part-time);
- normal hours of work per week for salaried employees in the
establishment;
- the weekly hours of work and the number of salaried employees
subject to different systems of normal working hours;
- the total number of hours of paid overtime worked by salaried
employees during the month in question.
These questionnaires are accompanied by detailed instructions on
definitions and calculation methods.
Substitution of sampling units
Not relevant (see also Adjustments for non-response).
Data processing and editing
A certain number of operations are carried out manually. Among these
are the preparation of lists of enterprises and establishments, and of
questionnaires, with an indication of the instructions on the total
number of wage earners to be selected and on the letter of the alphabet
to be used as a sample basis, the reception and initial editing of
questionnaires (number of wage earners, missing codes, calculation of
the average monthly wage per employee in each establishment, etc.),
reminder letters, calculation of average weekly working hours in each
establishment, average monthly wage, etc.
Computerized processing is carried out in respect of the editing of
establishment identification data, calculation, checking and comparison,
by establishment, of average gross hourly earnings, calculation of the
weighted average monthly wage by economic activity and by enterprise and
sex, and the production of the results.
Types of estimates
- average gross hourly earnings of wage earners;
- average gross monthly salaries of salaried employees;
- average weekly hours of work per wage earner and per salaried
employee.
In industry, the averages for wage earners are those for workers who
were not absent more than two hours per week.
The gross hourly earnings per wage earner are obtained by dividing the
total gross earnings by the number of hours worked during the pay
period.
For salaried employees in industry, the averages are established for
persons employed on a full-time basis only.
The average gross monthly salary per establishment, for men and women
separately, is obtained by dividing the total amount of monthly salaries
paid to all salaried employees of an establishment by the number of
employees concerned.
The same procedure is applied in estimating average earnings in the
services. However, part-time workers are included, after having been
converted to full-time equivalents.
Average weekly hours of work per wage earner are obtained by dividing
the number of hours actually worked by the each wage earner (present or
registered) in each establishment during the reference period, by the
number of working days during the period, and by multiplying the result
by five, i.e. the number of working days in a normal week.
For all the statistics, workers whose wages or salaries were reduced
because of absence for all or part of the reference period are not
included.
Construction of indices
Indices are calculated in respect of average gross hourly earnings of
wage earners in industry and average gross monthly salaries of salaried
employees in industry. These indices are compiled using as fixed
weights total employment based on the results of the survey of wage
structure and distribution (currently, that of 1978/1979).
Weighting of sample results
In industry, the data compiled at the establishment level are
extrapolated to the survey universe using two coefficients: one enables
the expansion from the sample of employees to all employees in the
establishment; the other enables the expansion from the establishments
observed to all establishments covered by the sampling frame.
Then for each industry group at three-digit level of the NACE, average
gross hourly earnings are calculated, for men and women aged 21 years
and over, and for men and women below the age of 21. These averages are
obtained by dividing the total of gross hourly earnings of the wage
earners concerned by the number of wage earners of each corresponding
sex and age group. All these data are then aggregated using current
weights so as to obtain, for each industry group, the average gross
hourly earnings for all men, all women, and for men and women of all
ages together.
For employees in industry, the data at the establishment level for men
and women are extrapolated to the universe and aggregated with a view to
presenting the results for each sex and for men and women combined at
the level of NACE industry classes. These aggregates are calculated
using current employment weights. To expand from the industry group
level to industry classes and divisions, a fixed weighting is used,
which reflects the estimated number of employees per sector.
The same procedure is used for weighting average wages in
the services.
Adjustments
Non-response
No adjustments are made in the case of non-response, since
enterprises are legally obliged to reply to the questionnaire
under the provisions of a royal decree enacted in 1974. In
principle, six months after the reference period, all
establishments should have sent the INS completed questionnaire
relating to the six-monthly survey on wages and hours of work.
Other bias
Not relevant.
Use of benchmark data
Not relevant.
Seasonal variations
None.
Indicators of reliability of the estimates
Coverage of the sampling frame
Complete, in principle. Registration with the ONSS is compulsory
for enterprises and establishments.
Sampling error / sampling variance
The standard error is not calculated.
Non-response rate
Not relevant. The INS tries to reduce non-response to a minimum by
maintaining close contact with establishments: written and telephone
reminders, visits by enumerators to collect information, etc.
Non-sampling errors
Not known.
Conformity with other sources
Not relevant.
Available series
The following series are published every six months:
- average gross hourly earnings of wage earners in industry for
Belgium, presented separately for adult men, adult women, for all men,
for all women and for all wage earners, by economic activity at the
three-digit level of NACE;
- same tables as above for the three regions and the nine
provinces;
- average gross hourly earnings of wage earners in industry by
size of establishment;
- average gross monthly salaries of salaried employees in industry for
Belgium, presented separately for men and for women, and for all
salaried employees together, by economic activity at the
three-digit level of NACE;
- same tables for the three regions and nine provinces;
- average weekly hours of work per registered wage earner, per wage
earner present, average weekly hours offered per wage earner and average
weekly hours per salaried employee, and by economic activity;
- indices of average gross hourly earnings of wage earners in
industry;
- indices of average gross monthly salaries of salaried employees in
industry.
History of the survey
The annual earnings surveys held in October of each year have
been conducted since 1952.
The current six-monthly survey started in April 1961 and,
since that date, has supplied data on earnings and hours of work
in industry, on the basis of a series of definitions that have
changed very little since that time. In 1964, the survey was
adapted to the new harmonized statistics of earnings established
by EUROSTAT. In October 1966, slight modifications were
introduced with the adoption of a new sampling frame and the NACE
industrial classification. Fresh modifications were introduced
in 1972 with the extension of the survey to include salaried
employees in industry and workers in the energy sector. The
results are published according to the new classification
of economic activities - the general Classification of Economic
Activities of the European Communities (NACE), Rev.1.
In 1978, the system was supplemented by six-monthly wage surveys in the
wholesale and retail trades, credit institutions and insurance
companies.
Documentation
Ministère des Affaires Economiques, Institut National de
Statistique: Statistiques Sociales (quarterly; Brussels).
Statistical Office of the European Communities (EUROSTAT):
Earnings - industry and services (twice a year, Luxemburg).
Methodological information is also available in:
idem: Harmonized earnings statistics - National survey
methodology (ibid., 1992).
The data are also available on diskette and magnetic tape.
Confidentiality / Reliability criteria
No data are published with respect to cells with fewer than four
establishments.
Other information
Data supplied to the ILO for publication
The following series are published in the Yearbook of Labour
Statistics:
- average hours of work per week (hours actually worked) of wage
earners, in October each year;
- average hourly earnings of wage earners - October;
- average monthly earnings of salaried employees - October;
in non-agricultural activities, mining and quarrying, manufacturing and
construction.
The six-monthly data are published in the corresponding tables of the
Bulletin of Labour Statistics.