Germany (1)

Title of the survey

Survey of Earnings in Industry and Commerce.

Organization responsible

Federal Statistical Office (Statistisches Bundesamt), in cooperation with the statistical offices of the Länder.

Periodicity of the survey

Quarterly (in January, April, July and October).

Objectives of the survey

The results of the survey are used mainly for the continuous observation of economic development and are thus one of the bases for decision-making for economic, social and business-cycle policies, as well as for use by employers' and employees' associations.

Main labour topics covered by the survey

Earnings and hours of work.

Reference period

The whole calendar month (or a minimum of four weeks).

Coverage of the survey

Geographical

The whole country. (Up to the reference month of October 1990, the survey covered only the territory of the Federal Republic of Germany prior to 3 October 1990. Since January 1991, it also covers the five new Länder and Berlin-East.)

Industrial

Industry (i.e. mining and quarrying, manufacturing, energy supply and construction, including handicraft establishments of structural and civil engineering), commerce, financing and insurance.

Establishments

In industry (except mining and quarrying): establishments with ten or more persons engaged; In mining and quarrying: all establishments. In civil engineering work and in commerce, credit institutions and insurance: establishments with five or more persons engaged.

Persons

Full-time wage earners and salaried employees subject to contributions to old-age pension schemes for wage earners or salaried employees (subject to a minimum earnings limit fixed at DM 15,000 since 1994).

Occupations

Not relevant.

Concepts and definitions

Employment

Data on employment refer to wage earners and salaried employees in industry; and to salaried employees only in wholesale and retail distribution, banking, credit institutions and insurance. Wage earners are all full-time dependently employed persons subject to contributions to the pension insurance fund for wage earners, except the groups of employees mentioned below under exclusions. Salaried employees are all full-time dependently employed persons subject to contributions to the salaried employees' pension insurance fund, including persons to whom special exemption provisions apply (i.e. all salaried employees receiving a regular gross monthly salary of less than DM 15,000, including those to whom the agreed pay scale does not apply (the so-called AT employees)). Excluded from the survey are the following groups of employees:

part-time wage earners and salaried employees, i.e. persons whose employment contract provides for a shorter work week than normal (e.g. certain hours, half-days or only certain days of the week);
wage earners and salaried employees who did not receive a wage or salary for the entire reference month, for reasons such as sickness (after the period of guaranteed payment of wages has elapsed), hiring or dismissal during the reference period;
wage earners and salaried employees who, for other reasons, were absent without pay for more than three days during the reference period;
wage earners and salaried employees who receive pensions from the statutory pension insurance fund and other pensions, and for this reason, receive a reduced remuneration;
unpaid family workers;
paid and unpaid trainees and voluntary workers;
apprentices in officially recognized (or exceptionally authorized) training occupations and specialized fields;
homeworkers and masters responsible for homework contracts;
statutory representatives of corporate entities (members of the management committee and their deputies and managing directors of limited liability companies);
senior managerial employees with full powers of supervision and decision (these employees form part of Performance Group I - see below);
staff members employed by private insurance enterprises in the advertising field organization or for the collection of premiums.

Both wage earners and salaried employees are classified into performance groups (or employment categories) (Leistungsgruppen), which are established for each collective agreement by the Federal Statistical Office in consultation with the social partners concerned on the basis of the various wage and salary categories. Performance groups are based on the level of training, experience and responsibility required for performing the tasks covered by the various earnings categories.

Wage earners are classified according to three performance groups (skilled, semi-skilled and unskilled).
Salaried employees are classified according to four performance groups (II to V - managerial employees of Group I being outside the scope of the survey) and two occupational categories: commercial employees, and technical employees and masters.

Data for each performance group and occupational category are collected by sex.

Earnings

Data refer to total gross wages or salaries and are collected for each performance group and occupational category, by sex. Gross wages or salaries include all regular payments which were actually paid to wage earners or salaried employees during the reference period, including remuneration for time not worked (e.g. for annual leave, vacation, public holiday). They comprise:

collectively or individually agreed wages or salaries;
performance and efficiency payments and bonuses, social and other payments such as payments for overtime worked during the reference month, whether or not provided for by collective agreements;
income tax payments and employees' social security contributions to funds voluntarily taken over by the employer;
commissions relating to the reference month, if commissions (related to sales) are paid in addition to a fixed wage or salary;
bonuses, end-of-year premiums, thirteenth month payments, profit-sharing payments, etc. which are paid in monthly instalments;
monthly contributions paid by the employers to workers' savings and capital formation schemes;
vacation pay of construction workers, even if refunded by the vacation funds;
reimbursement of travel expenses and relevant allowances for travelling to and from the place of work;
the tax value of free board and free accommodation, in so far as these benefits constitute all or part of the gross wage or salary (free full board means the supply of at least three meals per day; breakfast only and the occasional provision of meals are not taken into count);
deductions retained in the reference period for reimbursing loans, salary advances, etc.

Regular wages and salaries exclude the following:

social security contributions paid by the employers;
the employer's additional health insurance payments for those employees who are not obliged to take out health insurance in accordance with the Second Law amending the health insurance provisions;
bonuses paid on a one-off basis or at longer intervals, Christmas bonuses, profit-sharing schemes or payments, employer's contributions to workers' savings schemes, end-of-year bonuses, thirteenth month payments, holiday compensation pay;
additional holiday bonuses, on the basis of agreed rates or ex gratia;
savings bonuses payable to wage and salary earners by virtue of paragraph 12 of the Third Law on Capital Formation;
the miners' bonus paid in the mining industry (however, information on this type of bonus is collected separately);
the amounts to be paid by employers in the building industry to supplementary pension funds for supplementary benefits, pay compensation, vacation, vocational training and promotion of winter construction;
travel time remuneration, winter bonus and bad-weather compensation in the building industry;
reimbursement of expenses incurred and separation allowances, daily severance allowances, etc. (unless subject to wage tax);
payments made from operating or fiscal funds for support or welfare in the case of reduced working hours for operational reasons (e.g. short-time work);
statutory child allowances, even if they are paid by the employer, as in the civil service;
wage and salary advances and loans received during the reference period; payment of salary arrears and tax rebates;
payments in kind (except free meals and accommodation).

Each establishment covered by the survey is required to calculate:

the average gross salary of salaried employees, by dividing the total gross salaries for each performance group, occupational category and sex by the corresponding number of full-time salaried employees;
the average gross hourly earnings of full-time wage earners, by dividing the total gross wages for each performance group and sex by the corresponding number of hours paid for (see below).

In January of each year, data are collected on the collective agreements applicable in the surveyed establishments. Collective agreements are agreements applying to more than one enterprise. Enterprise agreements are wage agreements concluded between the enterprise or establishment and a partner capable of negotiating in collective bargaining (trade union). Establishment agreements are wage agreements concluded between the enterprise or establishment and the works council. Establishments whose working conditions are not regulated by one of the above three types of contract are classified under individual agreements.

Wage/salary rates

Not relevant.

Hours of work

Data are collected on hours paid for for wage earners only. They include hours actually worked during normal periods of work, overtime and hours paid for but not worked, such as for paid periods of absence due to sickness, public holidays, vacation, breaks, hours off for operating or personal reasons (employee meetings, annual works outings, visit to the doctor or family leave). If vacation pay is calculated on the basis of earnings for a past period, the number of paid hours (including overtime) is also ascertained for that period and indicated as paid vacation time. Hours paid for exclude the hours that were worked but were compensated only in the form of days off; rest days granted within the scope of reduced working hours (which are considered neither as working hours nor as paid hours lost); meal breaks; and unpaid inactive periods of time spent outside the workplace (e.g. for personal leave, military or related service, study or training leave, industrial disputes). Data are collected separately on overtime hours. These correspond to the hours in excess of the customary working hours in the establishment, that were not compensated by granting non-working time on other days, irrespective of whether or not a premium is paid for such hours. Hours that are worked in advance are not considered as overtime hours. Data are also collected on the regular working hours in the establishment. These refer to the working hours fixed in collective agreements (excluding overtime), or to the regular working hours fixed by the internal regulations in the establishment. Hours paid for public holidays are included. For employees whose remuneration is fixed on an hourly basis, the monthly regular hours correspond to weekly hours multiplied by four (for the number of weeks) plus the hours on the 29th, 30th and 31st day of the reference month. For employees whose remuneration is fixed on a monthly basis, they correspond to the fixed regular number of hours which constitutes the basis of the monthly wage or salary.

International recommendations

The definition of gross wages and salaries corresponds to the concept of gross earnings for current statistics contained in the international recommendations, in that it excludes irregular and ad hoc payments. It excludes the value of payments in kind (except for free board and free accommodation). Two concepts of hours of work are used: normal hours of work in the establishment, and hours paid for of wage earners.

Classifications

Industrial

Data on employment, earnings and hours of work are classified according to the Industrial Classification of Economic Activities (1979). This classification can be linked to a certain extent to the International Standard Industrial Classification of all economic activities (ISIC), Revision 2, 1968. Since January 1995, the results are classified according to the WZ93 Classification, which corresponds to the NACE Revision 1. This can be converted to ISIC Revision 3, 1990, at the division level.

Occupational

Not relevant.

Others

All data are classified by performance group, occupational category (for salaried employees) and sex.

Sample size and design

Statistical unit

The sampling and reporting unit is the establishment, defined as a local unit of an enterprise. (Up to April 1992, the sampling and reporting unit in the five new Länder and Berlin (East) was the enterprise.)

Survey universe / sample frame

The sample is selected from the Register of the production industries and the Register of the Census of non-agricultural local units. This census is conducted at more or less regular intervals, in general every ten years. The last Census was conducted in October 1991 (July 1992 for the five new Länder and Berlin (East)). The Registers are not updated between censuses.

Sample design

A stratified sampling design is used. The sample of establishments is first stratified by federal Land (16 Länder). Within each Land, the establishments are then stratified according to 83 economic activities and six size classes. The sampling rates within the individual strata are chosen so as to produce the minimum sampling error (the simple relative standard error for the average hourly earnings for the individual branches of activity should be not greater than 1 per cent). The size of the sample in each stratum is determined according to the Neyman-Tschuprov law: the number of units to be included in each stratum is proportional both to the size of the universe and to the standard deviation of the variable (gross hourly earnings) within the stratum. The sampling units are selected by systematic random sampling, in accordance with a defined system: if, during the process of selection, an establishment that was in the previous sample comes up again, the next establishment on the list is selected in its place, provided that it was not itself part of the previous sample. If it was, the next establishment on the list is taken, and so on. However, this procedure is not used if it ends up with an establishment that immediately precedes one which is selected by the system. If this occurs, the establishment originally selected by the system is retained in the sample so as not to jeopardize the principle of random selection. Since July 1992, the sample has comprised some 40,500 establishments, in both the old and the new federal Länder. The sample is changed following each Census. At that time, the survey is conducted in establishments in both the old and new samples, in order to maintain continuity in the statistical series.

Field work

Data collection

Data are collected by the staff of the statistical offices of the Länder, in accordance with uniform guidelines issued by the Federal Statistical Office. Mailed questionnaires are used.

Survey questionnaire

Separate questionnaires are used for wage earners and salaried employees, depending on the different collective or individual agreements and on the branch of activity covered. If the workers in the various production units of an establishment are paid under different collective agreements in force a separate questionnaire is filled in for each branch of activity and agreement. If there is only one collective agreement, the workers are grouped together on one form. In this case the establishment is classified according to its main activity, that is the one which employs most of the staff, or on the basis of the volume of production. The questionnaires are designed to collect the following information: For full-time salaried employees:

number of employees, by performance group, occupational category and sex,
total gross salaries for the above sub-groups,
average gross salary of each sub-group;

For full-time wage earners:

number of wage earners, by performance group and sex,
total number of hours paid for, including overtime, by sub-group,
total number of overtime hours, by sub-group,
total gross wages for the total number of hours paid for, by sub-group,
average gross hourly wages, by sub-group.

For salaried employees, data refer to the whole month because their salaries are usually calculated and paid monthly. In the case of wage earners, the establishment has the choice of supplying data for one or more pay periods, i.e. for a period of a month or at least four weeks which must fall entirely or mainly in the reference month. Thus, if wages are calculated for a period of three weeks, the declaration is made for two entire periods, that is, six weeks, and if the calculation is based on a period of ten days the survey period consists of three full periods of ten days. Detailed instructions on definitions, methods of calculation, items to be included or excluded, etc. are provided along with the questionnaires.

Substitution of sampling units

Not relevant (see under "Adjustments").

Data processing and editing

Data are processed manually and machine tabulations are used for data checking. In the case of missing or inconsistent data, the statistical offices of the Länder contact the respondents by telephone, and then communicate the information to the Federal Statistical Office.

Types of estimates

Average hourly, weekly and monthly earnings, and average hours paid for per week, according to a number of variables. Average earnings of wage earners are calculated on the basis of the total wage bill and hours paid for in the reference period. Average earnings of salaried employees are based on the total salary bill and the number of persons employed during the reference period. Average hours paid for, for wage earners, are calculated on the basis of the total number of hours paid for and the number of wage earners during the reference period. For each statistical unit (the establishment), average values are calculated for the individual variables (e.g. gross earnings per employee, average normal hours worked per week and per wage earner and average gross hourly earnings), and these are used for plausibility checks. If the current values deviate considerably from those for the previous quarter, this indicates an error. After each plausibility check, and in each stratum, total weekly hours, earnings and employment are calculated by adding the corresponding figures for all sampled establishments. The data are then extrapolated (see under Weighting of sample results).

Construction of indices

The following index series are calculated:

index of hours paid for,
index of gross hourly or weekly earnings of wage earners,
index of gross monthly earnings of salaried employees.

Indices are computed according to the Laspeyres formula, using as weights the number of employees, earnings and hours corresponding to the base year (1985=100), and multiplying the relatives by the relative weights of the weighting scheme. (Indices based on 1991=100 are available since mid-1995.)

Weighting of sample results

The sample results per stratum are multiplied by the raising factor (i.e. the ratio of the number of establishments in the universe to the number within the sample), corrected by the replacement factor. The extrapolated figures are then aggregated.

Adjustments

Non-response

A distinction is made between false and genuine non-response. False non-response relates to an establishment which no longer existed at the time the survey was conducted, was no longer involved in industry or commerce or no longer constituted an independent local unit. In such cases, the units concerned are crossed off the list and not replaced. Genuine non-response refers to an establishment which simply ignored the survey. In this case, a replacement factor is applied, which corresponds to the quotient of the number of establishments contacted less the number of non-existent establishments and the number of establishments that replied plus the number of non-participating establishments. The number of establishments contacted is the corrected number of establishments in the sample, i.e. after crossing off the false non-responding establishments. In the case of an establishment that deliberately ignores the survey (genuine non-response), but has also changed its area of business, this non-response is considered to be a false non-response, and consequently the establishment is eliminated without being replaced. The adjustment factor is not applied. In the new branch of activity, this non-response is offset by collecting and analysing the data for a replacement establishment of comparable structure.

Other bias

No adjustments are made for any other bias.

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

Not available.

Sampling error / sampling variance

The sampling rates within the individual strata are chosen so as to produce the minimum sampling error. The simple relative standard error for the average hourly earnings for the individual branches of activity should not be greater than 1 per cent.

Non-response rate

Not available. The obligation for employers to provide the required information comes from Art. 12 of the Law on Wage Statistics in connection with Art. 15 BStatG.

Non-sampling errors

Not available.

Conformity with other sources

Not relevant.

Available series

The following series are prepared quarterly:

for wage earners in industry: average hours paid for; average gross (hourly and weekly) earnings; index of average hours paid for and of gross (hourly and weekly) earnings;
for salaried employees in industry and commerce: average gross monthly earnings; and index of average gross monthly earnings.

These data are classified by sex, employment category, branch of activity and region. Annual averages of the results of the four quarters: average hourly or weekly earnings of wage earners; average monthly earnings of salaried employees; average paid weekly hours; by branch of activity, occupational category and sex.

History of the survey

The present quarterly survey was established in 1957. It replaced the former quarterly series on earnings of manual workers in industry. Reforms took place in 1964, January 1973, October 1982 and October 1991, when new samples were selected using the results of the most recent industrial census. In order to solve the problem of continuity in the series when the new sample is selected, the survey is conducted with respect to establishments in both the old and the new sample at the time of the change-over. The results are then compared and conversion coefficients established for each branch of economic activity. These can be used to convert the results for the previous period by simple multiplication, giving a valid comparison between the two sets of results. Indices have been computed on the following bases: 1958=100, 1962=100, 1970=100, 1980=100 and 1985=100. As from mid-1995, indices will be calculated using base 1991=100.

Documentation

Statistisches Bundesamt: Löhne und Gehälter, Arbeiterverdienste in der Industrie, Fachserie 16, Reihe 2.1 (quarterly; Wiesbaden). idem: Löhne und Gehälter, Angestelltenverdienste in Industrie und Handel, Fachserie 16, Reihe 2.2 (quarterly; ibid.). The data are published some three to four months after the survey reference period. Publications contain all available data, unless they are considered as confidential.

Confidentiality / Reliability criteria

Pursuant to Article 16, BStatG, the microdata collected are, as a general principle, kept secret. Microdata may be passed on only in exceptional cases specifically provided for by legal provisions. In accordance with Article 16, paragraph 6, BStatG, microdata can be made available to institutions of higher education or other institutions entrusted with tasks of independant research for carrying out scientific projects, if those microdata have been rendered relatively anonymous, to the extent that they can only be allocated to the respondents or parties concerned by employing an excessive amount of time, expenditure and manpower. The confidentiality provisions also apply to persons who receiving and using microdata.

Other information

Data supplied to the ILO for publication

Annual averages of hours paid for and hourly and weekly earnings of wage earners in non-agricultural activities, manufacturing, mining and quarrying and construction are published in Tables 11 to 14 and 16 to 19 of the Yearbook of Labour Statistics. Series relating to the old federal Länder and the five new Länder and Berlin (East) are shown separately. The corresponding quarterly series for non-agricultural activities and manufacturing are published in Tables 5 to 8 of the Bulletin of Labour Statistics.

Other sources of data

An annual Survey of Earnings in Industry and Commerce is also conducted during the first quarter of each year, with reference to the whole preceding calendar year. It covers the same categories of employees as the quarterly survey. Data are not collected on hours of work, but on total annual gross wages and salaries for each employee category (wage earners and salaried employees) and by sex. Total annual gross wages and salaries are based on the accounts which are maintained by the establishments for every employee, pursuant to Article 41 of the Income Tax Law. They include the following components:

employment compensation (cash compensation and tax value of the remuneration in kind, including the continued payment of wage or salary), without any deduction and without reduction by the general tax-free amount granted to all employees and the tax-free Christmas allowance; and end-of-year premiums, profit premiums, shares in profits, thirteenth month wage or salary, additional vacation bonuses, Christmas bonuses, compensation for travel expenses, etc.;
remuneration that is not part of the employment compensation subject to tax;
remuneration for inventions made by employees;
premiums paid under the employee suggestion schemes;
remuneration taxed at a fixed flat rate or at special flat rates and the relevant wage tax, if t

Excluded from the total annual gross wages and salaries are:

the employer's share of the contributions to social insurance funds;
the employer's allowance for the contribution to the health insurance fund, granted to employees not subject to contributions to the statutory health insurance;
employees' savings premia pursuant to paragraph 12 of the Third Law promoting Capital Formation by employees;
the miners' bonus granted in the mining industry;
amounts paid by the employers in the building industry to the supplementary pension fund of the building industry;
travel time remuneration in the building industry;
payments made from operating or fiscal funds for support or welfare in the case of reduced working hours due to operating reasons (short-time working);
statutory payments for dependent children even if paid by the employer, as in the public service;
reimbursement of expenses incurred, other transitory funds, separation allowances, any wage-tax free daily allowances;
wage advances granted in the calendar year for subsequent periods, and back payments and repayments in the calendar year concerning previous calendar years, as well as loans granted in the calendar year;
payments in kind (except free board and free accommodation).

This definition of total gross wages and salaries corresponds to the concept of gross earnings in cash contained in the international recommendations and used in annual or infrequent statistical inquiries. It includes all regular and irregular payments, but excludes the value of most payments in kind (except free board and free accommodation). The annual survey uses the same sample and estimation methods as the quarterly survey. The results of the annual survey are published in: Statistisches Bundesamt: Löhne und Gehälter, in Industrie und Handel, Fachserie 16, Reihe 2.1 and Reihe 2.2 (annual; Wiesbaden).