Statistics Estonia

Population and Social Statistics Department

Tatari 51, 10134 Tallinn Estonia

Regulation (EC) No 450/2003 of the European Parliament and of the Council of 27 February 2003 concerning the labour cost index
Commission Regulation (EC) No 1216/2003 of 7 July 2003 implementing Regulation (EC) No 450/2003 of the European Parliament and of the Council concerning the labour cost index
Commission Regulation (EC) No 224/2007 of 1 March 2007 amending Regulation (EC) No 1216/2003 as regards the economic activities covered by the labour cost index

The organization of the report follows the criteria on statistics quality, as defined by Article 2 and Annex 1 of the Commission Regulation (EC) No 1216/2003 of 7 July 2003.

Labour cost indices (wages and salaries; labour costs other than wages and salaries; total labour costs; total labour costs excluding bonuses) by economic activity sections (identified by letters B to S in EMTAK)

Estonian Classifcation of Economic Activities (EMTAK 2008) based on NACE Rev. 2

Economically active units – all enterprises, institutions and organisations (entrepreneurs without employees are excluded)

Wages and salaries index – index, which defines change of wages and salaries costs per hour.

Employers’ social and imputed contributions index – index, which defines change of employers’ social and imputed contributions costs per hour.

Labour cost index – index, which defines change of total labour costs per hour.

Labour cost index excluding bonuses – index, which defines change of total labour costs per hour excluding bonuses.

Economic unit

List of economically active units, excluding sole proprietors without employees. The list is generated from the Business Register for Statistical Purposes.

Estonia as a whole

Labour costs indices from 2000 onwards.

The base year for the indices is 2016 = 100.

Since 2009 the new classification of economic activity was required. That is why we recoded all our quarterly databases on the micro level back to year 2000.

Until the year 2002 the quarterly wages and salaries survey questionnaire didn’t include separate questions about social tax. Therefore the other labour costs and total labour costs of 2000 and 2001 were estimated by quarters using the respective proportions of the survey of 2002. Hourly labour cost and other hourly labour cost items were calculated.

Hourly labour costs are calculated according to the formula:

LCt = TLCt / THt, where

TLCt is the total labour costs in time t,

THt is the total number of hours worked in time t.

The labour cost index LCI_t is calculated as follows:

LCIt = LCt *100 / LCby, where

LCt is the hourly labour costs in a quarter t,

LCby is the hourly labour costs in the base year.

The similar formula is applied in order to calculate two sub-indices, i.e. the wages and salaries indices and other labour cost indices by economic activity.

Since 1997 the variable monthly hours worked and variable wages and salaries are collected from same sources — wages and salaries survey.

Definition of the variable hours worked is as follow:

Hours actually worked by full-time and part-time employees (incl. time spent on tasks such as work preparation, preparing, maintaining and cleaning tools and machines and writing up work cards and reports; time spent at the place of work during which no work is done owing to, for example, machine stoppages; tea and coffee breaks, etc.) and overtime working hours.

The procedures include a logical checking process after the data entry process.

Administrative data are not used.

Since 2002 all labour cost items and hours worked are collected from the same sources (wages and salaries survey). All labour costs items and hours worked are collected monthly - no different frequencies used. 

In wages statistics the statistical unit is an enterprise, institution or organisation. The population of a survey consists of all economically active units. The register of economically active units is based on the Estonian Enterprise Register. The register of economically active units is updated every year.

Since 1997 the completely surveyed part is formed of enterprises with 50 or more employees and of all state and municipal organisations and institutions. From the remaining part of the population, i.e. from enterprises with less than 50 employees a stratified simple random sample is selected. Non-profit organisations with less than 50 employees were excluded from the survey until 2018. Since the 1st quarter of 2018 non-profit organisations with less than 50 employees were included in the population. Since the 1st quarter of 2020 state and municipal organisations and institutions with less than 50 employees are not included in completely surveyed part.

• From the completely enumerated part, data are collected every month.

• For the sampled part the quarterly data is collected once in a quarter - the units are divided between the three months where for each unit one month represents the whole quarter.

For selecting samples and in order to rotate samples, permanent random numbers are used. Each year the number 0.02 is added to the starting point of a sample. It means that on the average in stratum where sampling fraction is 10%, one fifth of units are changed. So, after 5 years can expect that the whole sample is replaced entirely.

For selecting samples and in order to rotate samples, permanent random numbers and coordinating by response burden are used. Method accounts with response burden over all surveys with similar sample design beginning from year 2020. Sample is taken from sorted frame, where units are sorted from lower to higher response burden and permanent random numbers. Overlapping rate between two consecutive year samples is 50-70%, depending on changes in frame.

Data are collected electronically and a very small part by telephone.

Arithmetic and qualitative checks are used in the validation process, including checking that the population coverage and response rates are as required, comparison with the data of previous periods or other surveys and with administrative data sources.

The indicators to be published are calculated on the basis of the collected data. Weights are calculated for the statistical units which are used to expand the data of the sample survey to the population.

Time series are seasonally and calendar adjusted (i.e. working day effect is removed).

Seasonal adjustment of all time series was done with the aid of the program JDEMETRA+ and applying the method TRAMO/SEATS.

The time series were tested for ARIMA model, log-transformation, for correction of trading day and Easter effects, for outliers, and for decomposition specification.

To assure the quality of processes and products, Statistics Estonia applies the EFQM Excellence Model, EU Statistics Code of Practice and the ESS Quality Assurance Framework (QAF). Statistics Estonia is also guided by the requirements provided for in § 7. „Principles and quality criteria of producing official statistics” of the Official Statistics Act.

Statistics Estonia performs all statistical activities according to an international model (Generic Statistical Business Process Model – GSBPM). According to the GSBPM, the final phase of statistical activities is overall evaluation using information gathered in each phase or sub-process (this information includes, among other things, feedback from users, process metadata, system metrics and suggestions from employees). This information is used to prepare the evaluation report which outlines all the quality problems related to the specific statistical activity and serves as input for improvement actions.

Main users are Eurostat and other EU institutions. Other users are Ministry of Finance; Ministry of Social Affairs; Ministry of Economic Affairs; banks; scientific institutions; foreign and local employer’s associations; foreign entrepreneurs; trade unions; media and Statistical Office of Estonia itself. Above mentioned users are more interested in the average labour costs per hour worked and the average labour costs per employee. The LCI is still not well-known.

Since 1996 Statistics Estonia conducts reputation surveys and user surveys.

All results are available on the website https://www.stat.ee/en/statistics-estonia/about-us/user-surveys.

The Regulation (EC) No 450/2003 of the European Parliament and of the Council concerning the labour cost index is fully implemented.

The back data referred to Article 5 were transmitted in March 2005. At the same time the working-day adjusted and seasonally and working-day adjusted index series were transmitted.

100%

The type of survey and the data collection methods ensure sufficient coverage and timeliness.

Optional

Optional

See below

Since 2002 all labour cost items and hours worked are collected from the same sources (wages and salaries survey). All full-time and part-time employees are covered, except the employees of non-profit organisations with less than 50 employees and employees with the contract agreement (by civil code). The employees with the contract agreement are excluded because it is difficult to measure the working time of those employees and the Estonian Law does not require it to be measured.

Number of employees according to ESA 2010 is mainly based on the data of the labour force survey. Comparing data from wages and salaries survey and labour force survey (LFS), following differences between two surveys have to be considered.

• Due to the small sample size of the LFS the relative standard error by economic activities is relatively high, especially for quarterly results and smaller economic activities.
• In some cases the distinction between employees and self-employed can be different in LFS and wages and salaries survey. For example, if the owner of an enterprise, workshop, shop, office, etc., is also working as employee in own firm, he/she can be included as self-employed in LFS, but as employee in wages and salaries survey.
• Every person is included only once in LFS with his/her main job but in wages and salaries survey persons with more than one job are included multiple times.

Optional

Statistical population and sample size by economic activity is shown in the file attached

Statistics Estonia calculates standard deviation only for average labour cost and average monthly wages and salaries per employee. Labour cost index calculations are based on the sayed survey.

See below 6.3.4.1.

Optional

Is not measured

See below

Let us consider the part of the population which is investigated by sample survey, it is assumed that among the non-respondents of each stratum there are relatively as many economically active units as among the respondents of this stratum. The missing data of those assumed to be active non-respondents are imputed by the corresponding means of the active respondents. As a matter of fact, within a stratum equal response probability is assumed. Missing data of non-respondents belonging to the completely enumerated part of the population are imputed by using the hot deck method. If the number of employees of a non-respondent is bigger than 100, then the missing data of this unit are imputed by the data of the previous period of the same unit.

Procedures include also a logical checking process after the data entry process.

The table attached presents response rates in year 2022. Respondents include enterprises and institutions which have responded at least in one month of a quarter and enterprises and institutions which have informed us of their liquidation.

Optional

Seasonal adjustment of all time series was done with the aid of the program DEMETRA+ and applying the method TRAMO/SEATS.

Revisions are adjustments of statistical data which are:
a) planned

• regular (i.e. routine revisions)
• irregular (i.e. major revisions).

b) unplanned (correction of errors).

Regular revisions are related to the statistics production process – already published data are
revised upon the receipt of additional data (preliminary estimates are replaced with final values
derived from the data submitted by respondents). This is customary in certain subject areas
where preliminary data are published to ensure timeliness.
Irregular revisions are related to methodological improvements – definitions and classifications
are changed, the base year is changed, new data sources are introduced, etc. Irregular
revisions have to be scheduled so that several changes are published at once,
i.e. irregular revisions should be as rare as possible.

The data revision policy and notification of corrections are described in the dissemination policy of Statistics Estonia http://www.stat.ee/dissemination-policy.

The unadjusted labour cost indices by NACE Rev. 2 sections were not revised.

The final results were published and they were not revised after the publication for 2022.

Optional

The quarterly data of LCI of the year 2022 were transmitted by Statistics Estonia in compliance with article 6.2 in the Council Regulation (EC) No 450/2003 i.e. 70 days of the end of the reference period. No delays.

See above

See above

There was no time lag between the actual delivery of the data and the target date when it should have been delivered.

The data have been published at the time announced in the release calendar.

Internal coherence

The definitions and classifications used in Estonia comply with international definitions and classifications ensuring international comparability of results.

Optional

Since 2009 we used the reference year 2008=100. Using the respective proportions of the survey of 2002 and current quarterly wages and salaries data, the other labour costs and total labour costs of 2000 and 2001 were estimated by quarters. Hourly labour cost and other hourly labour cost items were calculated.

Since 2002 the other labour costs are measured, i.e. not estimated. Before the year 2002 the social tax was a single one kind of the employer’s statutory social security contributions, which formed 33% of the payroll. The Estonian law was changed and the unemployment security tax, which is a new kind of employer’s statutory social security contributions, was introduced in 2002.

Since 2002 the unemployment security tax is included in the other labour cost item, which formed 0.5% of the payroll. From January 2006 to May 2009, the unemployment security tax rate formed 0.3% of the payroll. From June to July 2009, the unemployment security tax formed 1% and from August 2009 to December 2012 it formed 1.4% of payroll. Since January 2014 it forms 1% of payroll. Since January 2015 it forms 0.8%.

Since the 1st quarter of 2018, changes have been made in the population and sample: the sample was optimised and non-profit organisations with less than 50 employees were included in the population.

Since the beginning of 2019, all data from 2000 Q1 were recalculated on the new basis (2016=100) and sent to Eurostat.

Since the 1st quarter of 2020, changes have been made in the population and sample: state and municipal organisations and institutions with less than 50 employees are not included in completely surveyed part.

Data on LCI are available since 2000 Q1. The data are comparable over time.

See 8.5

All NACE Rev. 2 sections are surveyed using the same methodology.

Optional

The working time account of ESA 2010 is mainly based on the data of the labour force survey. Comparing data from wages and salaries survey (LCI) and labour force survey (LFS), the following differences between two surveys have to be considered.

Due to the small sample size of the LFS, the relative standard error by economic activities is quite high, especially for quarterly results and smaller economic activities.

• Due to the hours worked of the LFS, the annual unadjusted growth rates of the ESA 2010 compensation of employees per hours worked by NACE Rev. 2 sections is quite different between years. There are large discrepancies when comparing the unadjusted growth rates presented in the table attached.

The outputs of the statistical activity are coherent.

Time series of labour cost indices starts with the 1st quarter of 2000 on Statistics Estonia website. Users are more interested in the average labour costs per hour worked and the average labour costs per employee. A description of the methodology and sources used is published in the Statistical Database https://www.stat.ee/en/find-statistics/methodology-and-quality

Not published

Not published

Data are published under the heading „Economy/ Wages and salaries and labour costs” in the Statistical Database in https://andmed.stat.ee/en/stat

Optional

The dissemination of data collected for the purpose of producing official statistics is guided by the requirements provided for in § 34, § 35, § 36, § 37, § 38 of the Official Statistics Act. Access to microdata and anonymisation of microdata are regulated by Statistics Estonia’s procedure for dissemination of confidential data for scientific purposes: http://www.stat.ee/dokumendid/51669.

Not used

Information on definitions and methodology can be found at Statistics Estonia website in database section Economy/Wages and salaries and labour costs/Labour costs indices /"table"/select "definitions and methodology".

Regulation (EC) No 450/2003 of the European Parliament and of the Council of 27 February 2003 concerning the labour cost index

http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2003:069:0001:0005:EN:PDF

Commission Regulation (EC) No 1216/2003 of 7 July 2003 implementing Regulation (EC) No 450/2003 of the European Parliament and of the Council concerning the labour cost index

http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CONSLEG:2003R1216:20090101:EN:PDF

Commission Regulation (EC) No 224/2007 of 1 March 2007 amending Regulation (EC) No 1216/2003 as regards the economic activities covered by the labour cost index

http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2007:064:0023:0024:EN:PDF

Data on the population, sample and respondents, and the error estimates are planned to be published in the Statistical Database.

Optional

Optional

Total time of filling in guestionnaires of the statistical activity, working days: 3499

Average time of filling in the questionnaires, hours per questionnaire: 0,72

The dissemination of data collected for the purpose of producing official statistics is guided by the requirements provided for in § 34 and § 35 of the Official Statistics Act.

The treatment of confidential data is regulated by the Procedure for Protection of Data Collected and Processed by Statistics Estonia.

No comment.