|
|
For any question on data and metadata, please contact: Eurostat user support |
|
|||
1.1. Contact organisation | Central Statistical Bureau of Latvia |
||
1.2. Contact organisation unit | Wage Statistics Section & Mathematical Support Division |
||
1.5. Contact mail address | Lacplesa Street 1 Riga, LV-1301 |
|
|||
2.1. Data description | |||
The Labour Cost Survey (LCS) is conducted every four years for a full calendar year. The survey is carried out in accordance with Council Regulations 530/1999, 1726/1999 and 1737/2005 and provides comprehensive and detailed information comparable at the EU level on labour costs, structure and development in various sectors of economic activity. The main indicators of the survey: number of employees, hours worked and paid, labour costs. LCS covers companies with at least 10 employees and all economic activities defined in sections B–N and B–S of the Statistical Classification of Economic Activities of the European Communities (NACE Rev. 2). Transmission of data on small enterprises (fewer than 10 employees) and enterprises belonging to NACE Rev. 2. For section O, optional. |
|||
2.2. Classification system | |||
NACE Rev. 2 |
|||
2.3. Coverage - sector | |||
NACE Rev 2 sections B-S. NACE Section O is optional, but it is also covered. |
|||
2.4. Statistical concepts and definitions | |||
All the concepts and definitions are in line to regulation 1737/2005 and other in force. — Total annual labour costs, distinguishing wages and salaries (broken down into direct remuneration and bonuses, payments to employees' savings schemes, payment for days not worked and wages and salaries in kind), the employer's social contributions, (broken down into actual and imputed social contributions), vocational training costs, other expenditure and taxes, and also subsidies directly related to labour costs, — The average annual number of employees, distinguishing fulltime employees, part-time employees, and apprentices, — The annual number of hours worked and the annual number of hours paid, in each case distinguishing full-time employees, part-time employees, and apprentices. |
|||
2.5. Statistical unit | |||
Enterprise |
|||
2.6. Statistical population | |||
Enterprises registered in Statistical Business Register (SBR) of Central Statistical Bureau of Latvia. The LCS applies to all activities in sections B-S (including section O, although it is optional) of NACE Rev.2 and represents all statistical units. |
|||
2.7. Reference area | |||
Latvia |
|||
2.8. Coverage - Time | |||
2020 |
|||
2.9. Base period | |||
not applicable |
|
|||
3.1. Source data | |||
The survey results are compiled from the labour survey and also use administrative data. |
|||
3.2. Frequency of data collection | |||
Every four years. |
|||
3.3. Data collection | |||
Enterprises could submit survey forms electronically using the CSB electronic system, where the form is already automatically integrated at the beginning of the year. |
|||
3.4. Data validation | |||
Mathematical and logical errors in the data are checked automatically at the moment they are received through the electronic system. Respondents see errors immediately and must correct them, otherwise the system cannot accept the report. After receiving the questionnaires, they are also checked by statisticians. In the future, the data are reviewed at the time of their processing by CSB experts. |
|||
3.5. Data compilation | |||
The survey results are compiled from the labour survey and also use administrative data. |
|||
3.6. Adjustment | |||
Not applicable. |
|
|||
4.1. Quality assurance | |||
Central Statistical Bureau Republic of Latvia works on the basis of several sources of quality requirements. |
|||
4.2. Quality management - assessment | |||
[Not requested] |
|
|||
5.1. Relevance - User Needs | |||
The main external users of the data of Labour Costs Survey (LCS) are Eurostat and other EU institutions, International Labour Organization, labour market research institutions as well as foreign companies. As the main national users should be mentioned the Central Bank of Latvia, governmental bodies - the Ministry of Finance and the Ministry of Economy, the National Accounts Department and other departments of Central Statistical Bureau (CSB), the Latvian Confederation of Employers, as well as mass media. |
|||
5.2. Relevance - User Satisfaction | |||
CSB has not received much information on the national and other data users satisfaction with the data offered by this survey. |
|||
5.3. Completeness | |||
[Not requested] |
|||
5.3.1. Data completeness - rate | |||
[Not requested] |
|
|||||||||||||||||||||||||||||||
6.1. Accuracy - overall | |||||||||||||||||||||||||||||||
1. Target population CSB of Latvia did not carry out a separate survey for LCS 2020, but main data source that contributed to the LCS 2020 was quarterly Survey on Employment and Wages (Nr. 2 ‑ LABOUR). Data in quarterly Survey on Employment and Wages were collected through a combination of census, sampling and information from the State Revenue Service (SRS) of Latvia. Enterprises that were covered by complete enumeration were:
Enterprises with 10-49 employees were covered through sampling. Small enterprises with less than 10 employees were not sampled, but information from State Revenue Service was used.
N – not sampled, but SRS data used
The official, up to date, Statistical Business Register (SBR) of Central Statistical Bureau was used for building up the sampling frame. Frame consisted of 110 184 units, the number of units in the sample 6 182, the number of units in the sampling frame 12 409.
2. Sampling design The sampling design for LCS 2020 survey was stratified simple random sample. Enterprises were stratified by three variables, namely, the size class of enterprise, the ownership form and economic activity by NACE Rev. 2 classification. Neyman optimal allocation was used to allocate units to strata which was slightly corrected, so that the design weights were not greater than 6. 3. Weights calculation method Design weights were calculated as reciprocals of inclusion probabilities, dividing the population size by sample size in each stratum.
To cope with the problem of changes in the target population, every quarter actual sample was updated adding:
Weights were adjusted for overcoverage, undercoverage and unit nonresponse, dividing target population size by the number of respondents in each stratum excluding enterprises covered by complete enumeration. The weights of enterprises with inclusion probability <1 were calibrated using auxiliary information from SRS data. The calibration was done in NACE Rev.2 divisions. Two auxiliary variables (the number of employees and the number of enterprises) were used for weight calibration. |
|||||||||||||||||||||||||||||||
6.2. Sampling error | |||||||||||||||||||||||||||||||
The coefficient of variation taking into account the sampling desing was calculated for annual labour costs and annual hourly labour costs. |
|||||||||||||||||||||||||||||||
6.2.1. Sampling error - indicators | |||||||||||||||||||||||||||||||
Tables in annex "Coefficients of variation" show estimates of coefficient of variation for annual labour costs and annual hourly labour costs. The precision estimation is done by the ultimate cluster method (Hansen, Hurwitz and Madow, 1953) with Taylor linearization for non-linear statistics and residual estimation from the regression model to take weight calibration into account. Annexes: Coefficients of variation |
|||||||||||||||||||||||||||||||
6.3. Non-sampling error | |||||||||||||||||||||||||||||||
See below |
|||||||||||||||||||||||||||||||
6.3.1. Coverage error | |||||||||||||||||||||||||||||||
Since under-coverage is difficulty to obtain this indicator is not calculated. |
|||||||||||||||||||||||||||||||
6.3.1.1. Over-coverage - rate | |||||||||||||||||||||||||||||||
Over-coverage relates to units presented in the frame which do not belong to the target population. These units were economically non-active enterprises (‘‘sleeping’’ or died) at the reference quarter of the survey. The over-coverage rate has been calculated as ratio of the number of non-active units over number of all units included in the sample. Over-coverage rates
The average overall over-coverage rate was 2.5%. |
|||||||||||||||||||||||||||||||
6.3.1.2. Common units - proportion | |||||||||||||||||||||||||||||||
[Not requested] |
|||||||||||||||||||||||||||||||
6.3.2. Measurement error | |||||||||||||||||||||||||||||||
The questionnaire for LCS 2020 was designed in such a way to eliminate the survey instrument (questionnaire) errors providing explanatory notes directly in questions and detailed explanations in the supplemented instruction for filling in the questionnaire.
The LCS 2020 data processing had been carried out using the Integrated Statistical Data Management System (ISDMS) where statistical metadata is the key element. The metadata module ensures maintenance of validation rules of statistical survey. Each validation rule description contains validation rule code, error message text and description of validation rule, validation rule conditions. It is possible to generate validation procedure automatically for each questionnaire directly after the entry of questionnaire. Main validation rules, both arithmetical and logical, were available during filling in e-questionnaire via website of CSB.
The validation programme consisted of 373 arithmetical and logical controls. After executing the validation procedure of respondent survey, data form with list of validation errors is available. This form contains the following information for each error – error number, error description, type (acceptable, unacceptable, not marked), error type reason (for some errors it is necessary to describe the acceptability reason). Special attention is paid to comparison of the values with the corresponding data of previous period (quarter, month).
Every statistician had a definite number of enterprises from which the questionnaires have to be collected, entered and verified. The responsible person could not finish the data entry of the questionnaire in case if any answer on the questionnaire’s variable was missed, or (according to the rules implemented in the data entry programme) filled in incorrectly. In these cases the responsible person contacted the enterprises once more and made the necessary changes in the questionnaire.
The variables that have been corrected most often were statutory social security contributions because in questionnaire those are only asked for employees having registered working time, but in accounts of bookkeepers contributions are calculated for all employed in unit. The question usually not clear for respondents is non-worked but paid hours.
In common with the previous surveys the most difficult question was hours worked for employees of specific occupational groups such as teachers, scientists, artists. Taking into consideration that this variable is very important statisticians gave it particular priority. Despite the detailed instruction provided respondents often asked about criteria for determining if an employee is full- or part-timer, about regular and non-regular payments. Some difficulties had arisen to define whether expenditure is labour cost or intermediate expenditure. |
|||||||||||||||||||||||||||||||
6.3.3. Non response error | |||||||||||||||||||||||||||||||
The overall un-weighted unit response rate (ratio, expressed as a percentage of the number of respondent units used in estimation over the total number of in-scope respondents) was 93.7%. |
|||||||||||||||||||||||||||||||
6.3.3.1. Unit non-response - rate | |||||||||||||||||||||||||||||||
The annual number of non-respondents was 481 enterprises. The 5% of those were enterprises not found; refused to submit questionnaire or delayed to do that comprised 46 % of non-respondents.
The reasons of non-response in percentage of the sample
|
|||||||||||||||||||||||||||||||
6.3.3.2. Item non-response - rate | |||||||||||||||||||||||||||||||
[Not requested] |
|||||||||||||||||||||||||||||||
6.3.4. Processing error | |||||||||||||||||||||||||||||||
The methods used to reduce size of the errors resulting from non-response were imputation and compensatory weighting.
The imputation has been carried out by combining automated and manual methods. The centralized automated imputation based on the calculating of mean value of the corresponding strata and number of employees as well as gross wages provided by the database of the State Revenue Service was applied for the large private and public enterprises and budgetary organizations. The other methods used – manual logical imputation of missing values during the period of data collection, manual imputation on the base of data available from previous period of the same survey or from other surveys. For small enterprises, which had not submitted the questionnaire, the weighting coefficients in the corresponding strata were changed. |
|||||||||||||||||||||||||||||||
6.3.4.1. Imputation - rate | |||||||||||||||||||||||||||||||
Imputation rate for units in population for which data are derived from administrative data sources – 60.8 %. |
|||||||||||||||||||||||||||||||
6.3.5. Model assumption error | |||||||||||||||||||||||||||||||
There was no problem to use a calendar year for the Labour Costs Survey 2020. The survey covered small enterprises (size class <10 employees).
For the collection of labour costs data the quarterly Survey on Employment and Wages (Nr. 2-LABOUR) as well as administrative databases of the State Revenue Service for enterprises with number of employees from 1 to 9 were used. |
|||||||||||||||||||||||||||||||
6.4. Seasonal adjustment | |||||||||||||||||||||||||||||||
[Not requested] |
|||||||||||||||||||||||||||||||
6.5. Data revision - policy | |||||||||||||||||||||||||||||||
[Not requested] |
|||||||||||||||||||||||||||||||
6.6. Data revision - practice | |||||||||||||||||||||||||||||||
[Not requested] |
|||||||||||||||||||||||||||||||
6.6.1. Data revision - average size | |||||||||||||||||||||||||||||||
[Not requested] |
|
|||
7.1. Timeliness | |||
Data on the LCS 2020 were transmitted to Eurostat in June 2022. |
|||
7.1.1. Time lag - first result | |||
[Not requested] |
|||
7.1.2. Time lag - final result | |||
[Not requested] |
|||
7.2. Punctuality | |||
The data that was subsequently used for the LCS was collected in accordance with the quality policy of the CSB within the time frame established by the internal rules. The LCS data was sent on time at the end of June 2022, however, after checking the data by Eurostat, an additional data transfer was required due to the verification of part of the data. Due to the timing of employee vacations and a mutual agreement with Eurostat, the final LCS data was sent on August 4, 2022. |
|||
7.2.1. Punctuality - delivery and publication | |||
[Not requested] |
|
|||
The coherence is analyzed using the data of all enterprises (including enterprises with number of employees < 10). |
|||
8.1. Comparability - geographical | |||
For LCS 2020 enterprises were used as sampling units instead of local units as defined in the Regulation.
There is no difference in the reference population and between national concepts and European concepts.
The CSB of Latvia uses the following international classifications:
The results on labour costs were transferred according to the European concepts. |
|||
8.1.1. Asymmetry for mirror flow statistics - coefficient | |||
[Not requested] |
|||
8.2. Comparability - over time | |||
LCS was conducted for the first time in 2004. All previous LCS and LCS 2020 were based on the quarterly Survey on Employment and Wages (2-LABOUR). The regular questionnaire of this survey had been supplemented by additional questions to provide all variables according to the Regulation. The stratification was made according to the groups of enterprises, the main line of economic activity of enterprise and number of employees to ensure the requirements of regulations for LCS as well as LCI, STS and national needs.
In all surveys the surveyed unit was enterprise instead of local unit. In order to avoid respondent’s burden and increasing of CSB costs the unit was not changed and tailor made survey was not carried out. |
|||
8.2.1. Length of comparable time series | |||
[Not requested] |
|||
8.3. Coherence - cross domain | |||
1. Coherence with statistics from Labour Force Survey (LFS) Graph and table in annex "Comparison LCS and LFS" shows differences between average hours actually worked per year by employee in all job LCS and in the main job LFS. The figures of LFS are higher than those in LCS in all sections of economic activities. The biggest gaps between both sources are in NACE S and I, the smallest – in K and O.
The differences can be explained by: 1) different surveying units: in LCS the observed unit is the enterprise, in LFS - household members aged 15-74. In the LCS the sampling frame of enterprises is based on the Statistical Register of Economic Entities, in LFS - private households; 2) in LCS individual enterprises (physical persons not registered in the Sate Business Register) are excluded, in LFS all employed persons irrespective of what kind of enterprise they are working for are covered; 3) hours worked per employee in LFS are calculated for employees in the main job, for LCS – for all jobs (full-time or part-time together); 4) results of LFS show also impact of hidden economy. Persons surveyed in LFS give self-perception of their status or inform about the real working hours, while the official status, in accordance with working agreement, is different. In LCS enterprises provide official figures reported also to the State Revenue Service; 5) sample size of LFS - 29 952 dwellings per year, meanwhile sample size of LCS - 6 182 enterprises (this fact influences also the results).
2. Coherence with Structural Business Statistics (SBS) Comparing wages and salaries per employee at NACE Rev.2 sections level in LCS and in SBS it is clear that both sources correspond well (see table and graph in annex "Comparison LCS and SBS"). The biggest differences between figures is stated in NACE D, M and N. In NACE D figures of SBS are higher than those in LCS, whereas in section N figures are higher in LCS.
The distinctions can be explained by: 1) in spite of the base for sample of LCS and SBS is the same – the Statistical Business Register – there are some differences for inclusion of commercial companies in the sample. In LCS all private companies with number of employees 50 and more are surveyed 100%, in SBS the other criterion is also used - level of turnover, as well as some specific branches. In 2020 mostly all enterprises of SBS with number of employees under 20 were not surveyed but imputed using administrative data sources. In LCS 2020 the administrative data sources were used for private enterprises with number of employees under 10; 2) SBS covers only commercial companies while LCS surveys also budgetary institutions, foundations, associations and funds; 3) employees for whom it is impossible to determine hours worked and paid are not included in results of LCS, but they are covered by SBS. Annexes: Comparison LCS and LFS 2020 Comparison LCS and SBS 2020 |
|||
8.4. Coherence - sub annual and annual statistics | |||
Coherence with Labour Costs Indices (LCI) The graph and table in annex "Comparison LCS and LCI" below shows growth rates of hourly labour costs in LCS 2020 in comparison with the previous survey in 2016 and growth rates of un-adjusted labour costs indices estimated for the same period. Since the data source for LCS 2020 and LCI is the same differences of growth rates are small. Those can be caused only by different variables included in calculation of hourly labour costs. Vocational training costs (D.2) as well as other labor costs (D.3) such as recruitment costs, working clothes provided by employer are not included in estimation of LCI, but are included in calculation of hourly labour costs in LCS. Annexes: Comparison LCS and LCI 2020 |
|||
8.5. Coherence - National Accounts | |||
The total difference of compensation of employees D.1 comprises 25.4%. The figures of NA are higher than those in LCS in many sections of economic activities. The biggest gaps between both sources are in NACE sections L and I, the smallest – in K and O. The reasons for differences: 1) There is no consistency between number of employees in NACE sections and compensation of employees in NA data. Main data source for employment in NA is Labour Force Survey, covering not only paid employees but also self-employed persons, as well as annual surveys (SBS and annual survey of budgetary institutions) providing information on employment by kind of activity units. Main data source for compensation of employees is the same as for LCS – quarterly Survey on Employment and Wages (2-LABOUR); 2) the definition of employees in both sources is different. In NA, the number of employees includes full-time and part-time employees in the main job, women on maternity or child care leave until three years, sales representatives, agents and other employees not on payroll lists as well as conscripts. The LCS covers all employees in payroll list and having registered working hours; 3) the compensation in NA includes employment costs of outworkers, of employees who perform tasks for enterprises on the basis of order agreements, commission wages and salaries; 4) NA estimates include also hidden economy, which is not taken into account in the LCS results.
Annexes: Comparison LCS and NA |
|||
8.6. Coherence - internal | |||
[Not requested] |
|
|||
Quarterly and annual data are available in databases of the website of CSB supplemented by metadata as well as in annual leaflet ‘‘Labour Costs’’ (in Latvian). |
|||
9.1. Dissemination format - News release | |||
|
|||
9.2. Dissemination format - Publications | |||
Publications: |
|||
9.3. Dissemination format - online database | |||
9.3.1. Data tables - consultations | |||
[Not requested] |
|||
9.4. Dissemination format - microdata access | |||
[Not requested] |
|||
9.5. Dissemination format - other | |||
[Not requested] |
|||
9.6. Documentation on methodology | |||
Methodology: Labour costs and their structure |
|||
9.7. Quality management - documentation | |||
[Not requested] |
|||
9.7.1. Metadata completeness - rate | |||
[Not requested] |
|||
9.7.2. Metadata - consultations | |||
[Not requested] |
|
|||
|
|
|||
11.1. Confidentiality - policy | |||
[Not requested] |
|||
11.2. Confidentiality - data treatment | |||
[Not requested] |
|
|||
N/A |
|
|||
|
|||