STATISTICAL OFFICE OF THE SLOVAK REPUBLIC

Labour and Education Statistics Department

Lamacska cesta 3/C,

P.O.BOX 17,

840 05 Bratislava,

Slovak Republic

Methodology corresponds to the target methodology of the European Labour Cost Index. It fully complies with the Regulation (EC) No 450/2003 of the European Parliament and of the Council of 27 February 2003 concerning the labour cost index (LCI) that establishes a common framework for the production and transmission of comparable labour cost indices in the Community. Labour Cost Index shows the short-term development of the total, direct, indirect or less irregular bonuses hourly costs for employers of employing the labour force in the Slovak Republic. It measures average labour costs in eur per hour worked in reference quarter. According to the target methodology, the indices have been calculated since 2023 with year 2020 = 100 (2016 = 100 before 2022; the base year changes in 4-year cycles).

1 digit level of NACE Rev. 2

Enterprises with 1 or more employees operating in the NACE Rev. 2 sections B to S are covered.

In the context of the LCI, labour costs are defined as expenditures borne by employers for the purpose of employing staff. They include employee compensation, with wages and salaries in cash and in kind, employers' social security contributions and employment taxes regarded as labour costs minus any subsidies received, but not vocational training costs or other expenditure such as recruitment costs and spending on working clothes (by contrast with multiannual and annual labour cost data). 

The quarterly LCI measures short-term trends in "average hourly labour costs", defined as labour costs divided by the corresponding number of hours worked in the quarter in question. Quarterly changes in hourly labour costs are calculated first for each section B to S by NACE Rev. 2 and then aggregated to the groups of sections  B-N, B-S, B-E, B-F, G-N, G-J, K-N, O-S and P-S keeping fixed weights for one year. All labour cost indices are annual chain-linked Laspeyres indices. 

Apart from the overall Labour Cost Index (LCI_TOT), indices are also available for the labour cost components "wages and salaries (LCI_WAG)", "employers' social security contributions plus taxes paid minus subsidies received by the employer (LCI_OTH)" and for "index excluding bonuses (LCI_TXB)".

The statistical unit is the enterprise. 

LCI applies to all activities in sectors sectors B to S of NACE Rev. 2 and represents all statistical units.

The whole country is covered by the data disseminated.

Data are available since the 1^st quarter of 1996 for NACE Rev 1.1 (until the 4^th quarter of 2008)  and since 1^st quarter of 2000 for NACE Rev 2.

The labour cost indices use the year 2020 as base year since 2023. The previous NACE Rev. 2 series for quarters prior to 2023 are available with the year 2016 and before 2019 with the year 2012 as base period. Historical NACE rev. 1.1 series are available with the year 2000 as base year.

Since 2008, the combination of two data sources has been applied as follows: 

• Enterprise surveys on labour

Data on earnings, statutory contributions to insurance schemes and hours worked is obtained from a quarterly full-scale survey (covering a complex of the quarterly enterprise surveys) on labour in organizations of financial intermediation and in all non-profit organizations irrespective of the number of employees as well as from a quarterly sample survey in profit organizations with 20 and more employees and in those with up to 20 employees which have a turnover of 5 million eur and more. Data for other small-size enterprises with up to 20 employees is obtained from a specific quarterly sample survey. Results are grossed up to the population.

Sample frame:  Enterprises surveyed are drawn from the Statistical Business Register administrated by the SO SR                             

                         („the Register“).

Sample size and design:

(1)  Full-scale quarterly survey for organizations of financial intermediation and  non-profit organisations (about 14 thous. units).

(2)  Sample quarterly survey for about 50 % sample of profit organizations with 20 and more employees and organizations with up to 19 employees, whose turnover has been 5 million eur and more (about 4 thous. units).

(2)  Sample quarterly survey for about 7 % sample of small-size organizations up to 19 employees, whose annual turnover has been less than 5 million eur (about 14 thous. units).

• Labour Cost Survey

Data on other labour cost components is obtained from the annual Labour Cost Survey. Covered are all local units regardless their economic activity, region, size, type of ownership and legal form. Sample in 2021 consisted of 9 221 statistical units that represents about 7% of the sample frame established on the base of the Register. Units with 100 and more employees are fully included. Collected and processed information concerned about 970 thous. employees (about 41% of the total employment in Slovakia). Results are grossed up to the population.

The source data are collected quarterly.

Collection, recording, checking and processing of input data is organized by the staff of the Statistical Office of the Slovak Republic. Data collection runs by making use of electronic questionnaires.

Data validation is undertaken by the staff of the Statistical Office of the Slovak Republic at the levels of collection, processing and disseminating data. Sets of global checks are used to validate completness of information, correct coding and formats in each of the surveys. Plausibility checks are used to verify correctness of values and logical relations among variables. If errors in completeness or reliability of the data file are apparent, the reporting units are contacted to complete or correct information.

Labour cost indices are calculated in a few consistent steps.

The first step to calculate the labour costs indices is the aggregation of quarterly data for all size groups of enterprises from the enterprise surveys on labour. The measured indicators are the total amount of gross wages and salaries, statutory social security contributions and the total number of hours worked classified by NACE Rev. 2.

Data on the collectively agreed, contractual and voluntary social security contributions, employers´ imputed social contributions and employers´ social contributions for apprentices are available from the Labour Cost Survey. These annual data, classified by economic activities, are recalculated to the quarters by means of the share of the quarterly statutory social security contributions to their annual amounts.

Estimation of the quarterly data on wages and salaries in kind, wages and salaries of apprentices, taxes and subsidies is made by recalculating the annual data from the Labour Cost Survey, classified by economic activities, using the shares of the quarterly wages to their annual amounts.

Seasonally and working-day adjusted data are also made available. 

LCI data are checked for completeness and consistency. Specific quality assurance activities comprising compliance monitoring, benchmarking and training are used.

Annual quality reports are compiled. Overall, the degree of compliance with the Regulation has increased since launching the Phare project in 2000. Currently, it complies with the quality requirements stressed in this field of statistics.

The key user obtaining data on the labour cost index (LCI) directly from the Statistical Office of the Slovak Republic (SO SR) is Eurostat. Other users who are interested in labour costs are ministries, financial institutions, employer´s associations, trade unions, research institutions and media.

Currently, there is no information on any lower level of user satisfaction concerning completeness (units, variables, breakdowns), accuracy or timeliness of the LCI.

The national concept of the LCI fully complies to Regulation (EC) No 450/2003.

The labour cost indices and weights were transmitted to Eurostat through GESMES/TS and STADIUM in the period 2005 – 2007 and through eDAMIS in the period 2008 – 2014q2. The reporting moved to the ESS Metadata Handler after testing in July 2014.

100% of data on labour cost indices and weights requested by the Regulation (EC) No 450/2003 is provided.

The accuracy of LCI depends on accuracy of data sources which it is compiled from.  Major part of labour costs and total volume of time worked have been taken from the quarterly enterprise surveys on labour. Remaining components come from results of the annual Labour Cost Survey, which are available later, therefore the LCI data are subject to revisions. The overall accuracy of basic data for LCI is satisfactory, particularly as regards big-size enterprises. Imputation method or adjustment of the weighting method is used for correction of unit non-response. Extent of revisions used to be generally small.

Optional

Optional

See below

Table 1 in the Annex shows the number of business units in the sample and the sample share - percentage of the business units in the sample (quarterly enterprise surveys) from the reference population (Business Register). It is broken down by size class. Sample fractions ranged between 10.7% - 10.9% for small-size units with less than 10 employees and 58.0% - 58.4% for other units over all quarters in 2022. The average sample size in 2022 was 30 044 units.

Optional

The proportion of common units can not be calculated since no administrative source is used to compile LCI.

Measurement errors occured more often at the beginning of the LCI data collection. It resulted mainly from the uncertainty of responding units how to fill in the questionnaires. Therefore, the explanatory notes, which are inseparable and important part of the questionnaires, were improved to clarify those issues (eg. definitions of variables were described in more detail to improve their clarity). Since that, the total amount of respondents’ errors resulted from wrong data inputting have been reduced. Main errors made by respondents refer to wrong aggregations of labour cost components, infilling of some variables in questionnaire, wrong calculation of variables in questionnaire and incorrect filling in the identification codes of units for economic activities. These errors have been identified and corrected through complex of software controls before data processing.

The electronic statistical data collection system was introduced in 2016. Responding units are obliged to provide data by making use of Internet, where first set of logic controls warn them against providing an incorrect or information. Regional statistical offices check and re-contact responding units in case of missing or dubious information (eg. if any inconsistencies occur concerning time series or relationships between variables). In case of mandatory checks all errors should be rectified while in case of information controls all finding should be verified at first and correct if necessary. A special attention used to be given during the periods of changes in the labour legislation to reduce the measurement errors. In order to avoid misunderstanding, the explanatory notes in the questionnaires are being carefully amended and consultation is being provided to responding units to clarify the issue.

The response rate in majority of quarterly enterprise surveys as well as in Labour Cost Survey exceeds 80 %.

According to the table 2 in the Annex a better response rate in the quarterly enterprise surveys was in big-size units. Best values were in services under education and human health due to full-size quarterly survey and dutiful approach of state bodies to statistics. High response rate over 90% in big industrial enterprises was very important as well. Total big-size unit response rate was 94.2 %, i.e. 15.6 pp. higher comparing to small-size units.

Missing questionnaires in the reference quarter for the active responding units, that delivered data in the previous quarter, were imputed by making use of the “hot-deck” method of the donor from the present period. The donor is an active responding unit, which had delivered correct completed questionnaire or it had not been identified any extreme or another problem with quality in the checking process. Selection of the comparable units was made on the basis of three common attributes:

• economic activity at the same 2-digit code of NACE Rev.2,
• the same code in “Categories of organizations by number of employees,
• “Added Value” indicator with keeping the same volume of production and of intermediate goods.

Item non-response has not appeared, since all missing data were consulted with the enterprises and corrected in the questionnaires.

See below

Missing quarterly information:

• some labour cost items

Methods used to estimate the missing information:

Quarterly data sources do not provide information on payments to employees saving schemes, wages and salaries in kind, wages and salaries of apprentices, collectively agreed, contractual and voluntary social security contributions, employers´ imputed social contributions, taxes and subsidies. 

At time of the labour cost index calculation, the missing variables mentioned above are taken from the annual Labour Cost Survey. Results of the survey relate to the period of 2 years ago. Share of the missing  items is relatively small (about 2 % of the total labour costs) and their progress mostly depends on trend of direct remuneration and bonuses. Due to this fact the disposable data from the latest LCS is extrapolated by making use of the growth rates of the current direct remuneration and bonuses. In order to keep coherence, data from the LCS is taken as ratio to  direct remuneration and bonuses. Absolute values are calculated on the basis of data on direct remuneration and bonuses from quarterly enterprise survey on labour, which is used as a benchmark.

Table 3 in the Annex demonstrates that primary source, the enterprise surveys on labour, covers 97.9 % of  information on total labour costs. Remaining components have been taken from the Labour Cost Survey.

Data on hours worked and data on labour cost items used for the LCI calculation is in consistency since it is obtained from the same primary source – the quarterly enterprise surveys on labour. Data on hours worked has a high degree of accuracy since it is recorded from company accounts according to definition which complies to the target definition used in a community labour cost statistics. Hours worked represent hours actually worked by employees during normal working hours and overtime in full-time jobs and part-time jobs. Time spent at the place of work on tasks concerning work preparation, refreshment breaks as well as paid time on business trips is included. Paid, but not worked time, such as paid annual holidays, public holidays, sick leave etc. is excluded.

Table 4 and graph 1 in the Annex show a change in trend of hours worked in manufacturing, which is the cornerstone of the Slovak economy, in the 3rd quarter of 2008 towards decrease due to the global economic and financial crisis. It had a negative impact on the LCI trend. The bottom of the volume of time worked was reached in the middle of 2009. Since the 2nd quarter of 2010 volume of time worked has been growing with positive impact on trend of total labour cost index. Annual growth rates of hours worked have returned to positive figures in the 2nd quarter of 2010. However, tendency in 2011 was slowed down and even reversed in 2012.  After year-on-year decline during 3 quarters of 2013 the situation in time worked positively turned at the end of the year and stabilized in 2014. Due to an economic growth in 2015 - 2018 the positive trend of hours worked continued. The slight economic slowdown in 2019 caused the hours worked in the industry did not reach the peak value of the previous year.

In 2020, due to a pandemic Covid-19, time worked in manufacturing fell sharply. The worst situation was in the 2nd quarter, when during the lockdown the volume of hours worked was 22.7% lower than a year ago. In other quarters, the decline was between 6 - 9%. In 2021, the quarterly number of hours worked were mostly lower in year-on-year comparison. The exception was a 15.7% increase in the 2nd quarter due to a low year-ago data base affected by the restrictive lockdown.

No data from administrative sources has been applied in the period 2008 – 2019 and from 2022q3. In the period 2020q2 – 2022q2,  the LCI_OTH growth rates were reduced in the majority of economic activities since data on government refunds treated as subsidies for wages due to the Covid-19 epidemic were deducted from social contributions. Data on government refunds treated as subsidies for wages due to a pandemic Covid-19 has been provided by the Central Office of Labour, Social Affairs and Family.

Imputation method or adjustment of the weighting method is used for correction of unit non-response.

The initial weights were calculated as N_h/n_h, where N_h was the total number of enterprises in stratum “h” of the population and n_h was the number of enterprises in the sample in stratum “h”, assuming that each unit in the stratum had the same inclusion probability. The initial weights were updated after collection of questionnaires. Units with extremely high or low value of turnover got the weight equals 1 and other units got the weight equals N_h’/n_h’, where N_h’ was total number of enterprises in stratum of the updated frame minus the number of enterprises with weight equals 1.

Weights were adjusted by reweighting or combination of imputation and reweighting due to non respondent units.

The original data are provided unadjusted, working-day adjusted, seasonally and working-day adjusted. Before 2014 data were adjusted by means of DEMETRA software using TRAMO/SEATS method. In 2014, the DEMETRA+ software had been introduced, which was replaced by JDEMETRA+ in the course of 2015. Direct method of seasonal adjustment has been replaced by an indirect method in 2017. From 2000 - 2022 the time series comprised 92 observed periods.  The file documenting the seasonal adjustment procedures and models used in the LCI is attached as the Quality report for LCI seasonal adjustment (SA) under Annex QR LCI2022 on SA.  

The Law on State Statistics defines the secrecy and confidential data protection. Without the approval of legal or physical persons providing the relevant individual data, these data cannot be published or announced to anybody or used for other than statistical purposes.

All data is preliminary until updating. Revisions are being made in dependence on availability of the annual Labour Cost Survey. Results are used to be disseminated 12 months after the end of the reference year. This data source is used for obtaining information on supplementary part of non-wage labour costs. Data on major components - wages and salaries as well as statutory social security contributions is collected quarterly.

Table 5 in the Annex shows time series of the 2020 -  2022 figures on growth rates of total labour cost indices, which were disseminated in March 2023.

In 2023, the revision has been undertaken as usual in June (revision history presents table 6 in the Annex). Preliminary data for all quarters of 2021 was revised on the basis of results of the Labour Cost Survey 2021.

Table 7 in the Annex demonstrates latest time series on growth rates of total labour cost indices covering last 3 years, which were disseminated in June 2023.

A major revision was made in 2019. All time series were rebased to the new base year 2016. This was reflected in an increase of the gaps between the original and revised absolute values of LCI. However, the year-on-year growth rates were not affected by this revision.

Extent of revisions of the year-on-year growth rates made in 2022 for the 2021 data was often negligible below 0.5 percentual points for 39 % of economic activities, in particular services. Values for two thirds of branches remained under 1 pp, while the most important manufacturing reached 1 pp. This level was slightly exceed in transportation and storage (1.2 pp). The biggest gap between 1.7 and 1.8 pp was recorded in administrative and in other service activities.

Time between data availability and the last day of the reference period ranged between 62 - 67 days for the 2022LCI.

2022Q1: t + 67 days

2022Q2: t + 67 days

2022Q3: t + 67 days

2022Q4: t + 62 days

First results are usually available approximately 60 days after the end of the reference period (table 8 in Annex).

Final results are available approximately 17 - 26 months after the end of the reference period depending on availability of the LCS results (table 8 in Annex).

Data for the all quarters of 2022 were transmitted without a delay before the deadline of 70 days after the end of the reference quarter.

2022Q1: t - 3 days

2022Q2: t - 3 days

2022Q3: t - 3 days

2022Q4: t - 8 days

Comparability of the labour cost indices and weights has been maintained across all dimensions.

The LCI statistics is comparable as regards the observed area of the Slovak Republic. All regions are covered. 

Optional

The LCI statistics is comparable over time. 

The overall length of comparable time-series LCI_TOT, LCI_WAG, LCI_OTH, LCI_TXB is 22 years.

Beginning from 2010q1 labour costs in section O include data on statutory social security contributions for army forces that is measured by statistical questionnaire. This payment, which is stipulated by specific law different from law related to other civil employees, were only estimated before. Therefore, backdata had been probably impacted by slight underestimation.

In LCI, coherence at subannual level between the quarters of the same year is achieved by using the same methods. Annual data of basic year for indices is calculated as average of quarterly data with full coherence.

Optional

At present, the LCI series are submitted exclusively to Eurostat. No specific publication media with this data or metadata is available at the national level.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Not applicable.

Information on methodology is involved in the quality report on LCI, which is prepared on annual basis.

Statistical Office of the Slovak Republic holds certificate, which confirms that the Office meets requirements of the international standard ISO 9001:2015 in organising, obtaining, processing and providing official statistics according to the current standards. At the same time it provides the evidence that the quality management system implemented in the Office creates appropriate conditions for further improvement of quality of services provided to users and for development of the Office towards higher effectiveness. In addition, the LCI statistics fully respect the provisions of the European Statistics Code of Practice.

Optional

Not applicable.

LCI is compiled on the basis of results coming from existing sources. Therefore, it does not create any burden of respondents concerning costs or time for completing the questionnaires.

Level of the LCI agregation and dissemination is sufficient for keeping the data confidentiality.

The Law on State Statistics defines the secrecy and confidential data protection. Without the approval of legal or physical persons providing the relevant individual data, these data cannot be published or announced to anybody or used for other than statistical purposes.

If the number of the statistical units in any cell of the table formed by aggregating the individual data is less than three the data in the concerned cell is considered confidential. Confidential data are under high protection from direct or indirect identification and cannot be published.

Not available.