Statistical Service of Cyprus (CYSTAT)

Labour

Statistical Service of Cyprus

CY-1444

Nicosia

Cyprus

The Labour Cost Survey (LCS) is carried out every four years. The Cyprus Statistical Service (CyStat) has carried out the LCS for the sixth time (2020, 2016, 2012, 2008, 2004, 2000), in compliance with the Council Regulation (EC) No 530/1999 concerning structural statistics on earnings and on labour costs and the Commission Regulation 1737/2005 of 21 October 2005 amending Regulation (EC) No 1726/1999 as regards the definition and transmission of information on labour costs. The purpose of the LCS is to measure the level and structure of labour costs, hours worked and hours paid.

Statistical Classification of economic activities NACE Rev.2. The size of the enterprise, in terms of the number of employees, is classified using the following bands: 10 to 49, 50 to 249, 250 to 499, 500 to 999, 1000 and more employees.

NACE Rev. 2 sections B to S are covered for all mandatory size classes, i.e. 10 to 49, 50 to 249, 250 to 499, 500 to 999, 1000 and more employees.

The LCS is carried out in compliance with the Council Regulation (EC) No 530/1999 concerning structural statistics on earnings and on labour costs and the Commission Regulation 1737/2005 of 21 October 2005 amending Regulation (EC) No 1726/1999 as regards the definition and transmission of information on labour costs.The purpose of the survey is the collection of data regarding the level and the structure of the labour costs, hours worked and hours paid. Data is broken down by economic activity and enterprise size.

Wages and Salaries reflect only payments made by the employer to employees (net of any subsidies received either by the employee or by the employer, in the context of covid schemes). Subsidies received by the employer reflect a special lump sum covid subsidy paid to enterprises in July 2020 that participated in the covid schemes and employed up to 50 employees.

The statistical unit is the enterprise. The government has a number of local units conducting different economic activities. Data were selected for each local unit separately.

All employees employed in enterprises with 10 or more employees.

The Republic of Cyprus.

Data refer to 2020. The Labour Cost Survey is carried out every four years since 2000 (previous vintages: 2000, 2004, 2008,2012 and 2016).

Not applicable.

Data are based on a survey. All employees employed in enterprises with 10 or more employees. The size classes 250-499, 500-999 and 1000+ and the central government were fully covered. The sampling frame was the Business Register. Concerning the private sector (including semi-governmental organizations and municipalities) the enterprises were first stratified by NACE.Rev.2 two digits divisions and within each NACE division, by employment size class i.e. 10-49 and 50-249. The sample sizes in strata were specified using proportionate stratification, i.e., the sample size of each stratum is proportionate to the population size of the stratum. The sample in each of the strata was selected using simple random sampling. The data are collected from the enterprises.

The survey is carried out every four years. 

A questionnaire was used which was completed either by personal interviews or through phone interviews. 

The data was validated by the auditors as the process ran, checking coverage and response rates, and comparing with data for previous cycles and with expectations After the end of the collection period, the data were analysed to produce the final results as requested by Eurostat. The data were validated using:

1. Checks on relationships between variables
2. Consistency checks
3. Cross checks between tables
4. Comparison of main aggregates to the Labour Cost Index, Labour Force Survey and National Accounts data.

Payments for days not worked (D1113) is estimated using information on direct remuneration paid in each pay period (D11111) and the number of days of paid absence due to holiday leave, public holidays, etc. Since all variables are collected at each collection, the estimate is considered quite accurate.  

When not provided by the enterprise, the part of guaranteed remuneration in the event of sickness (D1221) that refers to sickness is estimated similarly to payments for days not worked, using information on direct remuneration paid in each pay period (D11111) and the number of days of paid absence due to sickness. Since it is a small proportion of the employees who are entitled to sickleave during the year, the average (as estimated) may not provide a very accurate result, but the bias is considered to be small. In the case of maternity, the same applies as for sickleave in the private and broad public sector, but for the public sector, the monthly salary of the employee on maternity is recorded, therefore the estimate is considered to be accurate.

In all cases except public schools, the calendar year was used. In the case of public schools, the school year was used (September 2019-August 2020).

Not applicable.

The quality of statistics in CYSTAT is managed in the framework of the European Statistics Code of Practice which sets the standards for developing, producing and disseminating European Statistics as well as the ESS Quality Assurance Framework (QAF). CYSTAT endorses the Quality Declaration of the European Statistical System. In addition, CYSTAT is guided by the requirements provided for in Article 11 of the Statistics Law No. 25(I) of 2021 as well as Article 12 of Regulation (EC) No 223/2009 on European statistics, which sets out the quality criteria to be applied in the development, production and dissemination of European statistics.

Not provided.

The users of the LCS data are divided into Internal and External Users. The main user is Eurostat.

Internal Users include the following:

- The Statistical Service itself: The Labour Cost Index utilises the LCS to impute some variables. The National Accounts use the Hours worked for reference in estimating the Hours worked of Employees.

- Ministries and Other Government Bodies (i.e.  The National Central Bank)

- Enterprises

- Trade Unions

- Employers´ Associations

- Students, pupils

- Mass Media

- Research Institutes and Universities

External Users include the following: 

- Eurostat

- Other European Union Bodies: European Council, European Commission etc.

- International Bodies: OECD, Monetary International Fund, International Labour Organisation

- Foreign Employers

- Foreign Research Centres

Since 2008 (with the exception of 2010, 2013 and 2020) CYSTAT carries out an annual online “Users Satisfaction Survey”. The results of the surveys are available on CYSTAT’s website at the link attached below.

Overall, there is a high level of satisfaction of the users of statistical data published by CYSTAT.

• Results of CYSTAT’s User Satisfaction Surveys:

https://www.cystat.gov.cy/en/StaticPage?id=1144

Cyprus has complied fully with the EU Regulation as regards with the definitions of the variables collected in the LCS. All the variables included in the Regulation had been included in the questionnaire.

100% of the required data cells are available.

See below

The overall accuracy of the results is assessed in the following sections. As regards the sampling error for the private and broad public sector, the coefficient of variation for Annual Labour Costs is 1.1% and for Hourly Labour Costs is 1.2%. 

The fact that a sample of enterprises was selected for the survey, results in sampling error. The size classes 250-499, 500-999 and 1000+ and the central government were fully covered, therefore the concept is irrelevant in these cases. The sampling frame was the Business Register. The sample selection procedure for the private sector, including semi-governmental organizations and municipalities, took place separately from the Central Government. Concerning the private sector (including semi-governmental organizations and municipalities) the enterprises were first stratified by NACE.Rev.2 two digits divisions and within each NACE division, by employment size class i.e. 10-49 and 50-249. The sample sizes in strata were specified using proportionate stratification, i.e., the sample size of each stratum is proportionate to the population size of the stratum. The sample in each of the strata was selected using simple random sampling. In case a stratum is represented in the frame, but not in the sample, the sample size in this stratum is 1.

The coefficients of variation (%) for Annual Labour Costs and Hourly Labour Costs by NACE Rev.2 section and size class are shown in the tables below. 

The target population are all enterprises in NACE sections B to S with 10 or more employees in 2020.

The enterprises in the private sector, the broad public sector and municipalities have been sampled using the 2018 Business Register. The initial sample for the private sector, the broad public sector and municipalities was 1105.

The central government was fully covered. Each Ministry/Department/Service of the central government is considered a local unit. For the central government the initial units were 94. 

Cases of over-coverage and non response arose.

Private sector, broad public sector and municipalities

Under-coverage:

The use of the 2018 Business Register means that new enterprises in 2020 were not included in the sample.

A significant new enterprise, the State Health Services Organisation, was later included.

Over-coverage:

Out of the 1105 enterprises, the following instances of over-coverage were recorded:

1. Enterprises with less than 9.5 employees on average in 2020 - 122 cases
2. Enterprises that closed in 2020 - 29 cases
3. Enterprises which could not be located - 4 cases
4. Enterprises that merged with other enterprises - 1 case
5. Enterprises not active in 2020 - 11 cases
6. Enterprises out of scope according to domestic concept - 1 case

Due to misunderstaning, three extra enterprises not in the initial sample were included.

Central government

Over-coverage:

Out of the 94 units, the following instances of over-coverage were recorded:

1. Units with less than 9.5 employees on average in 2020 - 8 cases

Overall, the proportion of units accessible via the frame that do not belong to the target population, unweighted, is 14.7%.

Not applicable.

The Labour Cost Survey was conducted with computer-assisted personal and telephone interviews. Personal interviews was the approach that was preferred in previous vintages but telephone interviews were necessary this time, because of covid measures that applied during the collection period. Each questionnaire was a separate Excel file. Private sector, broad public sector and municipalities questionnaire differs from the central government questionnaire, for which many of the labour cost items were available from an administrative source (the Integrated Financial Information and Management Accounting System (FIMAS) which is used by the Ministries/Departments/Services for the accounting of the Government transactions and the monitoring of the Budget implementation). A large number of checks were carried out by the auditors (each auditor was responsible to check and validate each questionnaire for a number of enumerators). Errors in reading, calculating or recording numerical values though cannot be avoided completely.

The following instances were a cause of errors:

1. The recording of the impact of covid schemes on labour costs and hours worked was difficult in some cases. The questionnaire was designed and sent to the enterprises before the covid outbreak, therefore it did not incorporate for the covid schemes. And although the number of schemes was limited in number, many variations were found in practice. The enumerator had to understand how the calculations behind the questionnaire worked to record each case correctly and inevitably, errors were made.
2. Due to covid measures, there was a need to interview by phone. Given the complexity of the questionnaire, this was inevitably a cause of increased errors i.e. mainly through concentrating on getting answers for at least the most important variables.
3. Although the 2016 questionnare was modified to tackle the problem with the recording of the hours of employees working either shifts, either for 5 or 6 days but on a 7-day schedule or employees working flexible hours, these concepts remain difficult for enumerators to understand. Therefore, although there was certainly an improvement in the recording of the variables, a level of difficulty remained in calculating correctly the hours of work in these cases.
4. Recording absences, remained time consuming and as data are not readily available the enumerators had to make calculations before recording the values e.g. the number of days of sickleave refer in many cases to calendar days instead of the working days. In a number or cases, especially in the broad public sector, sickleave included absences recorded as sickleave, but were actually days not worked related to covid (vulnerable groups).
5. Labour costs not insured - sometimes the need to record costs was wrongly related to whether that cost was insurable or not. This led to two kinds of errors:
• labour costs not insured were not recorded
• items not belonging to labour costs (e.g. travelling expenses for business purposes) were recorded  as labour costs as they were insurable in some cases
6. Quarterly data were requested for some variables for the purposes of the Labour Cost Index and this increases the need for calculations and consequently the risk of errors.
7. In some cases an unsuitable person was assigned by the enterprise to provide the information requested, a person that may misunderstood or even ignored some concepts. In the cases it was realised, the necessary correction measures were taken, but this remains a risk factor.
8. In other cases the enterprises did not keep good or even no records of the information required for specific variables. An example is the Guaranteed Remuneration in the Event of Sickness. Since this is part of the employee's salary, it doesn’t appear as a separate item in the accounting books. The particular variable was requested from the companies this time, but in many cases the need remained to estimate it with a formula, based on the information collected on the specific enterprise's practice and the data on the number of paid sickleave and the average earnings for the enterprise.

There were cases of non response and inadequate data provision as follows:

 The non response cases were not replaced. The non response rate expressed as Non-response/(Sample size-Over-coverage), to take into account only in-scope units, is 4.5%.

See 6.3.3.

-

Programming and aggregation procedures were thoroughly checked in order to minimise processing error.

No imputation methods were used.

The formulae used is expected to accurately depict the unknown values.

Not applicable.

A data revision policy is in place at CYSTAT. It is published on CYSTAT’s website, at the following link:

https://www.cystat.gov.cy/en/StaticPage?id=1072

CYSTAT also publishes a list of scheduled revisions (regular or major revisions), also published on its website, at the following link:

https://library.cystat.gov.cy/NEW/List_Scheduled_Revisions-EN-190620.pdf

The data provided are final.

Not applicable.

The data collection period ran from January 2021 to August 2021. The units in the sample were informed from January 2020 in an attempt to first inform early so that they record throughout the year, the data necessary to complete the questionnaire and secondly create a contact with them in order to prompt a better response. The work done though was made difficult by the outbreak of covid, as the uncertainty for many firms, resulted in the loss of the initial interest and intention to complete the questionnaire gradually and in addition it made it more difficult for most to complete the questionnaire, as either labour costs or hours, or both, were affected by the covid schemes.

Another reason for contacting early, specific to the public sector, was that Cystat again helped Ministries/Departments/Services to extract data on absences by utilising the time recording system of the public sector. By the end of the collection period for the public sector, the majority of the labour cost items were recorded from the Integrated Financial Information and Management Accounting System (FIMAS) and government employees were allocated (number and absences) to the unit that incurs their labour costs (for example the Department of Public Administration and Personnel employed (and paid) employees who were allocated to all government units).

At the same time as the collection ran, a parallel process of validating by the auditors ran. After the end of the collection period of the private sector, the data were analysed to produce the final results as requested by Eurostat. 

Not applicable as final data are sent.

The final results are available at t+18months.

The data were transmitted to Eurostat on the 21st June 2022 i.e. before the deadline of 30th June 2022.

The data were sent to Eurostat within the time period required by the regulation.

It is possible to compare the national results with that of other European countries, as full compliance with the definitions of the variables laid down in the Council Regulation (EC) No 530/1999 concerning structural statistics on earnings and on labour costs is achieved.

Not applicable.

Statistics are comparable over time as the definitions used remain the same for all LCS vintages. However, it should be noted that besides the 2020 vintage, only the 2016 and 2008 vintages included data for the central government. Vintages 2012, 2004 and 2000 did not include data for the central government.

The data are comparable for years 2020, 2016 and 2008.

Analysis of the extent to which statistics are reconcilable with those obtained through other data sources or statistical domains follows:

1. Hourly labour costs
   1. Comparison with LCI
      • LCS2020/2016 growth rates of the hourly labour costs for all (part-time and full-time) employees including apprentices, working in an enterprise of 10 employee or more and belonging to NACE rev.2 sections B to S: 11.3%
      • 2020/2016 growth rates of the annual LCI: 10.6%
2. Total labour costs
   1. Comparison with SBS
      • LCS 2020 levels are not comparable with the SBS levels as for SBS purposes all enterprises are covered whereas for the purposes of the LCS, enterprises with 10 and more employees are covered.
      • The LCS2020/2016 growth rate of the total labour costs for sections B-N_S95_X_K: 29.7%
      • The SBS growth rate of the total labour costs for sections B-N_S95_X_K: 24.8%
3. Number of employees
   1. Comparison with SBS
      • The 2020 levels are not comparable with the SBS as for SBS purposes all enterprises are covered whereas for the purposes of the LCS, enterprises with 10 and more employees are covered.
      • The LCS2020/2016 growth rate in the number of employees: 25.6%
      • The SBS growth rate in the number of employees: 21.6%
   2. Comparison with LFS
      • The 2020/2016 growth rate in the number of employees taken from LCS, for all (part-time and full-time) employees including apprentices, working in an enterprise of 10 employee or more and belonging to NACE rev.2 sections B to S: 23.8%
      • 2020/2016 growth rate from LFS of the ‘Total number of employees’, all ISCED levels, age 15 to 64: 15.3%
4. Number of hours worked
   1. Comparison with LFS
      • The 2020/2016 growth rate in the number of hours worked taken from LCS for all (part-time and full-time) employees including apprentices, working in an enterprise of 10 employee or more and belonging to NACE rev.2 sections B to S: -6.4%
      • 2020/2016 growth rate from LFS of the ‘average number of usual weekly hours of work in main job’, employees, total worktime, all NACE sections: -0.5%

Notes on the differences between the different sources:

Differences with the LCI: 

In the above table, the labour cost per hour worked from the Labour Cost Survey was derived not taking into account vocational training costs (D2) and other expenditure paid by the employer (D3) i.e. (D1+D4-D5)/B1, as for LCI purposes total labour costs are defined as compensation of employees (D1) plus taxes (D4) minus subsidies received by the employer (D5).

Reasons for the differences:

• The LCS is a survey which gathers information from a sample of enterprises whereas the LCI is based on administrative data, the Social Insurance Services' Records, which provides data for all enterprises.
• Furthermore the Social Insurance Services' Records refer to all enterprise sizes, whereas the LCS refers to enterprises with ten or more employees.
• The hours worked used by the LCI refer to the hours worked produced for national accounts purposes and therefore differ from those produced by the LCS.

Differences with LFS:

• LCS is a business survey whereas LFS is a household survey.
• LFS covers full-time employees that worked during the reference week (hours worked > 0).

Difference with SBS:

• SBS covers all enterprises whereas for the purposes of the LCS, enterprises with 10 or more employees are covered.

Comparison with national source 

1. Average earnings {(D11-D1114+D1221)/A1}

Average earnings derived from the two sources are compared in the following table:

In the above table, the average earnings of an employee from the Labour Cost Survey was derived by subtracting from the wages and salaries (D11) the wages and salaries in kind (D1114) and adding guaranteed remuneration in the event of sickness (D1221) and dividing by the total number of employees (A1) i.e. (D11-D1114+D1221)/A1, as in the Social Insurance Services' Records figures the wages and salaries in kind are not recorded and the guaranteed remuneration in the event of sickness is included in earnings.

The reference year was 2020, which was the beginning of the covid pandemic. Covid measures’ impact on employment, labour costs and hours of work were recorded as per Eurostat’s guidance notes. The two sources are not comparable for many economic activities, as they closed during lockdown and employment was recorded differently for the purposes of the Labour Cost Survey and by the Social Insurance Service. As per the guidance notes “Under short-time work, temporary lay-offs and their many variants, employees remain within the scope of labour cost statistics. That is they should be included as employees of the enterprise.” For Social Insurance Services’ purposes, an employee is reported only in the month his/her employer paid a social insurance contribution. Therefore, the average number of employees recorded in the Labour Cost Survey and register differ due to the different definitions. As the number of employees is the denominator in calculating the average earnings, the average earnings, Labour Cost Survey estimate, are consequently lower the greater the effect of the pandemic measures for the specific economic activity. The economic activities with a large impact were B, I, L, R and S and some lesser impact on M.

Economic activity F was also affected, but F is also affected by another significant difference between the two sources: the Social Insurance Services’ records refer to enterprises of all sizes, whereas the Labour Cost Survey’s scope is enterprises with with ten or more employees.

A third significant difference occurs from the difference between the two sources in classifying the central government. In the Social Insurance Services' Records the central government is recorded in its entirety in Section O, whereas the Labour Cost Survey allocates each Ministry/Department/Service according to the unit's NACE Rev. 2 activity. This treatment affects sections O, P, Q significantly. Other sections affected are F, H, M and R.

A fourth difference that affects the broad public sector, municipalities and central government is that the Labour Cost Survey records salaries net of the salary deductions that apply, whereas the Social Insurance Services record gross salaries before the salary deductions.

Not provided.

Hourly labour costs

1. Comparison with NA - levels
   • LCS 2020 levels of the hourly labour costs for all (part-time and full-time) employees including apprentices, working in an enterprise of 10 employees or more and belonging to NACE rev.2 sections B to S: 18.1
   • Compensation of employees per hours worked, in euros: 15.3 (p)
2. Comparison with NA – growth rates
   • LCS 2020/2016  growth rate of the hourly labour costs for all (part-time and full-time) employees including apprentices, working in an enterprise of 10 employees or more and belonging to NACE rev.2 sections B to S: 11.3%
   • Compensation of employees per hours worked, 2020/2016  growth rate: 12.5%

Notes on the differences with the NA:

• The quarterly time series of the labour cost index and employment are used to estimate the Income Account and more specifically the wages and salaries (D11) and compensation of employees (D1). Data are finalised once SBS results are available. In the SBS, Labour Costs are only covered by two variables (wages and salaries and social contributions) without further details, whereas for the purpose of the LCS, these variables are requested in detail in accordance with the relevant regulations.
• The results for the LCS by NACE Rev.2 represent the size groups of more than ten employees, whereas the National Accounts represent all size groups and all economic activities.

Outputs are internally constistent.

CyStat publishes on its website a press release regarding the results of the Labour Cost Survey (link: https://www.cystat.gov.cy/en/PressRelease?id=67353).

CyStat publishes on its website a publication regarding the results of the Labour Cost Survey (link: https://library.cystat.gov.cy/NEW/Summary_Report_Labour_Cost_Survey-2020-EN-311022.pdf).

The main results are published on CyStat's online database (link: https://cystatdb.cystat.gov.cy/pxweb/en/8.CYSTAT-DB/8.CYSTAT-DB__Labour%20Cost%20and%20Earnings__)

-

Statistical micro-data from CYSTAT’s surveys are accessible for research purposes only and under strict provisions as described below:

 Under the provisions of the Statistics Law, CYSTAT may release microdata for the sole use of scientific research. Applicants have to submit the request form "APPLICATION FOR DATA FOR RESEARCH PURPOSES" giving thorough information on the project for which micro-data are needed.

 The application is evaluated by CYSTAT’s Confidentiality Committee and if the application is approved, a charge is fixed according to the volume and time consumed for preparation of the data. Micro-data may then be released after an anonymisation process which ensures no direct identification of the statistical units but, at the same time, ensures usability of the data. The link for the application is attached below.

• Link to the application for access to microdata on CYSTAT's website:

https://www.cystat.gov.cy/en/DataRequestContactForm?fid=7

CyStat has transmited the results of the Labour Cost Survey to Eurostat in accordance with the relevant regulations. Eurostat publishes the results on its on-line database. Relevant publications will also be available on Eurostat's website.

In addition to the above, CyStat publishes a relevant infographic on its website (link: https://library.cystat.gov.cy/Infographics/Cystat-Labour_Cost-EN-2020.pdf).

Methodological information are available on CyStat's website (link: https://www.cystat.gov.cy/en/MethodologicalDetails?m=2072).

Not provided.

Not provided.

Not provided.

Respondents have found the questionnaire burdensome and time consuming. They also frequently referred to the cost to the enterprises of employing staff who have to spend a large proportion of their time responding to Cystat's various requests. Those respondents that chose to fill in the questionnaire themselves, were asked to record (not obligatory) the number of hours taken to complete the questionnaire. This does not include all the time spent to complete the questionnaire as Cystat had to go back for clarifications and/or corrections on most occasions. The number of hours spent by our enumerators to complete the questionnaires was not recorded. Regarding the private sector, the average hours taken were 7.5 hours and of the public sector 47 hours. The average for the public sector may not be reliable as a small proportion of the public sector answered the question and may include the whole periods during which the employees spent only part of their time on preparing the data for Cystat. It has to be noted though that there are very large units in the public sector for which Cystat helped with the recording of the absences and took a considerably longer amount of time to record the necessary data.

The part that was more time-consuming was recording the absences. Cystat offered the help of enumerators employed for this purpose and informed all units in the sample before the beginning of the reference year, so the requirement were to their knowledge and if the informations was not readily available, they could take the necessary actions to record it. An additional purpose of informing early, was to help the respondents complete the questionnaire gradually during the year and alleviate the burden to complete it all at once. Unfortunately a minimal number of units followed this approach.

Comments were received on the practicallity of the questionnaire, but the data requested is complicated, and help was continuously availble with contact telephone numbers and emails provided, so that Cystat would provide answers and any help needed immediately.

Official statistics are released in accordance to all confidentiality provisions of the following:

• National Official Statistics Law No. 25(I) of 2021 (especially Article 16 on statistical confidentiality).
• Regulation (EC) No 223/2009 of the European Parliament and of the Council of 11 March 2009 on European statistics and its later amendments (especially Chapter 5 on statistical confidentiality).
• European Statistics Code of Practice (especially Principle 5 on statistical confidentiality).
• CYSTAT's Code of Practice for the Collection, Publication and Storage of Statistical Data.

Links:

• Statistics Law No. 25(I) of 2021:

https://www.cystat.gov.cy/en/StaticPage?id=1074

• Regulation (EC) No 223/2009 on European statistics (consolidated text):

http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:02009R0223-20150608&qid=1504858409240&from=EN

• European Statistics Code of Practice:

http://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/-/KS-32-11-955

• Code of Practice for the Collection, Publication and Storage of Statistical Data:

 https://www.cystat.gov.cy/en/StaticPage?id=1066

The treatment of confidential data is regulated by CYSTAT's Code of Practice for the Collection, Publication and Storage of Statistical Data.

• Code of Practice for the Collection, Publication and Storage of Statistical Data:

 https://www.cystat.gov.cy/en/StaticPage?id=1066

No comment.