Labour cost index (lci)

National Reference Metadata in ESS Standard for Quality Reports Structure (ESQRS)

Compiling agency: Statistics Iceland


Eurostat metadata
Reference metadata
1. Contact
2. Statistical presentation
3. Statistical processing
4. Quality management
5. Relevance
6. Accuracy and reliability
7. Timeliness and punctuality
8. Coherence and comparability
9. Accessibility and clarity
10. Cost and Burden
11. Confidentiality
12. Comment
Related Metadata
Annexes (including footnotes)
 



For any question on data and metadata, please contact: Eurostat user support

Download


1. Contact Top
1.1. Contact organisation

Statistics Iceland

1.2. Contact organisation unit

Wage Statistics

1.5. Contact mail address

Borgartún 21A, 105 Reykjavík, ICELAND


2. Statistical presentation Top
2.1. Data description

The Icelandic LCI is not based on a survey of its own, but administrative data and data from already existing surveys. The index reflects the quarterly development of total labour costs for all employers with 10 or more employees. 

2.2. Classification system

NACE Rev. 2.

2.3. Coverage - sector

NACE Rev. 2 Sections B-S.

2.4. Statistical concepts and definitions

The LCI measures quarterly developments in labour costs per hours worked.

2.5. Statistical unit

The statistical units are the local units by division of the NACE Rev. 2.

2.6. Statistical population

Enterprises with ten or more employees.

2.7. Reference area

Iceland.

2.8. Coverage - Time

First quarter of 2008 and onwards

2.9. Base period

Base year 2016=100.


3. Statistical processing Top
3.1. Source data

The main original source for wages and salaries is based on the administrative records for Pay-As-You-Earn (PAYE). PAYE is monthly records of wages and income tax on an employee-employer level reported to the tax authorities. The dataset used is post-processed, records reviewed and classified, internally for statistical usage. The PAYE records are on average available with up to a three month lag.

The main original source for hours worked is a business survey, the Icelandic survey on wages, earnings and labour costs (ISWEL). The target population in ISWEL includes all legal units with more than 10 employees, both in the public and private sector. Data is collected directly from the business units’ pay-roll systems. Among the information which is collected which is relevant to hours worked are; normal hours, hours with shift premium, hours in piece work, overtime pay, sickness pay, and contractual working hours. The ISWEL records are on average available with a one month lag. The information from ISWEL is applied to the main micro-dataset, which is primarily based on the PAYE. 

Other major data sources include the national register (daily information about gender, marital status, legal domicile, place of birth, citizenship and child bearing), individual tax returns, the graduation register (records of school, line of study, degree, mode of teaching and date of graduation), register for days at sea (date of departure and arrival of individual on vessels), and a database of collective agreements in Iceland (terms of the agreements and parties for which it applies).

Table 1: Variables and sources

Variable

Data

Source

Units

Number of employees and amount of wages and salaries 

Administrative data

PAYE

Local units

Taxes and employers contributions other than wages

Composite based on administrative data and business survey. 

PAYE; ISWEL; estimations

Variable, mainly estimated. Local units from employers with 10 or more employees for survey data.

Worked hours

Business survey

ISWEL 

Local units from employers of 10 or more employees

Subsidies received by the employer

Composite on business survey and a database of collective agreements in Iceland

ISWEL; estimations

Variable, mainly estimated. Local units from employers with 10 or more employees for survey data.

3.2. Frequency of data collection

The wages and salaries component are collected monthly as part of the PAYE. All cost items as well as working hours are collected monthly as part of the ISWEL, directly from the business unit's payroll system. See further in section 3.1.

Information on sick hours paid is available from approximately fifty percent of the business units in the ISWEL survey sample. For the remaining part of the sample hours paid for sick leave are part of normal hours paid. The proportion of sick hours paid was estimated for the sample data and the same proportion then assumed to apply to the rest of the sample. The data collection is annual.

Table 2: Origin, frequency and timelag of information on labour cost components

LCI component

Data collected

Frequency of uptdates

Source

Timelag

Wages and salaries

Monthly

Quarterly

PAYE

60 days

Hours worked and employers contributions other than wages

Monthly

Quarterly

ISWEL

30/60 days

Sick pay and absence hours due to sickness

Yearly

Yearly

ISWEL

+365 days

Taxes

Monthly

Quarterly

ISWEL; estimations

30/60 days

3.3. Data collection

See table in section 3.2.

3.4. Data validation

Source data from the PAYE and the ISWEL goes through a standardized quality check monthly and is prepared for statistical processing. Each quarter a validation procedure analyses LCI data and final indices quality. The data is, among other things, validated by comparison between time periods and other data sets.

3.5. Data compilation

The raw and unprocessed pay as you earn data register (PAYE) used in LCI is not usable for statistical processing. Therefore, Statistics Iceland has built an extensive processing system to handle the data. For example, the data does not show the type of payment. A strong emphasis is made to split payments into wages and other payments. Yearly the data is consolidated with individual tax returns and statements of earnings to validate the correct classification of payments. This data check has shown that the payment classification in the processing system is of high quality and payments are correctly classified. 

Working hours are estimated with a multimodal statistical evaluation, based on various data from Statistics Iceland's surveys, taking into account parameters of individuals, legal units and jobs. The parameters are applied to the statistical population based on the sample. To evaluate the parameters, several models are used which are integrated and evaluated based on the ISWEL. Each parameter is evaluated in a multimodal manner. The data is pre-processed, adjusted and post-processed to ensure that each group of entries is consistent with pre noted criteria of internal consistency and consistency with other sources. When parameters have been predicted for each entry in the population, hours worked are evaluated with multimodal models, based on criteria in the register data

3.6. Adjustment

Indices are working day adjusted and seasonally adjusted.


4. Quality management Top
4.1. Quality assurance

Statistics Iceland follows the principles in the Code of Practice for European Statistics (CoP) and uses the Quality Assurance Framework of the European Statistical System (QAF) for the implementation of the principles.

4.2. Quality management - assessment

Not available.


5. Relevance Top
5.1. Relevance - User Needs

The LCI is not published by Statistics Iceland at the moment. In addition, the index is flagged as preliminary data on Eurostat's website.

5.2. Relevance - User Satisfaction

Not applicable.

5.3. Completeness

The Labour Cost Index is available for all NACE sections B to S, starting from first quarter 2008.

5.3.1. Data completeness - rate

Not applicable.


6. Accuracy and reliability Top
6.1. Accuracy - overall

The overall reliability of the index is considered to be reasonably high. The component of wages and salaries is based on administrative records (PAYE) covering the whole economy. The hours worked and labour cost other than wages is mainly based on a business survey (ISWEL) and estimations. The survey includes the economic activities C, D, E, F, G, H, I, J, K, O, P, and Q in 2019. Data for economic activity I (Accommodation and food service activities) dates back to 2018, while data for other economic activities go back to 2008. Within the economic activities included in the ISWEL, there are approximately 85% of the labour market (B-S, enterprises with 10 or more employees) covered. In some economic activities certain sub-groups are missing. In the economic activity J small enterprises and enterprises in information technology (divisions 62 and 63) are missing. Economic activities O, P, and Q include only the public sector.

The ISWEL-data goes through thorough validation checks before the production to make sure to sort out observations with errors. This is especially true for the largest NACE sections of economic activity. However, for the smallest sections of economic activity the accuracy is lower. E.g. economic activity B (Mining and quarrying) is a remote part of the Icelandic economy and always held confidential due to a very small population.

The information from ISWEL is applied to the main micro-dataset, which is primarily based on the PAYE to obtain the estimated component of the LCI, such as hours worked and labour cost other than wages.

6.2. Sampling error

Not available at current moment.

6.2.1. Sampling error - indicators

Not available at current moment.

6.3. Non-sampling error

See below.

6.3.1. Coverage error

Not available at current moment.

6.3.1.1. Over-coverage - rate

Full coverage for all NACE sections for wages and salaries. Hours worked based on sample and estimation (see section 6.1).

6.3.1.2. Common units - proportion

Not available at current moment.

6.3.2. Measurement error

Not available at current moment.

6.3.3. Non response error

Not available at current moment.

6.3.3.1. Unit non-response - rate

Not available at current moment.

6.3.3.2. Item non-response - rate

Not available at current moment.

6.3.4. Processing error

See below.

6.3.4.1. Imputation - rate

There are currently no adjustments for unpaid overtime.Estimation method for hours worked:

Normal hours are generated for each job-record based on job specific characteristics. Hours paid are generated for each job-record through a process predominantly based on modelling information from business surveys. Absence due to holidays, annual leave, and sickness leave are generated for each job-record and subsequently subtracted from hours paid in order to generate hours actually worked.

Adjustments made to transform the original source to adapt it to the concept of working hours as defined:

  • Accounting for holidays and annual leave:
    Holidays and annual leaves are generated for each job-record. Initial assignment of hours paid but not worked due to holidays and annual leaves (separate processes) are based on employee and employer specific characteristics within the job-record. Holidays, whether that is a national holiday or other specific holidays, and rights for annual leaves are assigned if specifications in laws or regulations and in collective agreements, are applicable to the job-record. These figures for hours paid but not worked due to holidays and annual leaves are then adjusted in relation to hours paid within the job-record and in relation to hours-paid for the employee in other jobs within the month and within the year.
  • Accounting for sickness leave:
    Hours paid but not worked due to sickness leave are assigned to each job-record based on estimates from specific business surveys in relation to ISWEL.
    Currently these estimates are annual percentages for certain sections (NACE Rev. 2) from the Labour Cost Survey (LCS). Sickness leave on a more granular level is in development.
  • Accounting for strikes and temporary lay-offs:
    Strikes and temporary lay-offs are monitored, but no specific adjustments are deemed necessary since their effect is reflected in the PAYE.
  • Accounting for paid but unreported overtime:
    Paid but unreported overtime is estimated.
  • Accounting for unpaid overtime:
    There are currently no adjustments for unpaid overtime.
6.3.5. Model assumption error

Not available at current moment.

6.4. Seasonal adjustment

The data is seasonally adjusted with the R package RJDemetra using the TRAMO-SEATS method with specification RSA1.



Annexes:
Quality report on seasonal adjustment
6.5. Data revision - policy

No revision policy has been made, the Icelandic LCI is a preliminary series as it is still in development stage.

6.6. Data revision - practice

The Icelandic LCI is a preliminary series as it is still in development stage. Every quarter all data is revised.

6.6.1. Data revision - average size

Not available at current moment.


7. Timeliness and punctuality Top
7.1. Timeliness

Data is published 70 days after the end of the reference period.

7.1.1. Time lag - first result

Not available at current moment.

7.1.2. Time lag - final result

Not available at current moment.

7.2. Punctuality

Table 3: Data transmission

Quarter

Actual date of transmission

Planned date of transmission

Difference

2022-Q1

+ 71 days

 + 70 days

+ 1 days

2022-Q2

+ 70 days

 + 70 days

  0 days

2022-Q3

+ 70 days

 + 70 days

  0 days

2022-Q4

+ 55 days

 + 70 days

- 15 days

7.2.1. Punctuality - delivery and publication

Not available at current moment.


8. Coherence and comparability Top
8.1. Comparability - geographical

Iceland has one NUTS level.

8.1.1. Asymmetry for mirror flow statistics - coefficient

Not applicable.

8.2. Comparability - over time

The results broken down by sections and divisions of NACE Rev. 2 are comparable in time since 2008.

8.2.1. Length of comparable time series

From first quarter 2008.

8.3. Coherence - cross domain

Full comparability over all NACE Rev. 2 sections.

8.4. Coherence - sub annual and annual statistics

Estimates of the LCI and Labour Cost Levels (LCL) for labour cost variables and hours worked have complete internal consistency as they are based on the same corpus of microdata and administrative data. Moreover, they are calculated using the same methods of estimation.

8.5. Coherence - National Accounts

The Icelandic LCI is a collaboration project with the Icelandic National accounts on hours worked.

8.6. Coherence - internal

Overall internal coherence with the 4-yearly Labour Cost Surveys (LCS) is not feasible because the Icelandic LCI is compiled with wider range of data sources (see section 3.1) whereas the LCS has one main source, the ISWEL. The LCI coverage has full comparability over all NACE sections. The Icelandic LCS 2020 includes the economic activities NACE Rev. 2; C, D, E, F, G, H, I, J, K, M, O, P, Q and R, which cover about 90% of the labour market (B-S) according to the Pay as You Earn data (PAYE). Economic activities J and M are not fully covered and therefore flagged. The economic activities L, N and S are missing in LCS 2020 for Iceland. The economic activity B is only a remote part of the Icelandic economy.


9. Accessibility and clarity Top
9.1. Dissemination format - News release

Statistics Iceland has not yet published the LCI. Eurostat is the only institution that releases the Icelandic LCI. 

9.2. Dissemination format - Publications

Not applicable.

9.3. Dissemination format - online database

Eurostat's database.

9.3.1. Data tables - consultations

Not applicable.

9.4. Dissemination format - microdata access

Not applicable.

9.5. Dissemination format - other

Not applicable.

9.6. Documentation on methodology

Not applicable.

9.7. Quality management - documentation

Not applicable.

9.7.1. Metadata completeness - rate

Quarterly LCI metadata accompanies every quarterly submission of data to Eurostat. 

9.7.2. Metadata - consultations

Not applicable.


10. Cost and Burden Top

The Icelandic LCI is mostly based on administrative data and of existing data at Statistics Iceland. Cost and burden has not been estimated for the LCI.


11. Confidentiality Top
11.1. Confidentiality - policy

Regulation (EC) No 223/2009 on European statistics (recital 24 and Article 20(4)) of 11 March 2009 (OJ L 87, p. 164), stipulates the need to establish common principles and guidelines ensuring the confidentiality of data used for the production of European statistics and the access to those confidential data with due account for technical developments and the requirements of users in a democratic society. See also Statistics Iceland's Rules of Procedure for Treating Confidential Data:  https://statice.is/about-statistics-iceland/laws-and-regulations/statistics-icelands-rules-of-procedure-for-treating-confidential-data/

11.2. Confidentiality - data treatment

In those cases where the number of legal units is less than 5, the value is not published. That is why the economic activity Mining and quarrying (B) is not published.


12. Comment Top

Currently the time series starts at first quarter 2008. Plans are for extending the time series in successions to 2005 and 2000, in accordance to breaks in the source data.


Related metadata Top


Annexes Top