Annual Report on Structure of Earnings Survey Reference Metadata in the Euro SDMX Metadata Structure (ESMS) Agency for Statistics of Bosnia and Herzegovina
	1. Contact
	2. Metadata update
	3. Statistical presentation
	4. Unit of measure
	5. Reference period
	6. Institutional mandate
	7. Confidentiality
	8. Release policy
	9. Frequency of data dissemination
	10. Accessibility and clarity
	11. Quality management
	12. Relevance
	13. Accuracy and reliability
	14. Timeliness and punctuality
	15. Coherence and comparability
	16. Costs and burden
	17. Data revision
	18. Statistical processing
	19. Comment
	1. Contact
	1.1. Contact organisation	Agency for Statistics of Bosnia and Herzegovina
	1.2.  Contact organisation unit	Sector of Demography and Social Statistics – Labour Market Statistics Unit
	1.3. Contact name	a) Vladimir Mijović b) Tijana Popić
	1.4. Contact person function	a) Head of the Labour Market Statistics b) Senior Adviser for Labour Market Statistics
	1.5. Contact mail address	Zelenih beretki 26, 71 000 Sarajevo, Bosnia and Herzegovina
	1.6. Contact mail address	vladimir.mijovic@bhas.gov.ba tijana.popic@bhas.gov.ba
	1.7. Contact phone number	+387 33 911 991;   +387 33 911 962
	1.8. Contact fax number	+387 33 220 622
	2. Metadata update
	2.1. Metadata last certified	11/11/2025
	2.2. Metadata last posted	11/11/2025
	2.3. Metadata last update	11/11/2025
	3. Statistical presentation
	3.1. Data description	The Structure of Earnings Survey provides data on average monthly and annual earnings, as well as on average hourly earnings, by individual characteristics of employees such as occupation, sex, level of education, age, length of service with the enterprise, and type of employment contract (open-ended and fixed-term contracts, as well as contracts for temporary and occasional work). Data are also broken down by characteristics of the enterprise in which employees work, including economic activity, size, and form of ownership. These data enable employers and economic policy makers to gain insight into the level and structure of labour costs in Bosnia and Herzegovina. The results of this survey also allow employers and economic policy makers to assess labour cost structures in the domestic labour market and to compare them with those in the international labour market. The methodological framework of the survey is based on the recommendations and definitions of the Statistical Office of the European Union (EUROSTAT), as well as on the Regulations of the Council of the EU and the European Commission. According to the regulations, the Structure of Earnings Survey is conducted every four years.
	3.2. Classification system	Classification of economic activities of Bosnia and Herzegovina: NACE Rev. 2 (KD BiH)
	3.3. Sector coverage	The survey covers all economic sectors, except Sector A – Agriculture, forestry and fishing.
	3.4. Statistical concepts and definitons	Data for the reference year: Annual earnings include all payments made to employees on which the relevant taxes and social contributions are paid. Annual earnings comprise payments for work performed on a full-time, part-time or longer-than-full-time basis (overtime), arrears or wage differentials, allowances for on-call duty, night work and shift work, work on Sundays and public holidays, bonuses, regular monthly bonuses, compensation for transport costs to and from work, meal allowances during working hours, as well as compensation for hours not worked (annual leave, paid leave, public holidays, sick leave paid by the employer, leave for professional training, and work stoppages not caused by the employee). Total annual bonuses include the sum of all quarterly and annual bonuses in gross amounts (bonuses not paid every month). They include bonuses paid on the basis of interim and final accounts, the thirteenth salary, holiday allowance, and other periodic payments that have the character of earnings, as well as jubilee awards and severance payments to employees upon retirement or redundancy. The number of weeks to which the gross annual salary relates is calculated by dividing the total number of days (including weekends) covered by the salary by 7. The total annual net amount of regular monthly cash earnings of an employee for the reference year includes the sum of all monthly net wages and other monthly earnings of the employee (meal allowance, transport allowance) for the reference year on which taxes and contributions are not paid. The number of days of annual leave represents the number of days of annual leave to which the employee was entitled in the reference year, regardless of whether those days were used or not. When calculating average annual earnings, median annual earnings, average bonuses, average transport cost compensation, and the average number of days of annual leave, only employees who worked 30 or more weeks during the reference year are taken into account. If employees who received earnings for October of the reference year did not work throughout the entire year (due to sick leave, unpaid leave, starting work during October of the reference year, etc.), data for the full year are obtained through appropriate extrapolation. Furthermore, in order to eliminate differences in earnings resulting from different numbers of working hours between full-time and part-time employees, the earnings of part-time employees are extrapolated to amounts corresponding to full-time work. All mentioned variables collected for the reference year are extrapolated to correspond to full-year data (in cases where employees did not work the entire year). Additional extrapolations for part-time employees are carried out for annual earnings and bonuses to obtain amounts corresponding to full-time employment. Data for the reference month (October): Total gross monthly cash earnings of an employee include the gross wage for October (regardless of the month in which it was paid), meal allowance and transport allowance for October of the reference year. This amount also includes bonuses paid in the reference month, overtime supplements, as well as supplements for shift or night work, work on Sundays or public holidays, if applicable. Special supplements for shift work, night work, and work on Sundays and public holidays represent part of the total amount of earnings paid for October. In accordance with Commission Regulation (EC) No 1738/2005, supplements for overtime, shift or night work, and work on Sundays or public holidays are shown separately. Total net monthly cash earnings of an employee include the net wage for October (regardless of the month of payment) and other monthly payments (meal allowance, transport allowance) on which taxes and social contributions are not paid. The total number of paid hours for October of the reference year includes effective hours worked, overtime hours, and hours not worked but paid by the enterprise (e.g. sick leave paid by the employer, public holidays, etc.), i.e. all hours for which wages were paid for the month of October. The number of paid overtime hours represents the number of paid hours worked by the employee, at the employer’s request, beyond the agreed working time. The gender pay gap represents the percentage share of the difference between the average hourly earnings of male and female employees in the average hourly earnings of male employees. The share of low-paid employees in total employment is defined as the proportion of employees earning two thirds or less of the median hourly earnings in the total number of employees.
	3.5. Statistical unit	The units of observation are active legal entities with 10 or more employees selected in the sample.
	3.6. Statistical population	The target population of the Structure of Earnings Survey consists of all active enterprises from all activity sectors according to KDBiH 2010 (except Sector A) with 10 or more employees, as recorded in the Statistical Business Register.
	3.7. Reference area	The data refer to Bosnia and Herzegovina.
	3.8 Time coverage	Data from the Structure of Earnings Survey are produced and disseminated on a four-year basis.
	3.9. Base period	-
	4. Unit of measure	Data from the Structure of Earnings Survey are expressed in convertible marks and percentages.
	5. Reference period	The Structure of Earnings Survey is conducted for the reference year and for the reference month of October of the same year.
	6. Institutional mandate
	6.1. Legal acts and other agreements	The Structure of Earnings Survey is conducted on the basis of: • the Law on Statistics of Bosnia and Herzegovina (“Official Gazette of BiH”, Nos. 26/04 and 42/04), the Law on Statistics of the Federation of Bosnia and Herzegovina (“Official Gazette of FBiH”, Nos. 63/03 and 9/09), and the Law on Statistics of the Republic of Srpska; • the Annual Work Programme of the Agency for Statistics of Bosnia and Herzegovina; • Council Regulation (EC) No 530/1999 of 9 March 1999 on structural statistics on earnings and labour costs; • Commission Regulation (EC) No 1916/2000 of 8 September 2000 implementing Council Regulation (EC) No 530/1999 on structural statistics on earnings and labour costs, as regards the definition and transmission of information on the structure of earnings; • Commission Regulation (EC) No 1738/2005 of 21 October 2005 amending Commission Regulation (EC) No 1916/2000 as regards the definition and transmission of information on the structure of earnings; • Methodological guidelines of the Statistical Office of the European Union (EUROSTAT).
	6.2. Data sharing	Data from the Structure of Earnings Survey are transmitted to EUROSTAT, and are available to other users upon request.
	7. Confidentiality
	7.1. Confidentiality – policy	Statistical data confidentiality is regulated by law, and the staff conducting statistical surveys are legally obliged to protect confidentiality on the same legal basis. The Law on Statistics of Bosnia and Herzegovina (“Official Gazette of BiH”, Nos. 26/04 and 42/04 – Chapter XI, Articles 23–29) establishes the principle of confidentiality as one of the main principles. The Agency for Statistics of Bosnia and Herzegovina disseminates statistics in accordance with the statistical principles of the European Statistics Code of Practice, in particular the principle of statistical confidentiality.
	7.2. Confidentiality-data treatment	The document “Rules on the Protection of Statistical Data in the Agency for Statistics” sets out procedures to ensure confidentiality during data collection, processing and dissemination, including protocols for securing access to individual data, rules for defining confidential cells in output tables, and procedures for detecting and preventing secondary disclosure, as well as access to microdata for research purposes. By signing a Memorandum of Understanding with the Agency for Statistics of Bosnia and Herzegovina, data users who have access to individual data are obliged to: treat individual anonymised data as confidential in accordance with the relevant rules, regulations and procedures; ensure appropriate protection of individual data in line with these rules, regulations and procedures; safeguard the transmission of microdata; and destroy the media on which the data are stored, as well as the accompanying documentation, within five days after the publication of the results.
	8. Release policy
	8.1. Release calendar	BHAS publishes a Release Calendar every year with the exact dates and times of statistical releases for the following year.
	8.2. Release calendar access	The release calendar is available at:
		https://bhas.gov.ba/Calendar/
	8.3. User access	Data are disseminated to all users via the official website of the Agency for Statistics of Bosnia and Herzegovina:
		https://bhas.gov.ba
	9. Frequency of dissemination	The results of the Structure of Earnings Survey are published on a four-year basis, in accordance with the data release calendar.
	10. Accessibility and clarity
	10.1. News releases	Data from the Structure of Earnings Survey in the form of press releases are available on the website of the Agency for Statistics of Bosnia and Herzegovina:
		https://bhas.gov.ba/Calendar/Category?id=13&page=5&statGroup=13&tabId=2
	10.2. Publications	There is no publication containing the results of the Structure of Earnings Survey.
	10.3. On-line database	Data are not available in the BHAS online database.
	10.4. Micro-data access	Anonymised microdata datasets may be obtained under certain conditions prescribed by Article 18 of the Rulebook on the Protection of Statistical Data of the Agency for Statistics of Bosnia and Herzegovina, Chapter VI – Regulated access to microdata for research purposes, which is available at the following link:     https://bhas.gov.ba/data/Dokumenti/pdf/PRAVILNIK_O_ZASTITI_STATISTICKIH_PODATAKA.pdf
	10.5. Other	Websites of other institutions, written requests, telephone mediation, digital media, publications of international organisations, databases intended for internal use, and databases available to external users.
	10.6. Documentation on methodology	Structure of Earnings Survey:
		https://bhas.gov.ba/Calendar/Category/13
	10.7. Quality documentation	Izvještaji o kvalitetu se dostavljaju Eurostatu.
	11. Quality management
	11.1.Quality assurance	The Agency for Statistics of Bosnia and Herzegovina is committed to ensuring quality in the production of official statistics, in accordance with Article 19, paragraph 1, item “ ” of the Law on Statistics of Bosnia and Herzegovina (Official Gazette of BiH, Nos. 26/04 and 42/04). The Quality Management Policy, as an element of the quality management system, generally refers to the following components: the quality of business processes as well as other elements of the quality management system, such as planning and control. In order to ensure the quality of processes and products, the Agency for Statistics implements a Total Quality Management (TQM) model developed using the CAF tool. The Common Assessment Framework (CAF) is a general quality management tool recommended for public sector institutions, inspired by the excellence model of the European Foundation for Quality Management (EFQM). The self-assessment process identifies and monitors all organisational results and provides feedback on organisational capacities and the results of policy and planning processes. The self-assessment scoring tool includes 9 key indicators and 28 criteria, on the basis of which it is possible to measure the achieved level of management and identify critical points for future improvement. This approach views the organisation from different perspectives simultaneously, i.e. it applies a holistic approach to the analysis of organisational performance. BHAS uses the “CAF Implementation Guidelines” and a 10-step implementation plan developed to help organisations use the model in the most effective and efficient way. The self-assessment process identifies and monitors all important organisational results and provides feedback on organisational capacities and the results of policy and planning processes. This quality monitoring model requires the development of a standardised questionnaire that includes all principles of the model. In order to ensure the quality of statistics, the Agency has developed the document “Quality Management Policy and Programme”, which is available on the website:
		https://bhas.gov.ba/data/Dokumenti/Kvalitet/QM_Politika_i_Program_BHAS.pdf
	11.2. Quality assessment	The quality of official statistics products (main indicators/statistics) is regularly assessed in quality reports through the calculation of quality indicators. On the other hand, the processes involved in statistical production are assessed using specific process quality indicators in quality reports and additionally through GSBPM documentation.
	12. Relevance
	12.1. User needs	The key users of data from the Structure of Earnings Survey are: the Council of Ministers of Bosnia and Herzegovina, the Directorate for Economic Planning, the Directorate for European Integration, the Agency for Labour and Employment of Bosnia and Herzegovina, the Ministry of Finance of Bosnia and Herzegovina, Eurostat, UNDP, the World Bank, the International Monetary Fund, embassies and consular representations of foreign states, business entities, natural persons, the general public, the media, etc.
	12.2. User satisfaction	One of the key elements in ensuring the quality of statistical data is monitoring user satisfaction. For this survey, there is no specific user satisfaction survey. The results of the User Satisfaction Survey are available on the website:
		https://bhas.gov.ba/
	12.3. Completeness	The Structure of Earnings Survey is conducted on the basis of Council Regulation (EC) No 530/1999 and Commission Regulation (EC) Nos 1916/2000 and 1738/2005. The Structure of Earnings Survey is fully harmonised with EU and European Commission regulations.
	13. Accuracy and reliability
	13.1. Overall accuracy	The Structure of Earnings Survey is a sample-based survey; therefore, the results obtained from this survey are subject to the usual types of errors associated with sampling and survey techniques: sampling errors, non-sampling errors, measurement errors, processing errors, and non-response errors.
	13.2. Sampling error	For the allocation of business entities by strata, the “Neyman rule” was used on the variable number of employees (from the final accounts), with certain modifications of a practical nature. The Neyman rule states that sample units should be selected proportionally to the number of elements in the frame by stratum and multiplied by the standard deviation of the number of employees from the Business Register (SPR) by stratum. The Neyman rule was (modified) as follows: - Business entities with 50 or more employees are fully covered; - If a stratum contains fewer than 5 business entities, all are covered; - If the Neyman rule determines a sample of fewer than 5 business entities (e.g. 2 or 3) in strata that contain 5 or more business entities, then 5 business entities are included. Based on the described allocation, a stratified random sample was drawn from the total frame. R software was used as the programming package for the calculation of estimates, while the calculation of variables was carried out in SPSS. Sampling errors arise from the fact that the survey is not conducted as a full census, i.e. it does not cover all business entities. The survey is conducted on a sample, i.e. on a part of the business entities. It is also known that for a single survey it is possible to select a large number of different samples, with each sample producing estimates of key indicators that may differ to a greater or lesser extent. Sampling errors—unlike non-sampling errors, which are not measurable—indicate how reliable the obtained indicators are. As a check of indicator reliability, the coefficient of variation (CV) and confidence intervals are used. In the SES survey, the following notation is applied: 1. () the data are less reliable if the coefficient of variation (CV) is less than 0.20 and equal to or greater than 0.10, or expressed in percentages, 10% ≤ CV < 20%; 2. (()) the data are unreliable if the coefficient of variation (CV) is less than 0.30 and equal to or greater than 0.20, or expressed in percentages, 20% ≤ CV < 30%; 3. the data are extremely unreliable if the coefficient of variation (CV) is equal to or greater than 0.30, or expressed in percentages, CV ≥ 30%.
	13.3. Non-sampling error	Coverage errors (or frame errors) arise due to differences between the population covered by the frame and the target population. Three types of coverage errors are distinguished: undercoverage, overcoverage, and multiple listing (duplication). The fact that the Statistical Business Register as at 31 December of the reference year was used as the frame for this survey, and that the requested data refer precisely to the reference year, reduces the possibility of coverage errors. However, coverage errors are inevitable in all statistical surveys, including the Structure of Earnings Survey. Overcoverage in this survey can be measured through the reasons for non-response. Namely, if the reason for non-response is that a business entity was closed at the beginning of the reference year, liquidated, or did not operate in the reference year, it is considered that it should not have been included in the frame. In the absence of precise information on the reasons for non-response for business entities that are in some way considered as “overcoverage”, it was decided that these companies remain classified as non-respondents.
	14. Timeliness and punctuality
	14.1. Timeliness	Preliminary data from the Structure of Earnings Survey are published 18 months after the end of the reference period, while final results are published after 24 months.
	14.2. Punctuality	/
	15. Coherence and comparability
	15.1. Comparability-geographical	Level of Bosnia and Herzegovina, entities, and the Brčko District of Bosnia and Herzegovina
	15.2. Comparability-over time	/
	15.3. Coherence-cross domain	/
	15.4. Coherence-internal	/
	16. Cost and burden	The annual operational costs for conducting the survey are reflected in material costs and in the number of hours worked by employees of the Agency for Statistics of Bosnia and Herzegovina and the Agency’s branch offices in the Brčko District.
	17. Data revision
	17.1. Data revision-policy	The Agency for Statistics of Bosnia and Herzegovina currently does not have an official document related to a revision policy.
	17.2. Data revision-practice	Revisions of published data are carried out as needed, for example in cases where an error is detected. In such cases, the revised data are clearly marked, and users are informed through notes accompanying the press release.
	18. Statistical processing
	18.1. Source data	The observation units in this survey are business entities selected in the sample – legal persons (companies, enterprises, institutions, associations, foundations, and other organisations) registered in Bosnia and Herzegovina. The observation units are legal persons from all areas of activity according to KD BiH 2010, except for the following area: • A – Agriculture, forestry and fishing. Data were collected for legal persons as a whole, and not by their constituent units. The observation units are simultaneously the reporting units.
	18.2. Frequency of data collection	Data are collected every four years. For the implementation of the Structure of Earnings Survey, a paper-based questionnaire and a link for electronic completion of the questionnaire are used.
	18.3.  Data collection	Data are collected every four years. For the implementation of the Structure of Earnings Survey, a paper-based questionnaire and a link for electronic completion of the questionnaire are used. Reporting units submit the completed questionnaires to the entity-level regional departments within a period of two months. Legal persons complete the questionnaires on the basis of data from accounting and personnel records. Personnel records are used to provide data on the number of employees in the reporting units, while accounting records are used to provide data on wage payments and other labour costs. The questionnaires are submitted by selected business entities whose headquarters are located in the territory of Bosnia and Herzegovina. The deadline for business entities to submit the report to the entity-level regional departments is 60 days. After all received questionnaires have been entered, the entity-level regional departments submit the databases to the central statistical offices. The data processing procedure in the central statistical offices includes data checking, comparison, editing, and imputation. This period also includes the publication of preliminary and final results and the preparation of tables required by Eurostat, while the deadline for the preparation of the quality report, which is also prescribed by EU Regulation, is 24 months after the end of the reference period.
	18.4. Data validation	Entity-level regional departments provide professional assistance to reporting units in case of ambiguities when completing the questionnaire, collect the questionnaires, and control the coverage of reports for the regional department in which the units operate. They contact by telephone business entities that do not comply with the deadlines specified in the questionnaire, check the correctness of completed questionnaires, enter the data from the questionnaires, and correct errors. The data entry application includes built-in controls that prevent arithmetically and logically incorrect data from being entered into the database. This method of data entry ensures a database with logically and arithmetically correct data. The second stage of data processing for the Structure of Earnings Survey is carried out in the central statistical offices. Imputation represents the process of filling in missing or unacceptable values in survey data, with the aim of reducing the impact of incomplete data on the quality and reliability of the statistical results of the Structure of Earnings Survey. Missing values for which respondents were unable or unwilling to provide answers are imputed. The share of imputed data is negligible, as one of the editing phases involves contacting reporting units and completing missing data. After the completion of this processing phase, the database is forwarded to the Department for Sample Design and Survey Methodology, which calculates the variables required for this survey. Weights are calculated at the stratum level and adjusted for non-response rates by stratum. Calibration, post-stratification, and non-response weighting are not performed. Finally, based on the calculated variables, the Labour Market Statistics Department produces the results of the Structure of Earnings Survey.
	18.5. Data compilation	Procedures are applied to the data in order to derive new information in accordance with a given set of rules.
	18.6. Data compilation	No statistical procedures for series adjustment (seasonal adjustment methods) are used.
	19. Comment	/