Turnover in industry

Statistical Office of the Republic of Serbia (SORS)

Industry and Energy Department,

Industrial Statistics Unit

Dunja Filipovic Jacovic

Statistician analyst for Industry Statistics Unit

Statistical Office of the Republic of Serbia, 5 Milana Rakica St, 11 000 Belgrade, Serbia

dunja.f-jacovic@stat.gov.rs

+381 11 2412 922, ext. 214

Industrial Turnover Index gives measure of the receipts of sales including the sales of goods, products and services provided to other parties. Industrial Turnover Indices measure changes in turnover between two different periods of time. Turnover in industry is an important indicator of demand for industrial products and services. Information about turnover in industry are particularly important for short-term analysis.

All disseminated data concerning turnover in industry are indices. Turnover indices give information on turnover in: total market, domestic market and non-domestic market.  

For each of these markets are shown: the total turnover in industry (sections B and C), MIGs (MIG Energy excluding sections D and E) and sections level (B and C separately) of NACE Rev. 2 classification (Statistical Classification of Economic Activities in the European Community). The reporting method is used for data collection. Data are collected by the Monthly Survey on the Turnover in Industry (IND-2).

- National classification of economic activities – CA (2010) based on NACE Rev. 2 classification (Statistical Classification of Economic Activities in the European Community).

- Main Industrial Groupings (MIGs).

Enterprises with principal activity classified in sections B (Mining and quarrying) and C (Manufacturing) of NACE Rev. 2. On the level of NACE Rev. 2 divisions smallest enterprises according to annual turnover in financial statements for 2021 are excluded.

The following information is collected from each reporting unit for each reference month: total income from sales of goods; income from sales of goods on non-domestic market; total income from sales of products and services; income from sales of products and services on non-domestic market.

Industrial turnover index is Laspeyres type index. Indices are calculated by aggregation of industrial turnover value data – sales in reporting month. The objective of the turnover in industry index is to show the dynamics of the market for industrial products and services.

Turnover comprises the totals invoiced by the observation unit during the reference period, and this corresponds to market sales of goods, products and services supplied to third parties. Turnover also includes all other charges (transport, packaging, etc.) passed on to the customer, even if these charges are listed separately in the invoice. Turnover excludes VAT and other similar deductible taxes directly linked to turnover as well as all duties and taxes on the goods, products or services invoiced by the unit. The indices of domestic and non-domestic turnover require turnover to be split according to the first destination of the product based on the change of ownership. The destination is determined by the residency of the third party that purchased the goods, products and services.

Units of statistical observation are enterprises with main activity in sections B and C of NACE Rev. 2. 

Statistical population consists of enterprises with main activity in sections B and C of NACE Rev. 2.

The sampling frame is based on statistical business register (SBR) which includes about 30 500 of enterprises in sections B and C that have reported annual financial statements for 2021. The sampling frame was constructed from this list of industrial enterprises by cutting off the smallest enterprises according to the auxiliary information on the annual turnover for 2021. The sampling frame includes about 5 730 enterprises that cover at least 95% of the total turnover and more than 90% of turnover on the level of NACE Rev. 2 divisions. About 604 units are surveyed every month.

Republic of Serbia

Since 1999, the Statistical Office of the Republic of Serbia doesn’t dispose of certain data for AP Kosovo and Metohia, so they are not contained in the data coverage for the Republic of Serbia (total).

The turnover of the units located outside the territory of the Republic of Serbia is not included. 

There are the time series indices for the period beginning in January 2000 on the 2015 base year.

For the period from January 2000 to December 2006 the data were estimated. Statistical survey of turnover in the industry has been conducted since January 2007.

Time series are available on dissemination database as: unadjusted indices (raw series), only calendar adjusted indices and seasonally and calendar adjusted indices.

Base period is average of previous year. For the year 2023, indices are published relative to the average 2022.

For the period beginning in January 2000 to December 2023 indices are recalculated with the 2015 base year. The referent year for indices transmitted to Eurostat is 2015.

From the publication of statistical release for January 2024 (from 15 March 2024 onwards), the time series of industrial turnover index since January 2000 are recalculated to the base year 2021 and submitted to Eurostat.

Indices.

Calendar month. Reporting units complete questionnaires until the 20^th of the current month for the previous month.

National legal bases for conducting the Monthly survey on the turnover in industry (IND-2) is regulated by the Official Statistics Law ("Official Gazette of RS", No. 104/2009), as well as Decision on the Program of the Official Statistics over 2021-2025 ("Official Gazette of RS", No. 24/2021), Regulation on defining the plan of official statistics for 2023 ("Official Gazette of RS", No. 141/2022) and Law on the classification of activities ("Official Gazette of the RS", No. 104/2009). Official statistics law states that all persons who manage or represent legal persons are obliged to provide the SORS with reliable data. There is obligation on units to provide data related to turnover in industry. The Мonthly survey on turnover in industry is included into the Regulation on defining the plan of official statistics for 2023 – it is compulsory.

Legal acts are available in SORS website: http://www.stat.gov.rs/en-US/o-nama/dokumenti

EU Legislation:

• Regulation (EC) No 1893/2006 of the European Parliament and of the Council of 20 December 2006 establishing the statistical classification of economic activities NACE Rev. 2,
• Regulation (EU) No 2019/2152 of the European Parliament and of the Council of 27 November 2019 on European business statistics (EBS-Regulation),
• Commission Implementing Regulation (EU) No 2020/1197 of 30 July 2020 concerning short-term statistics as regards the definition of variables, list of variables, frequency of data and the classification of the main industrial groupings (MIGs).

Statistical Office of the Republic of Serbia makes available the non-confidential data on the SORS website. Data are transmitted to Eurostat through EDAMIS in form unadjusted indices, only calendar adjusted indices and seasonally and calendar adjusted indices.

Data confidentiality is stipulated by:

-Official Statistics Law ("Official Gazette of RS", No. 104/2009), Articles 44, 45, 46, 47, 48 and 49.

(www.stat.gov.rs/media/2322/zakon_o_statisticie.pdf)

-Rulebook on statistical data protection in the Statistical Office of the Republic of Serbia

https://www.stat.gov.rs/en-US/o-nama/dokumenti under Rulebooks and guidelines

-Guidelines on measures of data and information protection in the Statistical Office of the Republic of Serbia

https://www.stat.gov.rs/en-US/o-nama/dokumenti under Rulebooks and guidelines

Data which enable statistical units to be identified, either directly or indirectly are not published. Data are published for the following levels: industry in total, sections B and C and the Main Industrial Groupings (MIGs). Data for these levels are not confidential.

Release calendar is publicly available on the SORS website: http://www.stat.gov.rs/en-US/calendar/

Data are collected monthly. Dissemination of results is also monthly.

Statistical release is available each month to all users, free of charge, at 12 o'clock on stated day in the Release calendar on SORS website.

Statistical release is published every month at 12 o`clock on the day specified in Release calendar. This release is available in paper and electronic form, bilingual (Serbian, English) and accessible on SORS website. Statistical release contains unadjusted monthly turnover index for total (B+C), MIGs, section B and section C separately for turnover in total industry, for turnover on domestic market and for turnover on non-domestic market. There are two types of indices: referent month compared to average of previous year and referent month compared to the same month of the previous year. Statistical release contains graph with unadjusted monthly indices for turnover in total industry, turnover on domestic market and turnover on non-domestic market, for referent month and the last 24 months compared to average of the previous year. Also, the table "Indices of turnover in industry, by MIGs and sections" is available in statistical release and it contains unadjusted monthly indices for turnover in total industry, on domestic market and on non-domestic market, in the last 13 months compared to the 2015 base year.  

The Monthly Statistical Bulletin, is a complex bilingual (Serbian, English) publication in paper and electronic form and it is accessible on SORS website. Table that presents indices of turnover in the industry is in the Industry chapter. The table contains unadjusted monthly turnover index for total industry, domestic market and non-domestic market for the current and previous year. For the current year monthly indices compared to the average of the previous year are presented. For the previous year, monthly indices compared to the average of that year are presented.

The database contains: unadjusted indices, only calendar adjusted indices and seasonally and calendar adjusted indices of turnover in the industry within the theme Industry. Indices of turnover in industry are given for the total market, domestic market and non-domestic market. For each of these markets indices for total (B+C), MIGs and Sections B and C were shown. For all the mentioned variables are presented monthly indices since January 2000 on 2015 base year. Also, annual indices from 2000 to 2022 on 2015 base year are presented. The database is available on-line to all users. The frequency of updating is monthly. From the publication of statistical release for January 2024 (from 15 March 2024 onwards), the time series of industrial turnover index since January 2000 are recalculated to the base year 2021 and submitted to Eurostat.

Link to database on SORS website: http://data.stat.gov.rs/?caller=SDDB&languageCode=en-US

Current data are sent monthly to Eurostat via EDAMIS. Also, there is dissemination on ad-hock request.

National methodological document, based on Eurostat documents, is available on SORS website: 

https://www.stat.gov.rs/en-us/istrazivanja/referentni-metapodaci/ ,

http://www.stat.gov.rs/en-US/istrazivanja/methodology-and-documents/?a=06&s=0

Not available.

Turnover in industry is based on reliable survey data. The level of accuracy is high. Publication of results is punctual, and easy assessable. Industrial Turnover Index calculation is based on recommended international methodology used in EU countries. It is internationally comparable since follows EU and relevant international concept, definitions and classifications. Comparability over time is good. The Republic of Serbia has required length of time series. Comparability with other statistical domains primarily the Quarterly Structural Business Survey (SBS-03) is satisfactory.

Data of turnover in the industry provide statistical information necessary to monitor the competitiveness and performance of the business community. It is used by SORS's National Accounts Department as input in compiling the components in national accounts. Therefore data are used by national users (Serbian National Bank, Ministry of finance, other government bodies, economic analysts in institutes, companies and financial institutions, media) and international users (Eurostat).

On biennial basis (once in two years), by the means of web-interview, implemented is the User Satisfaction Survey.

The survey results are available at SORS website: http://www.stat.gov.rs/en-US/o-nama/sistem-upravljanja-kvalitetom

"The Results of the User satisfaction survey, 2023" is the last survey conducted and available on link: https://www.stat.gov.rs/media/377354/results-of-the-user-satisfaction-survey-2023.pdf

Special user satisfaction survey regarding short-term indicators was not conducted so far.

All time series of Industrial Turnover Indices are calculated for: total industry, MIGs, sections, on total market, on domestic and on non-domestic market. Indices of turnover in industry are available since 2000. Тhey are based on NACE Rev. 2, and calculated in relation to the base year 2015. Time series are provided for levels according to STS Requirements, for countries belonging to the Group I.

As for all surveys, errors are inevitable. In this survey, there are sampling and non-sampling errors. Sampling errors are a consequence of observing a random sample instead of the whole population, while main non-sampling errors are coverage errors, non-response errors, measurement and processing errors. In spite of their existence, the survey design is such that they are kept under control and do not violate the estimates.

This survey is based on stratified simple random sample. Stratification of the sampling frame was according to MIGs (five classes: Energy; Intermediate Goods, energy excluded; Capital Goods; Durable Consumer Goods; Non Durable Consumer Goods), annual financial accounts size measure (four size classes: micro, small, medium and large enterprises), turnover in 2021 (two size classes: large enterprises on the level of MIG classes; other) and NACE Rev. 2 section (Mining and quarrying (B); Manufacturing (C)). Units with largest turnover were census units. Allocation of sample was done using the Bethel algorithm (Bethel, J. "Sample allocation in multivariate surveys", Survey methodology, 15, 1989: 47–57). Turnover was used as the auxiliary information. This way a sample of 604 units was allocated.

Sampling error is not estimated for the key indicator – Laspeyres index, but only for the monthly turnover in industry. For the total monthly turnover, estimates of the coefficients of variation (CVs) are from 1.59% to 2.27%.

The monthly surveys are based on the same sample (with differences due to non-response errors) or on coordinated samples with rotation of 1/5 of sampling units, for samples of successive years. That is why errors of indices are not expected to be high.

In the Annex referring to Accuracy is a table with monthly estimates of standard errors by main classes of MIG.

The survey sampling frame was constructed from SBR in such a way that the smallest enterprises were excluded, according to the turnover in annual financial statements for year t-2, where t is the survey reference year. Also statistical enterprises that are not obliged to report the balance sheets (mainly small enterprises that correspond to entrepreneurs in legal world) are not covered, as well as newly established.

Under-coverage arising from these causes is not measured, but it is expected that it does not have a significant impact on total turnover index in industry.

Over-coverage units (out of scope, ineligible) are resolved units determined not to belong to the target population for the survey. These are, for example units that are closed, not working, in bankruptcy, in liquidation or units whose economic activity is not covered by the survey. Over-coverage error is estimated from the survey. The average rate of over-coverage was 1% in 2023.

Based on industry turnover data collected in 2023 and the results obtained from processing them, the un-weighted response rate on the level of the Republic of Serbia on average was 98.01%. The size-weighted unit response rate was obtained with the size-weight that is the product of design sample weight and a size measure: reported turnover in industry in the given month. The size-weighted unit response rate on the level of the Republic of Serbia in 2023 was between 98.63% and 98.92%. The size-weighted response rate in 2023 on the level of Republic of Serbia on average was 98.85%.

Non-response units are in-scope units that are either non-respondents or those units that provide information below the "usable" threshold ("no contacts", refusals and other non-responding units). The un-weighted non-response rate on the level of the Republic of Serbia in 2023 was between 1.84% and 2.67%. The average un-weighted nonresponse rate on the level of the Republic of Serbia in 2023 was 1.99%. The average size-weighted non-response rate (size-weight is the product of design sample weight and a size measure: reported turnover in industry in the given month) on the level of the Republic of Serbia in 2023 was 1.15%.

In order to speed up data collection or reduce the non-response, non-responding units are followed up by telephone contacts or visited by field officers.

There is no case of item non-response as it is already checked in the stage of organizing the data whether the questionnaires are filled out completely. If this is not the case, the unit concerned is contacted by telephone and the missing data are obtained from it.

More information can be found in the Annex.

The Turnover Indices (total, on domestic and on non-domestic market) are firstly published 45 days after the end of the reference month.

Monthly indices are transmitted to Eurostat in a proper form within required deadlines.

All data are published according to the announced Release calendar on link https://www.stat.gov.rs/en-US/calendar

Every deviation from the plan published in the Release calendar is previously announced and explained.

Also, there are no deviations from the planned deadlines at any stage of survey implementation.

Turnover indices are representative only for the country level, they are not representative for the regions of the country.

SORS applies methodology, which is in line with international standards to calculate turnover indices. The application of the STS methodological guidelines ensures a good comparability between national data and European aggregates. 

Мonthly indices are presented as time series from January 2000 to December 2023 on 2015 base year. From the publication of statistical release for January 2024 (from 15 March 2024 onwards), the time series of industrial turnover index since January 2000 are recalculated to the base year 2021 and submitted to Eurostat.

SORS has started to collect data on turnover in industry since 2007. National Classification of economic activities CA (2010) (based on NACE Rev. 2) has been implemented in short-term statistics since 2011. Historical data for all months back, ie. from January 2007 to December 2010 were recalculated according to the NACE Rev. 2 (National Classification of economic activities CA (2010)).

Monthly turnover indices for the period January 2000 – December 2006 were estimated. In order to estimate the turnover in the industry for the period 2000–2006 was analyzed, compliance data available to the National Accounts Department at SORS and the data obtained from the survey IND-2 on turnover at the annual level for the section B + section C for all years since 2007. Results were showed a high degree of compliance and then the turnover in the industry was estimated in the following manner:

On the basis of monthly data for the period January 2007 – December 2012, from the survey on turnover in industry were calculated the average amount turnover by month and the average amount of turnover on the year. After that, were calculated the share of the average monthly turnover in the average annual turnover for the same period. This calculation was performed for the total turnover in industry, for the turnover on domestic market and for turnover on non-domestic market.

Total turnover of the sections B and C of NACE Rev. 2 per year, for the period 2000–2006, has been taken from the National accounts. The structure that was obtained from the statistical survey on turnover in the industry for the period 2007–2012 was applied to the turnover from the national accounts for the period 2000–2006.

After the calculation of the absolute amounts of monthly turnover for the period 2000–2006, turnover indices were calculated – monthly amounts ​​were compared with the average turnover in 2010 (average turnover for 2010 was obtained from the survey on turnover in the industry).

The time series from January 2000 are posted on the SORS website and transmitted to Eurostat.

Monthly survey on the turnover in industry (IND-2), Annual Structural Business Survey (SBS-01) and Quarterly Structural Business Survey (SBS-03) provide data of total industry turnover. Data may differ due to differences in coverage. Аnnual Structural Business Survey (SBS-01) is conducted exhaustively and all enterprises are legally obliged to submit their Annual Reports. Monthly survey on the turnover in industry (IND-2) is a sampling survey, and it is a part of sample of Quarterly Structural Business Survey (SBS-03) .

Also, data may differ due to differences in methodology. The "definitions" of turnover collected in SBS-01 and SBS-03 surveys and in Monthly survey on the turnover in industry (IND-2) are not the same.

Aggregates are always consistent with their sub-aggregates:

- turnover on domestic market and turnover on non-domestic market are consistent and coherent with total industrial turnover,

- turnover by MIGs is consistent and coherent with total turnover, and

- turnover by sections is consistent and coherent with total turnover.

There is no precise information about the costs and burden of respondents and statistics related to Monthly survey on the turnover in industry (IND-2).

Currently, turnover data are collected by using paper questionnaires which requires a lot of money and human resources. Also, web-based questionnaires has being introduced in order to reduce burden on statistics and on respondents. Respondents can choose to fill out web-based questionnaires or paper questionnaires.

As a consequence of the introduction of a web-based questionnaire, the costs of the respondents have been reduced because there are no costs associated with deliveries of questionnaires.

The SORS general revision policy constitutes the global frame ensuring that each statistical domain shall define its own revision policy in compliance with its specific nature.

The general SORS revision policy determines:

- general rules of revisions of the published data,

- forms of informing users as regards the possible causes of revisions,

- categories of revisions, and

- documents covering all aspects of revisions.

The General Revision policy is available at https://www.stat.gov.rs/en-US/o-nama/dokumenti (under Policy)

Revision calendar is available at: https://www.stat.gov.rs/en-US/kalzarev

The same revision policy is applied for publishing of national data and for data transmitted to Eurostat.

Also, there are some procedures in the Industry Statistics Unit. There is a confrontation with Quarterly Structural Business Survey (SBS-03) data. Some revisions are realized quarterly, when data from Quarterly Structural Business Survey (SBS-03) are available.

Regular revisions are realized every 5 years due to change of base year. Users are informed on those revisions in Note which follows statistical database.

Turnover indices for all months of the current year are published and treated as "preliminary results". At the end of the year, after possible changes they "become" final. Indices are available into statistical database posted on the SORS website.

Revisions are realized monthly. Revisions due to errors are implemented immediately after detection. The most of revisions of Turnover in industry indices are realized because of the availability of more complete data, or additional corrections from respondents. Estimates are replaced with definitive figures when they are available. The most of minor revisions are performed at the beginning of the year. Monthly, there are approximately 5–10 changes on reporting unit level. The most of them does not affect higher levels (total industry, MIGs, sections).

SORS has no practice to publish final indices on Turnover in industry in "new" Release form, but put them into statistical database. The users are informed that indices published in Statistical Release are "preliminary data" that might be changed. Minor revisions of Turnover in industry indices are not explaining to the users.

There is a confrontation with turnover from Quarterly Structural Business Survey (SBS-03) on reporting unit level. Some revisions are realized quarterly, when data from Quarterly Structural Business Survey (SBS-03) are available.

Regular revisions are realized every 5 years due to the change of base year. Users are informed on those revisions in Note which follows statistical database.

Mean Revision (MR) and Mean Absolute Revision (MAR) are used as revision indicators. They are calculated on year-on-year growth rates for unadjusted and calendar adjusted data. The calculated values for the last 36 months indices on turnover in industry, for the period from January 2020 to December 2023, are:

• Total industry for unadjusted data series MR = -0.05,  MAR = 0.56,

• Total industry for calendar adjusted data series MR = -0.14,  MAR = 0.85

The data source is statistical survey known as Monthly survey on the turnover in industry (IND-2). That survey covers enterprises registered in the Sections B or C of NACE Rev. 2. Survey is conducted on a stratified simple random sample. Strata are defined by cross classifying: 5 classes of MIGs (Energy; Intermediate goods, energy excluded; Capital goods; Durable consumer goods and Non-durable consumer goods); 4 size classes as defined in balance sheets (micro, small, medium and large enterprises); 2 size classes according to the reported turnover in balance sheets (census units and non census that are randomly selected) and 2 classes defined on NACE Rev. 2 sections (B (Mining and quarrying) and C (Manufacturing)).

Data are collected monthly.

Data are collected both with a paper questionnaires and with web-based questionnaires (IND-2). Questionnaires are sent to reporting units by post for all 12 months at the beginning of the year. Also, the questionnaire is located on the website and reporting units can use it. Reporting units may send completed questionnaires by post or e-mail. Telephone contacts are taken in order to increase response rate.

Data validation is done during data processing: first, during data entry, then through the IT process of calculating the aggregates. Data editing is based on rules designed to detect errors. Some errors can be automatically corrected, but the correction of some errors requires contact with the reporting units.

Finally, data validation is conducted during analysis of the aggregates.

Before transmission to Eurostat, the structure of the data file and format is checked in details. The data that should be transmitted to Eurostat are first transferred to the SORS dissemination database, from which they are automatically mapped. Than an SDMX-ML data file is formed and submited to Eurostat via EDAMIS.

The starting point for the data compilation, are the data on turnover collected from reporting units. Steps of processing data can be summarized as follows:

- Data entry from IND-2 questionnaire,

- Detection and correction of errors and inconsistencies,

- Reduction of non-response, i.e. contacting the reporting units in order they submit the data,

- Estimation procedure consisting of calculation of weights (original weights are corrected for non-response), detection and treatment of outliers and estimation of totals and their standard errors by defined domains of interest. During estimation, data suspected to be incorrect (for example, turnover expressed in RSD instead of thousands of RSD) are marked and returned to be checked,

- Calculation of indices for all aggregates (starting from divisions under sections B and C). Indices are simple Laspeyres indices. No weighting is done. Indices are calculated on the average of the 2015 base year, on the average of the previous year and compared to the same month last year and

- In quarterly periodicity, check of aggregated data by comparison of estimates with corresponding estimates from Quarterly Structural Business Survey (SBS-03).

Estimates of totals and their standard errors are calculated using the Horvitz-Thompson estimates according to the procedure for stratified simple random sample. That means that the estimates of totals are calculated by summing the weighted values from the sample. Sample design weights are corrected for non-response.

The data is seasonally and calendar adjusted. More information can be found in the Annex.

No comment.