Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
INS (National Institute for Statistics) is responsible for the data collection and dissemination.
1.2. Contact organisation unit
Department of Short term Business Statistics
1.3. Contact name
Confidential because of GDPR
1.4. Contact person function
Confidential because of GDPR
1.5. Contact mail address
Institutul National de Statistica (INS) Bulevardul Libertatii 16, Bucharest 5 RO-050706
1.6. Contact email address
Confidential because of GDPR
1.7. Contact phone number
Confidential because of GDPR
1.8. Contact fax number
Confidential because of GDPR
2.1. Metadata last certified
12 June 2025
2.2. Metadata last posted
12 June 2025
2.3. Metadata last update
12 June 2025
3.1. Data description
Monthly Index of Services Production (ISP). ISP aims to follow the monthly change in the services production.
ISP is compiled from deflated turnover value in the different service activities.
3.2. Classification system
NACE Rev. 2.
3.3. Coverage - sector
The following activities are covered divisions 49, 50, 51, 52, 53, 55, 56, 58, 59, 60, 61, 62, 63, 68, 69, 71, 73, 74, 77, 78, 79, 80, 81, 82, 92 of NACE Rev. 2 Group 70.2, as well as classes 9313, 9329, 9603, 9604, 9609, 9313, 9329.
In order to increase the efficiency of the sampling process, the possibility of including very small units in the selection base (in terms of number of employees and turnover) was reduced, thus units with less than 4 employees and turnover were eliminated from the base insignificant, trying to ensure a representativeness threshold of 90%, for the turnover indicator, at the level of each CAEN Rev.2 division.
3.4. Statistical concepts and definitions
Services turnover as defined in REGULATION (EU) 2019/2152 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 27 November 2019 on European business statistics, repealing 10 legal acts in the field of business statistics and COMMISSION IMPLEMENTING REGULATION (EU) 2020/1197 of 30 July 2020 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistics:
The sampling frame used for the sample selection is drawn from the Romanian Business Register (REGIS), which contained all enterprises, authorities and organisations as well as their local units that carried out any economic activity irrespective of their size or if they belong to the private or public sector. The Business Register is updated, yearly, on the balance sheets and contains the latest information on each enterprise in the statistical population such as the identification items (unique identification code, address etc.), economic activity codes (NACE Rev.2), number of employees (giving the enterprise size class). There are around 6200 units which are surveyed each month.
3.7. Reference area
All regions of Romania are covered.
No data of statistical units that are located abroad are included.
3.8. Coverage - Time
2015
3.9. Base period
2021=100
Indices (%)
Monthly data.
6.1. Institutional Mandate - legal acts and other agreements
REGULATION (EU) 2019/2152 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 27 November 2019 on European business statistics, repealing 10 legal acts in the field of business statistics and COMMISSION IMPLEMENTING REGULATION (EU) 2020/1197 of 30 July 2020 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistics:
Statistical Law no. 226/2009 regarding the organisation of official statistics. According to Law no. 226/2009 all individuals and legal units are required to submit data to the INS on request. Sanctions are possible but the INS prefers to use persuasion to gain the necessary data. All statistical data are collected on the basis of national legal acts approved by Romanian Governmet (i.e the "Programme of Statistical Surveys of National Statistics).
6.2. Institutional Mandate - data sharing
Under the provision of the statistical law, INS cannot publish, or otherwise make available to one individual or organisation, statistics that could enable the identification of data for any individual person or entity. No government official outside the National Institute of Statistics has access to the data before release to the public.
7.1. Confidentiality - policy
All statistics collected and published by INS are governed by the statistical Law no. 226/2009 regarding the organisation of official statistics. The Law specifies that: the official statistical services are authorised to ask for and receive free of charge statistical data and information from all natural and legal persons who are located, hold any kind of capital or carry out any type of activity on the Romanian territory. Also under the provision of this law, INS cannot publish, or otherwise make available to one individual or organisation, statistics that could enable the identification of data for any individual person or entity. INS has been repeatedly and publicly recognised by the government as fully intellectually independent: it decides by itself which data to collect, which methodology to use, and which results, methodological notes and commentaries are published. The most important aspects on the organisation of the official statistics in Romanian are posted on the National Institute of Statistics Internet website NIS website. Copies of the statistical laws are available in Romanian from the library of the National Institute of Statistics, 16, Libertatii Avenue, Bucharest, District 5.
7.2. Confidentiality - data treatment
No data are considered to be confidential.
8.1. Release calendar
There is an advance release calendar of data dissemination which can be found on the website Release calendar. The results of the monthly survey are disseminated at M+49 (for turnover services) days by press release and via INS Internet website NIS website and published at M+57 days in the Monthly Statistical Bulletin. The data can be found also in the TEMPO data base.
8.2. Release calendar access
The release calendar is available on the INS website NIS website.
8.3. Release policy - user access
Monthly data are available to all users simultaneously at the time of release through the press release and via INS Internet website NIS website. No organisation or person outside the INS has access to the data before release to the public. Data are transmitted to Eurostat as an index by e-DAMIS using SDMX format within 52 days (calendar days) after the reference month.
Monthly.
10.1. Dissemination format - News release
The results of the monthly survey are disseminated at M+49 (for turnover services) by press release and via INS Internet website NIS website.
10.2. Dissemination format - Publications
The specific publications: "Sinteza principalilor indicatori economico-sociali" only for I and N79 (Main Economic and Social Indicators) and “Buletinul Statistic Lunar” (Monthly Statistics Bulletin).
10.3. Dissemination format - online database
The turnover index are available also in our on line data base TEMPO which is available both in Romanian and English.
10.4. Dissemination format - microdata access
No microdata are disseminated.
10.5. Dissemination format - other
Data on services turnover indices are sent monthly to Eurostat both to be used in European aggregates as well as to be released as national data.
10.6. Documentation on methodology
Brief details of the methodology are included with the published data.
10.7. Quality management - documentation
The statistical process is monitoring regarding relevance, accuracy, timeliness, punctuality coherence and comparability. Self assesment of the survey manager are carried out from time to time.
The documentation on quality management is available on the NIS website.
11.1. Quality assurance
The system of official statistics follows international and European quality standards. NIS Romania has adopted the UN Fundamental Principles for Official Statistics as well as the European Statistics Code of Practice.
Data are relevant because are required by Europen legislation and Eurostat.
The main national users are grouped as following: Presidential administration, Governmental bodies; National Commission for Prognosis; mass-media; other users.
Their main needs: The Monthly Survey on Short-Term Statistics outcomes, in relation with the information provided by other surveys, are used by policy makers in establishing the actions and measures for economic development strategies.
NIS Romania has conducted general surveys of user satisfaction in 2024, and plans to continue doing so every three years. The purpose is to know the perception of users regarding the quality of the information they use and if their needs are covered.
The users’ specific needs are considered when methodological revisions are carried out in order to tailor the content of the survey to the specific requirements of its users, thereby increasing their levels of satisfaction.
Data are compliant with European legislation and Eurostat requirments. All breakdowns and variable forms are provided, according to the EU regulations in force.
13.1. Accuracy - overall
There are negligible non sampling errors concerning coverage, measurement and non response. The response rate is 92.4%.
13.2. Sampling error
There is an exaustive survey for enterprises employing 50 persons and more. For the remainder a sample is drawn. In general, units with less than 4 employees are excluded. See also point 3.3 for thresholds aplied for each activity.
13.3. Non-sampling error
Usually the typing errors could appear in the processing stage and the data are revised the next month. The difference between initial figures and revised figures represents about +/-3.0 % of the initial figures for total retail trade index , gross series.
The sampling volume was determined by imposing that at the total country level the precision estimator should be 0.03, at a confidence level of 95%.
14.1. Timeliness
All the activities carried out by INS during a year are scheduled through the Activity Plan with several months earlier. Annually, INS issues for the users a press release calendar and a catalogue of statistical products and services containing the names, periodicity, short descriptions and release dates for all publications. The press release is issued 45/49 days after the end of the reference month. The publication is ready within 57 days from the end of the reference month. The questionnaires are prefilled with identification data about reporting units and sent to them at the beginning of each quarter of the year. The enterprises are requested to answer by the 25th day following the reference month.
14.2. Punctuality
The process concerning the deadlines for the respondents to reply was performed as follows:
the units have to fill in the paper questionnaires and send them to the territorial statistical offices until 15th of the month following the reference month.
the units can also fill in the on line questionnaires until around 25rd of the month following the reference month.
The process concerning the covering recalls and follow-ups was performed as follows:
from 15th till the 25th of the month following the reference month, the territorial statistical offices had to perform data entry and data checking.
In this phase, the covering recalls and follow-ups are taking place. The statisticians from the local offices contact or re contact the units/respondents who did not sent the data, irrespective of their size classes, but the stress was more on the enterprises with more than 50 employees, who are exhaustively surveyed and on those from the public sector. Since not all respondents comply with the deadline requested, a small proportion of questionnaires reach with delay the territorial statistical offices.
Data processing steps and the related periods were:
At local level (regional/county statistical office) the data processing consisted in:
data collection – until 15th of the following month of the reference month;
data entry and validation- from 15th of the following month of the reference month till the 25th day of the month;
At central level (INS) the data processing consisted in:
data validation – errors were detected and sent for correction or explanations to the statistical offices;
comparative analysis with the previous month, the same month of the previous year and with other data sources (surveys and administrative data);
the analysis of the non-weighted data;
the weighting procedures;
the analysis of the weighted data;
data tabulation;
monthly data preparation for dissemination in the press releases, the Monthly Statistical Bulletin, the TEMPO (on-line) database;
transposing into the STS standard format for data transmission designed by Eurostat;
transmission of final results to Eurostat.
The publication dates for the monthly indicators were disseminated in accordance with the legal requirements as follows:
European level (Eurostat): M+52 of each month;
National level:
Press releases - M+49 (for turnover services) of each month;
Monthly Statistical Bulletin (MSB) - M+57 and TEMPO data base - M+59 of each month.
15.1. Comparability - geographical
Indices are representative for the whole country; they are not representative by regions of the country. NSI applies methodology, which is in line with international standards. In this sense, the comparability of the indicators across countries can be considered as very good.
15.2. Comparability - over time
No major methodological changes over the last years.
15.3. Coherence - cross domain
Confrontation is made with the annualy SBS data. Usualy the monthly data are not revised as a result of confrontation but some revision of data could be done when we change the base year as a result of confrontation with annual SBS data.
15.4. Coherence - internal
All data are coherent at different aggregation levels of NACE.
Number of units =6200
Cost (NSI) hours per year=16479
Burden (respondents) hours per year= 14637
17.1. Data revision - policy
Data revisions follow the Revision Policy of the NIS Romania, which represent one of the key documents of the office. The same revision policy is applied to STS data released nationally and transmitted to Eurostat. Calendar of standard revisions is being prepared together with the release calendar Revision policy.
Routine revisions: The revised indices are published together with the preliminary indices for the following month. Notice is given at the time the change is introduced.
Major revisions: In the case of regular index rebases (as a rule every five years), the results and changes in the calculation methodology or major changes as classification notice are presented in detail in the Monthly Statistical Bulletin and on the NIS website well in advance of the change.
Vintage database are not available.
17.2. Data revision - practice
The data are preliminary when first released and are revised three months after the end of the reference month.
The difference between initial figures and revised figures represents about +/-3 % of the initial figures for services turnover index , gross series.
Major changes in methodology are announced in all statistical publications at the time of the change.The status of data is clearly indicated in the publications.
The quality indicators used in data revision are: MR (Mean Revision) and MAR (Mean Absolute Revision). MAR and MR values were obtained with the information of its last 36 final months being December 2024 the last one.
The Mean Revision values of the differences between the general index first released and the last version, in the period January, 2022 to December, 2024 (36 months) was for WDA series was MR=0.13461
And the Mean Absolute Revision for this period, for WDA series was MAR=0.344711
18.1. Source data
Statistical survey. The frame used for sampling is business. There is a stratified sample survey. The sample is updated annualy.
All units with 50 employees or more are included. For the remainder a sample is drawn.
The market services sample includes approximately 1200 enterprises and the sample for services rendered mainly to enterprises includes approximately 5000 enterprises.
18.2. Frequency of data collection
Data are collected monthly.
18.3. Data collection
One combined questionnaire regarding turnover, number of employees, wages and salaries, investment is used (SERV TS – Short term indicators in services in month ……, Chapter CA Turnover in retail trade according to annex D of the STS regulation). Data are collected in principal using a web questionnaire. For the rest of the respondent units paper questionnaire or electronic format by e-mail is used. The paper questionnaires are sent to the reporting units and then collected by county statistical offices. From 2010 onwards data are collected online using the eSOP portal for statistical data collection.
18.4. Data validation
The IT solution for Monthly survey on short-term statistics was developed to find out the measurement and processing errors occurred in different stages of the survey. The application was designed for on line data collection and validation.
The IT solution allowed to perform on line data entry and validation at unit level. Also, the IT solution allowed to perform data entry and validation for questionnaires received on paper by post/email at county level (42 counties in total – NUTS3 level).
The IT solution contained the following categories of logical tests that check:
the primary data from the questionnaires;
the logical flows among the questionnaire chapters;
the data integrity and correctness;
the data comparability with the previous month.
The logical tests have the scope:
to follow and check the logical flows of the questionnaire;
to correlate data from related chapters of the questionnaire;
to define and check certain limits to which the indicators should belong compared to their evolution over time and between them;
to find out the outliers or non-logical values.
The tests for data integrity check the identification information from the sample of units with the information from respondent files.
At central level more accurate and detailed procedures were developed for checking data integrity and correctness:
data comparison with the previous and corresponding months;
non-weighted procedures;
data tabulation;
automatic transformation of final results into the STS standard format designed by Eurostat;
data transmission to Eurostat.
Logical tests were performed for data checking and correlation on all the indicators collected through the survey. Since the same tests were conducted at local level (unit and county) as well as at central level, the largest part of the errors found was solved from the beginning (at local level). The figures were corrected after re-contacting the corresponding enterprises by the statistical offices.
The questionnaire design, detailed explanatory notes and the IT solution developed had as result a complete and of good quality data set.
18.5. Data compilation
Non-response is imputed using the estimated turnover from the D300 declaration (VAT declaration) and for cases where the information is not found in the VAT, the average calculated based on the historical series of the unit is used. If the lack of response is due to the closure or temporary suspension of the activity, no imputation is made. The production indices of turnover are Laspeyres-type indices and are calculated under comparable methodological and price conditions. Deflator price indices are used to express the values from the current period in the prices of the reference period. The base year and the weighting system are for the year 2021. The weights used for aggregation are calculated based on the turnover according to the results of the Structural Survey of Enterprises from the reference year (2021).
18.6. Adjustment
In order to adjust the series Jdemetra + v2.2 was used (TRAMO/SEATS method) estimating the effect of the number of working days, different from one month to another and the calendar effect (orthodox Easter, leap year and other national holidays) as well as the identification and correction of outliers (occasional, transitional or permanent changes) and interpolation of missing values. Series adjusted for the number of working days was obtained by eliminating from unadjusted series those effects by means of correction coefficients set up according to the regression model used (additive or multiplicative). The regression models used for each series are set up at the beginning of each year involving the recalculation of adjusted series calculated in the previous year (recalculation due to the change in the adopted models, number of regressors used, number of available observations). Aggregated levels were adjusted by the direct method namely direct adjustment of aggregated series. The estimation of unobserved components (trend-cycle, seasonality and irregular components) is done by the SEATS program based on ARIMA models. The series adjusted for working days and seasonality is obtained by eliminating the seasonal component from the working-day adjusted series.
Monthly Index of Services Production (ISP). ISP aims to follow the monthly change in the services production.
ISP is compiled from deflated turnover value in the different service activities.
12 June 2025
Services turnover as defined in REGULATION (EU) 2019/2152 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 27 November 2019 on European business statistics, repealing 10 legal acts in the field of business statistics and COMMISSION IMPLEMENTING REGULATION (EU) 2020/1197 of 30 July 2020 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistics:
The sampling frame used for the sample selection is drawn from the Romanian Business Register (REGIS), which contained all enterprises, authorities and organisations as well as their local units that carried out any economic activity irrespective of their size or if they belong to the private or public sector. The Business Register is updated, yearly, on the balance sheets and contains the latest information on each enterprise in the statistical population such as the identification items (unique identification code, address etc.), economic activity codes (NACE Rev.2), number of employees (giving the enterprise size class). There are around 6200 units which are surveyed each month.
All regions of Romania are covered.
No data of statistical units that are located abroad are included.
Monthly data.
There are negligible non sampling errors concerning coverage, measurement and non response. The response rate is 92.4%.
Indices (%)
Non-response is imputed using the estimated turnover from the D300 declaration (VAT declaration) and for cases where the information is not found in the VAT, the average calculated based on the historical series of the unit is used. If the lack of response is due to the closure or temporary suspension of the activity, no imputation is made. The production indices of turnover are Laspeyres-type indices and are calculated under comparable methodological and price conditions. Deflator price indices are used to express the values from the current period in the prices of the reference period. The base year and the weighting system are for the year 2021. The weights used for aggregation are calculated based on the turnover according to the results of the Structural Survey of Enterprises from the reference year (2021).
Statistical survey. The frame used for sampling is business. There is a stratified sample survey. The sample is updated annualy.
All units with 50 employees or more are included. For the remainder a sample is drawn.
The market services sample includes approximately 1200 enterprises and the sample for services rendered mainly to enterprises includes approximately 5000 enterprises.
Monthly.
All the activities carried out by INS during a year are scheduled through the Activity Plan with several months earlier. Annually, INS issues for the users a press release calendar and a catalogue of statistical products and services containing the names, periodicity, short descriptions and release dates for all publications. The press release is issued 45/49 days after the end of the reference month. The publication is ready within 57 days from the end of the reference month. The questionnaires are prefilled with identification data about reporting units and sent to them at the beginning of each quarter of the year. The enterprises are requested to answer by the 25th day following the reference month.
Indices are representative for the whole country; they are not representative by regions of the country. NSI applies methodology, which is in line with international standards. In this sense, the comparability of the indicators across countries can be considered as very good.
No major methodological changes over the last years.
