National Statistical Institute

 Short-term Business Statistics and Transport Statistics Department  

Mariana Krusteva

Chief Expert

2, P. Volov Str.; 1038 Sofia, Bulgaria

[email protected]

+359 2 9857 174

The index of turnover is a key indicator for assessing the current state and current trends in ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’. It reflects the development of the production of goods and services taking into account the impact of the ongoing economic processes.

The monthly turnover index measures changes in the turnover of enterprises ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’. This information is suitable for monitoring of current economic developments and short-term forecasts.

Classification of Economic Activities (CEA-2008, for international use NACE.BG-2008).

Section G ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’ under NACE.BG 2008. The indices are calculated by groups defined in Regulation (EC) 1893/2006 on NACE.BG 2008.

The turnover includes net revenues from sales of products, goods and services survey units, including all taxes and fees (excluding VAT and other deductible taxes), and revenue from raw materials sold.

A enterprise of section G ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’.

The population of firms in section G ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’ includes about 111 000 units, which form a sampling population of approximately 33 000 commercial enterprises with an annual turnover exceeding over 10 thousand BGN. Of thes observed a sample of about 5 000 enterprises which form approximately 74% of trade turnover.

At national level: The indices are calculated and published total for the territory of Bulgaria
At European level: Eurostat calculates and publishes aggregated data for EU Member States (EU27) countries and the euro area (EU17). Eurostat publishes data received from the 27 EU Member States, Norway and the 3 candidate countries in the EU (Turkey, Croatia and the former Yugoslav Republic of Macedonia).

The indices are published as time series from January 2000 to the last month of the current reporting year.

The indices are calculated and published at 2015 as a base year.

Per cent (%)

Month

• COUNCIL REGULATION (EC) № 1165/98 of 19 May 1998 concerning short-term business statistics (STS);

• REGULATION (EC) № 1158/2005 of the European Parliament and of the Council amending, Regulation (EC)  № 1165/98;

• COMMISSION REGULATION (EC) № 1503/2006 of the European Commission as regards of definitions and the periodicity of indicators on STS;

• REGULATION (EC) № 1893/2006 - translation to NACE Rev. 2 and groups of activities of STS;

• COMMISSION REGULATION (EC) № 472/2008 for the conversion of data base according to NACE Rev. 2.

The indicators are transmitted to Eurostat before their national publication.

Regulation (EC) № 223/2009 on European statistics (article 20, paragraph 4) of March 11, 2009, stipulates establishing common principles and guidelines to ensure the confidentiality of data used for the production of European statistics and the access to those confidential information as required for scientific and technological development and needs of users in a democratic society.
National legislation: Law on Statistics includes provisions relating to confidentiality and protection of secrecy, which are in compliance with the EU legislation and standard European practice. Law on Statistics is stated that the individual data received and collected at the time of statistical surveys shall be statistical confidentiality and shall be used only for statistical purposes. Individual data obtained for the purposes of the statistical surveys, shall not be used as evidence to the bodies of the executive powers and the judicial authorities. Requirements, obligations and measures for data protection, as well as fines that can be imposed, are treated in separate chapters of the Law on Statistics, namely Chapter 6 ‘Protection of secrecy’ and Chapter 9, ‘Administrative penalty provisions’.

Article 25 of the Law on Statistics specifies the criteria according to which certain data cannot be made public or released - individual data, data which could be indirectly identified, as well as aggregations of less than three statistical units and/or when the relative share of one parameter is more than 85% of the total volume of this parameter for all population units.

Data which according to the Law on Statistics are confidential not published.

At the national level - according to the terms set out in the Release Calendar presenting the results of the statistical surveys carried out by the National Statistical Institute. The turnover indices are published about 37 - 40 days after the end of the reference month.

The calendar is available on the NSI website: http://www.nsi.bg/en/node/480

Data are published on the NSI website, section Business statistics in accordance with the Law on Statistics and the European Statistics Code of Practice respecting the professional independence and aimed at objectivity, transparency and equal treatment of all consumers.

Monthly

 NSI publishes regular monthly press release Turnover indices 37 - 40 days after the end of the month.

 The indices are not presented in publications.

Data on G ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’ are available to all users of the NSI website under the heading Statistical data› Bisiness statistics› Short-term Statistics› Monthly data

http://www.nsi.bg/en/node/6798

Micro data are neither distributed nor provided to the users.

Not applicable.

• Methodology Monthly Turnover Indices ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’:

http://www.nsi.bg/sites/default/files/files/metadata/dTrade_Method_1_en.pdf

• Eurostat Methodology of short-term business statistics. Interpretation and guidelines:

https://ec.europa.eu/eurostat/web/short-term-business-statistics/methodology

Regulation (EC) № 1165/98 and Regulation (EC) № 1158/2005 on Short-term statistics required by the European Commission (Eurostat) regularly (every three years) to report to the European Parliament on the applicability of short-term data, their quality and revisions of the data. Every year Member States should present to Eurostat detailed information about the quality of particular short-term indicators, which is part of the Principal European Economic Indicators.

The legal framework that regulates the production of short-term business statistics indicators states that statistics collected within the EU system must be of satisfactory quality and this quality and administrative burden that comes from what must be comparable across Member States. These criteria have been drawn up to ensure that these requirements are met, and that is reflected in the quality reports of the individual short-term indicators.

The quality of the short-term variables is measured by each Member State according to common criteria. The quality of the variables is measured regularly by comparing them with other statistical information, if it is possible some quality indicators are calculated and in addition these indicators are checked for internal consistency.

• Bulgarian state institutions and local authorities;
• Business and NGOs;
• Information and polling;
• Analytical and associate professionals;

• Internal users: Directorate ‘Macroeconomic Statistics’.

User reviews are collected by:

• Direct contact - via phone and email;

• Indirect contact - through regional statistical offices.

The users’  satisfaction is not measured.

All levels of detail that are required by the Regulations concerning STS is calculated and published.

The total population in section G ‘Wholesale and retail trade, repair of motor vehicles and motorcycles’ includes about 111 000 units, which form a sampling population of approximately 33 000 commercial enterprises with an annual turnover exceeding 10 thousand BGN. Of these monthly observed sample about 5 000 units, which form approximately 74% of the turnover of trade.

The main source of errors is the measurement errors which arise at the data collection stage for variety of reasons, for example respondents misunderstanding what information is required or non-providing precisely the desired information.

Shall be made afforrs to avoid these errors by providing assistance to the reporting units, improv the instructions attached to the questionnaire, assist the statisticians in Regional statistical offices. Most of measurement errors are detected and corrected through the logical and arithmetical controls of the Information system ’Short-term statistics‘ (Online Information system for data collection, editing and processing of the STS information). The system carries out the statutory and warning controls and in a case there are errors, it displays a warning dialog error in the report. The system does not allow completion of the report until they have corrected these errors by statutory controls.

The specific sampling method is ‘Stratified Random Sample’ in which first enterprises stratify by groups and then in each group by turnover. The stratums are: 1 – complete, 2 and 3 – sampling. Coefficient of variation is used as an indicator on precision. It is computed as a percentage.
Coefficient of variation for 2020:

Non-responders companies are valued according to the stratum in which they are faced. In stratum comprehensive assessment is done individually for each company using available information on trends in the establishment and adjusted trends in reported units of the same group and stratum. Sampling strata for each cell group and an overall assessment of non-response using extrapolation coefficients calculated from the data reported in the enterprise group and stratum after eliminating extreme values. In these cases it is assumed that there are no significant differences in the behavior of responders and non-responders enterprises. The response rate for July 2020 is 94% in terms of the number of enterprises and 93% in terms of turnover. Actions to reduce non-sampling errors: sending informative reminders and phone calls to collect data.

NSI provides information on turnover index from 37 to 40 days after the end of the month.

Turnover indices present ontime, according to the terms in the Release Calendar presenting the results of the statistical surveys carried out by the National Statistical Institute.

The turnover indices are representative for the whole country, they are not representative for the regions of the country. To calculate the indices of turnover NSI applies a methodology in accordance with international standards. In this sense, the comparability of the indicators across countries can be considered as very good.

The monthly indices are presented as time series from January 2000.

According to the requirements of the Regulation (EC) № 472/2008, since January 2009 the new version of Classification of economic activities (NACE.BG-2008) has been implemented in short term statistics. Historical data for all months back to 2000 were recalculated according to the new classification NACE.BG-2008 and published on NSI WEB-page in the Internet. For recalculation of the time series back to 2000, the ‘macro’ method was applied. The indices at the old classification (NACE.BG-2003) are recalculated according to the new classification (NACE.BG-2008) using transition coefficients, which were calculated from transition matrixes, based on micro data.

Regulation (EC) № 1165/98 requires ‘Every five years, Member States shall rebase the indices using as base years the years ending with a 0 or a 5. All indices must be rebased on the new base year within three years after the end of the new base year’. In compliance with these requirements since January 2018 the base year for Short-term business statistics has been changed. All short-term indicators presented in the form of index are calculated and published at 2015 as a base year. The time series has been recalculated according to the new base year and are posted on the website of NSI

The turnover indices are compared with the turnover indices calculated based on data from the annual survey on Structural Business Statistics. The differences are due to different coverage. SBS survey is conducted exhaustively and all enterprises are legally obliged to submit their Annual Reports to the National Statistical Institute. The STS surveys are sampling surveys. Sampling population includes companies with a turnover of over 10 thousand BGN. SBS is the source of turnover for calculation (reconciled) STS indicators.

Not applicable.

Since the beginning of 2012 NSI started the developing the online information system for data collection, data editing and data processing of STS information. Information system ’Short-term statistics’ provides more efficient way of collecting data from enterprises. This new system save recourses, reduce the burden of the reporting units and the quality of short term indicators is improved.

The turnover indices usually are revised two times. Each month data for the current month are published as preliminary and for the previous month as final. At the beginning of each year the indexes for the previous year are reconciled on the basis of the comprehensive annual data from SBS.

Revision policy / practice in case of turnover indices in trade:

The revised data are published on the NSI and transmitted to Eurostat.The MR and MAR are calculated from the year-on-year growth rates  for calendar adjusted data of 2018, 2019 and 2020 year(G47 - deflated turnover).The base year is 2015 MR=0,83 (Mean Revision) and  MAR=1,91 (Mean Absolute Revision).

The growth rates for G45 and G46 are published at first time during February 2021 and QPI for these sections aren't calculated.

The indexes are based on monthly survey ‘Sales in industry, construction, trade and services’.
In calculating the turnover index includes all receipts from sales (including revenues from sales of raw materials). Includes information about lateral not trade activities.

Monthly

Sampling survey. Total population includes about 111 000 units from which a sampling population is formed about 33 000 enterprises with an annual turnover of more than 10 thousand BGN. Of them observed monthly sample of approximately 5 000 enterprises that form approximately 74% of the trade turnover. The specific method is ‘Stratified Random Sampling’ in which units in each group stratify in three intervals depending on turnover. Data collection media:

• Online through Information system ’Short-term business statistics’;
• Paper questionnaire. The questionnaires are sent out to the respondents and collected by regional statistical offices, where the information is entered into IS STS.

To reduce non-response IS STS has functionality to send an informative and a reminder letters to respondents:

• Automatic formation of electronic letters informing respondents about the upcoming observations, in which they are included. To enable them to attach and reporting forms in PDF format and official letters;

Provide automatic sending of e-mail reminders to respondents who fail to report the time (when approaching the deadline for submission of reports by the subsystem).

Validation of data in the report:

When recording the data, the system carries out obligatory and recommended controls according to set with validation rules.

The system carries out the statutory and warning controls. In case there are errors, it displays a warning dialog error in the report. The system does not allow completion of the report until they have corrected these errors by statutory controls.

Validation of the output data:

Output data is verified at group level with regard to the previous month, the same month of the previous year and also with regard to extreme values, if necessary data is validated at the level of statistical units.

The turnover indices are calculated for units from the trade section for all groups defined in Regulation (EC) 1893/2006 on Classification of Economic Activities (NACE.BG 2008). Price effects are eliminated by using deflators derived by regrouping of products in the consumer basket in accordance with the content defined in the regulation groups and calculated based on average prices of the base year (2015).

The software used for seasonal and working day adjustment is DEMETRA+ (TRAMO-SEATS).

First data rows are tested for calendar effects, which usually includes:

• the different number of working days in a specific month;
• the composition of working days;
• the leap year effect, moving holidays such Easter.

In working day adjustment the following approach is applied

• RegARIMA approach, with all checks of significance and plausibility of the effects;
• RegARIMA approach, with all pre-tests for number of regressors, length and composition of month, check of plausibility of effects (sign and size of estimated coefficients), etc.;
• RegARIMA approach, with pre-tests for Easter and other moving holiday effects;
• The use of national calendar prepared by NSI experts.

In seasonal adjustment the following approach is applied:

• The series are checked for outliers of different types, but with the complete automatic procedure for detecting outliers;
• Complete automatic procedure for automatic selection within a large number of models according to the options of the tool, after checking for model adequacy using standard statistical tests (e.g. normality, heteroskedasticity, serial correlation, etc.);
• Fully automatic decomposition scheme selection using information criteria.The decomposition scheme specifies how the various components - trend-cycle, seasonal and calendar component and irregular – combine to form the original series. Usually, the default decomposition scheme is multiplicative, because in most economic time series, the magnitudes of the seasonal component vary proportionally to the level of the series;
• Direct seasonal adjustment is performed. All time series, including aggregates, are seasonally adjusted on an individual basis.

Revisions to seasonally adjusted data are published in accordance with accepted revision policy and release calendar that is aligned with the revision policy and the revision calendar for the unadjusted data.