Annexes I, II, III: National Statistical Institute (INE)

Annex IV: Ministry for Transport, Mobility and Urban Agenda

Annexes I, II (except series 2H, 2I, 2J and 2K), III: Subdirectorate General for Economic Sectors Statistics. INE

Annex II (series 2H, 2I, 2J and 2K): Subdirectorate General for Environmental, Agricultural and Financial Statistics. INE

Annex IV: General Department for Economic Programming. Ministry for Transport, Mobility and Urban Agenda

Annexes I, II, III: Avenida de Manoteras 50, 52. 28050 Madrid. Spain

Annex IV:  Paseo de la Castellana, 67. Nuevos Ministerios. 28071 Madrid. Spain

Structural business statistics (SBS) describes the structure, conduct and performance of economic activities, down to the most detailed activity level (several hundred economic sectors).

SBS covers all activities of the non-financial business economy with the exception of agricultural activities and personal services. Limited information is available on banking, insurance and pension funds.

Main characteristics (variables) of the SBS data category:

• Business demographic variables (e.g. Number of enterprises)
• "Output related" variables (e.g. Turnover, Value added)
• "Input related" variables: labour input (e.g. Employment, Hours worked); goods and services input (e.g. Total of purchases); capital input (e.g. Material investments)

Statistical Classification of Economic Activities in the European Community (NACE): NACE Rev.1 was used until 2001, NACE Rev. 1.1 since 2002, and NACE Rev 2 is used from 2008 onwards. Key data were double reported in NACE Rev.1.1 and NACE Rev.2 for 2008. From 2009 onwards, only NACE Rev.2 data are available.

The regional breakdown of the EU Member States is based on the Nomenclature of Territorial Units for Statistics (NUTS). Detailed information about the consecutive NUTS Regulations can be found at Eurostat's website

The product breakdown is based on the Classification of Products by Activity (CPA) as stated in the Regulation establishing CPA 2008 and its amending  Commission Regulation (EU) No 1209/2014  (from reference year 2015 onwards)

The SBS coverage was limited to Sections C to K of NACE Rev.1.1 until 2007. Starting from the reference year 2008 data is available for Sections B to N and Division S95 of NACE Rev.2. With 2013 as the first reference year information is published on NACE codes K6411, K6419 and K65 and its breakdown.

The statistical characteristics are defined in Annex I of Commission Regulation (EC) No 250/2009

National data refer to enterprises. The enterprise is the smallest combination of legal units that is an organisational unit producing goods or services, which benefits from a certain degree of autonomy in decision-making, especially for the allocation of its current resources. An enterprise carries out one or more activities at one or more locations. An enterprise may be a sole legal unit.   

Until the 2017 reference year, Spain identified, for operational purposes of SBS, the statistical unit Enterprise with the Legal Unit. Thus, each Legal Unit formed an Enterprise. However, the progressive complexity of the way in which enterprise groups operate internally nowadays caused the European Statistical System to search for an improvement as regards the way in which the activity of these groups is reflected in enterprises’ official statistics. Legal Units that belong to enterprise groups sometimes sell their products or provide their services exclusively or mainly within the group, without being market-oriented or having decision-making power over the entire production process.

For these reasons, and in accordance with the European Statistical System, from the SBS exercise 2018 onwards, the statistical unit 'enterprise' can be formed by one or several legal units. For enterprises consisting of more than one legal unit, new compilation methods, based on guidelines and recommendations of Eurostat, have been applied in order to measure the economic characteristics of the Enterprise, considering the set of their legal units and eliminating intra-enterprise transactions. Due to these changes in methodology, figures of SBS-2018 are affected, to a greater or lesser extent, by a  break in the time series compared to previous years. It should be noted that most of the Enterprises are independent Legal Units, so the Enterprise = Legal Unit identity remains valid. The change will only affect Legal Units that are part of Enterprise Groups.

Regional data refer to local units. The local unit is an enterprise or part thereof (e.g. a workshop, factory, warehouse, office, mine or depot) situated in a geographically identified place. At or from this place economic activity is carried out for which - save for certain exceptions - one or more persons work (even if only part-time) for one and the same enterprise.

The target population consists of enterprises (companies, societies, and natural persons) that have been active during the reference year and whose main activity is in the following sections and divisions of NACE Rev.2.:

• Industrial Sector: Sections B, C, D, E
• Construction Sector: Section F
• Trade Sector: Section G
• Services Sector: Sections H, I, J, L, M, N and Division S95

The current Regulation (EC) nº 295/2008 concerning structural business statistics groups the information in various Annexes depending on the sector of activity. This report summarizes the characteristics of the following Annexes:

• Annex I:   Services Sector
• Annex II:  Industrial Sector
• Annex III: Trade Sector
• Annex IV:  Construction Sector

Coverage: No threshold is applied regarding the size of enterprises.

All the enterprises in the entire national territory of Spain are considered.

Foreign affiliates resident in Spain are included (INWARD FATS).
Spanish affiliates abroad are excluded (OUTWARD FATS).

1998-2020

Not applicable.

• Number of enterprises and number of local units are expressed in units.
• Monetary data are expressed in millions of €.
• Employment variables are expressed in units.
• Per head values are expressed in thousands of € per head.
• Ratios are expressed in percentages.

2020

Annexes I, II, III:  The reference period for the data is generally the calendar year. Exceptionally, companies that operate seasonally or during campaigns that cover two different years and have their data recorded this way, provide the information of the season or campaign that finishes that year.

Annex IV: The reference period for the data is the calendar year.

Year 1995 was the first year for the implementation of the Council Regulation No 58/97 (SBS Regulation).

The Council Regulation No 58/97 has been amended three times: by Council Regulation No 410/98, Commission Regulation No 1614/2002 and European Parliament and Council Regulation No 2056/2002. As a new amendment of the basic Regulation it was decided to recast the Regulation No 58/97 in order to obtain a new "clean" legal text. The European Parliament and Council Regulation No 295/2008 was adopted on 14/02/2008 and the provisions of this Regulation are applicable from the reference year 2008. Regulation No 295/2008 has been amended by Commission Regulation (EU) No 446/2014.

Not applicable.

The Statistical Law No. 12/1989 specifies that the INE cannot publish, or make otherwise available, individual data or statistics that would enable the identification of data for any individual person or entity. Regulation (EC) No 223/2009 on European statistics stipulates the need to establish common principles and guidelines ensuring the confidentiality of data used for the production of European statistics and the access to those confidential data with due account for technical developments and the requirements of users in a democratic society.

Annexes I, II, III:

- Primary confidentiality: Frequency method. Data are flagged as confidential if they are reported by less than 4 enterprises.
- Secondary confidentiality: Minimize the number of suppressed cells according to the hypercube method provided by t-argus software, assuming that all the cells have the same cost.

Annexes IV: Frequency method. In 2020 there were no cases

The advance release calendar that shows the precise release dates for the coming year is disseminated in the last quarter of each year.

Annexes I, II, III: The calendar is disseminated on the INEs Internet website: 

http://www.ine.es/daco/daco41/calen.htm

Annex IV: The calendar is disseminated on the Ministry’s Internet website:

               https://www.mitma.es/informacion-para-el-ciudadano/informacion-estadistica/calendario-de-disponibilidad-de-las-estadisticas

In addition to the annual publication of the online database, INE and the Ministry for Transport, Mobility and Urban Agenda delivers to users interested in SBS other available data or details under custom requests, always protecting confidential data.

Annual

Annexes I, II (except series 2H, 2I, 2J and 2K), III: For each of the sectors (Industrial, Trade and Services), press releases with Final Results for the reference year 2020 were published on the INE website in t+18 months (June 9, 2022).

Industrial

https://www.ine.es/en/prensa/eesi_2019_d_en.pdf

Trade

https://www.ine.es/en/prensa/eesc_2019_d_en.pdf

Services

https://www.ine.es/en/prensa/eess_2019_d_en.pdf

Annex II (series 2H, 2I, 2J and 2K): For reference year t, news release is published on the INEs Internet website in t+23 months. Last news release: November 2022, for reference year 2020.

https://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736176846&menu=ultiDatos&idp=1254735976612

Annex IV: It is published on the date of publication of data. It is not issued every year.

Annexes I, II (except series 2H, 2I, 2J and 2K), III. There are no paper or electronic publications.

Annex II (series 2H, 2I, 2J and 2K). There are no paper or electronic publications.

Annex IV: Electronic publication (PDF) on the ministry website. Link to the online publication is provided in 10.3.

Annexes I, II (except series 2H, 2I, 2J and 2K), III.

        For reference year t, INE disseminates the online database in t+18 months

        Last update: June 9, 2022 for reference year 2020.

 Annex II (series 2H, 2I, 2J and 2K):

For reference year t, INE disseminates the online database in t+23 months
Last update: November 2022 for reference year 2020.

Annex IV: There is no online database

Annex I:
http://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736176865&menu=ultiDatos&idp=1254735576778

Annex II (Except Series 2H, 2I, 2J and 2K):
http://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736143952&menu=ultiDatos&idp=1254735576715

Annex II (Series 2H, 2I, 2J and 2K):
http://www.ine.es/dyngs/INEbase/es/operacion.htm?c=Estadistica_C&cid=1254736176846&menu=ultiDatos&idp=1254735976612

Annex III:
http://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736176902&menu=ultiDatos&idp=1254735576799

Annex IV:
https://www.mitma.es/informacion-para-el-ciudadano/informacion-estadistica/construccion/estructura-de-la-industria-de-la-construccion/estructura-de-la-industria-de-laconstruccion 

For these statistical operations no microdata are disseminated. Only other statistical institutions (Statistical Institutes of Autonomous Communities) are provided with microdata files. These institutions have the legal obligation to protect statistical secrecy.

Users can demand specific information and it is released, always saving the confidentiality, under custom requests.

Also, data are transmitted to Eurostat (compilation of European aggregates) and disseminated at the level of Member States in Eurostat's database.

Annual publications include methodological notes and comments on the results to facilitate understanding of the data.

A catalogue of publications is available on the INE website, including an informative bulletin, specialized journals and working documents on various statistics calculated by the Spanish statistics institute, among which data on the Structural Business Statistics could be found. This information can be consulted and downloaded in the following link:

https://www.ine.es/ss/Satellite?L=es_ES&c=Page&cid=1254735110606&p=1254735110606&pagename=ProductosYServicios%2FPYSLayout

Annexes I, II (except series 2H, 2I, 2J and 2K), III. Documentation on methodology is available on the Website (electronic version).

Methodology Structural Business Statistics Spanish and English languages
https://www.ine.es/en/metodologia/t37/metodologia_eee2020_en.pdf

Standardized Methodological Report. Spanish and English languages  
http://www.ine.es/dynt3/metadatos/en/RespuestaDatos.html?oe=30048

https://www.ine.es/dynt3/metadatos/en/RespuestaDatos.html?oe=30232

https://www.ine.es/dynt3/metadatos/en/RespuestaDatos.html?oe=30177

Annex II (series 2H, 2I, 2J and 2K): Documentation on methodology is available on the Website (electronic version).

Methodology Structural Business Statistics. Spanish and English languages
https://www.ine.es/en/daco/daco42/ambiente/metoemin_en.pdf

Standardized Methodological Report. Spanish and English languages  
https://www.ine.es/dynt3/metadatos/en/RespuestaDatos.html?oe=30066

Annex IV: Documentation on methodology is available on the Website (electronic version).

https://www.mitma.es/informacion-para-el-ciudadano/informacion-estadistica/construccion/estructura-de-la-industria-de-la-construccion/estructura-de-la-industria-de-laconstruccion

INE statistics are produced in accordance with the Code of Good Practice for European Statistics, which is the basis for the institution's quality policy and strategy.
In addition, INE has introduced a quality assessment and monitoring system made up of a series of instruments and methods allows to measure the quality of statistical work.
Some of these elements are the following:

• Quality reports and indicators. Methodological description of quality reports and indicators of the different statistical products of the INE.
• User surveys.
• Peer Review external evaluation.

For more information see the section on Quality on the INE website (in english) (link under 10.7)

Standardised Methodological Report for SBS Industry, Commerce and Services can be consulted (in english) at the links already indicated in section 10.6

These reports include, in addition to the methodological description of the statistical operation and methods for its processing, some quality indicators (Coefficient of variation, Non-response rate, Time between the reference period and the publication date of the final results, Punctuality, Comparable length for time series)

Based on regulations 275/2010 of the Commission and 295/2008 of the European Parliament and the Council, the European Commission (Eurostat) evaluates the quality of the transmitted data and publishes reports about the quality of European Statistics.

Quality assurance framework for the INE statistics is based on the ESSCoP, the European Statistics Code of Practice made by EUROSTAT. The ESSCoP is made up of 15 principles, gathered in three areas: Institutional Environment, Processes and Products. Each principle is associated with some indicators which make possible to measure it. In order to evaluate quality, EUROSTAT provides different tools: the indicators mentioned above, Self-assessment based on the DESAP model, peer review, user satisfaction surveys and other proceedings for evaluation.

The statistic development process has certain established controls to detect and correct the errors in order to guarantee its quality. Data collection in the Centralized Collection Units has serious and second level errors programmed, so as to filter the data as it is introduced by the user. In addition, during data collection and filtering, measures are taken to reduce the non-response. Finally, once the data reaches the Central Services, the data is subjected to additional micro and macro-filter controls. It is also compared to other short-term surveys and to information coming from administrative sources, which allow guaranteeing the highest quality final product.

Regulation 275/2010 of the Commission establishes the procedures to evaluate the data sent by the different member states for statistics of attachments I to IX of the SBS Regulation (EC) no. 295/2008, based on the quality criteria listed in article 12, section 1, of Regulation (EC) no. 223/2009 of the European Parliament and the Council: relevance, accuracy, timeliness and punctuality, accessibility, clarity, coherence and comparability.

The statistic may be considered high quality and it’ s based on a stratified sample design with a sample that is representative enough of the object population at national and Autonomous Community level for each size classes and activity stratum. Moreover, the data are subjected to interannual adjustment and verification criteria, as well as to comparisons with administrative register data, like the one provided in the Mercantile Register.

INE and Ministry of Development try to meet the needs of main users.

Main national users:

- Within the INE: National Accounts, Business Register (BR), Social Statistics, Environmental Statistics, Science and Technology Statistics, Price Statistics (Output Prices for Services), CIS-R&D statistics, ICT statistics, short-term statistics, foreign direct investment etc.
- Other national users: Ministries and other public institutions, Central Bank, Chamber of Commerce, Associations for economic activities in question, Institute of Economic Research and other research institutes, Universities, Students, Media, Businesses etc.
- Autonomous Communities
- In general anyone interested in the economic structural analysis of Industry, Construction, Trade and Services.

External users of SBS: Eurostat (European Commission), OECD, etc..

Conceptual differences between SBS data published at national level and data sent to Eurostat:

• Annexes I, II (series 2H, 2I, 2J and 2K), III y IV: No differences.
• Annex II (except series 2H, 2I, 2J and 2K): From 2015 onwards, no differences

Until the reference year 2014 included, industry data published by INE at national level were presented sectorized (100 industrial subsectors adapted to Spanish industry) according to the main activity of the local units of the industrial enterprises. From 2015 onwards, to harmonize industrial statistics with those of other sectors and with Eurostat SBS data, the criterium has been changed and data are already presented broken down by activities depending on the main activity of the enterprise

Additional indicators published at the national level to fulfill the national needs:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III y IV: Productivity, average salary, rate of value added, rate of personnel costs, rate of investment, rate of employees, female participation rate for employees.
• Annex II (series 2H, 2I, 2J and 2K):  No additional indicators

Annexes I, II (except series 2H, 2I, 2J and 2K), III. The INE conducted three general user satisfaction surveys in 2007, 2010, 2013, 2016 and 2019, and plans to perform them regularly. These surveys try to obtain an average user rating of the quality of the information sector, taking into account aspects such as relevance, accuracy, punctuality, consistency and accessibility. According to satisfaction survey 2019, statistics from industry and services have a positive or very positive assessment of users in terms of relevance, accuracy, consistency and comparability.

          Satisfaction surveys 2019 http://www.ine.es/en/ine/codigobp/InformeEncuesta2019_en.pdf

Annex II (series 2H, 2I, 2J and 2K): The user satisfaction surveys conducted to date assess the Environmental sector (environmental statistics and accounts) in which the statistical operation is set in, provide direction with regard to the opinion users have of it.

Annex IV: Not available

_______

Number of accesses in 2021 to the on-line databases:

• Annex I: 70,943
• Annex II (except series 2H, 2I, 2J and 2K): 83,832
• Annex III: 52,038
• Annex IV: Not available

The completeness is the extent to which data are available compared with the requirements in terms of characteristics, geographical and activity breakdown, as specified in the SBS Regulation.

Availability rate

• Annexes I, II, III and IV. Availability rate: 100%

Use of the quality flag ‘Contribution to European totals only (CETO-flag) foreseen in the SBS Regulation

Until the reference year 2015 included, in the Trade and Services Sectors, certain 4-digit activities of the NACE rev.2 were classified as CETO (Contribution to European Totals Only) when designing the samples for these sectors. This measure, provided for in Article 8.3 of SBS Regulation 295/2008 and applicable only to a limited number of activities with a smaller weight on the totals of each sector, made it possible to mitigate as far as possible the increase in sample sizes. This implied that the results of these activities were only representative for the elaboration of aggregates at the European level and the commitment not to publish such data at the national level and, therefore, not at a more disaggregated level of Autonomous Community.

From reference year 2016 included, thanks to the reform and optimization of the sample designs, all the activities to 4 digits of the NACE rev.2 of the sectors under study have their corresponding sample support, so it has been possible to cease applying the CETO figure. Thus, from year 2017 onwards, the publication of the final SBS data offers, for all activities not affected by confidential data, statistical results up to the 4-digit level of the NACE rev.2.

The sampling design tries to minimize sampling errors and the different survey processes are intended to eliminate or decrease its errors in the collection stage (response rate and filter control) as well as in edition and imputation stages.

Collection, coverage control, error filtering and non-response imputation allow obtaining a highly reliable statistic

Annexes I, II (except series 2H, 2I, 2J and 2K), III.

Annex II (series 2H, 2I, 2J and 2K):  Sampling errors are calculated of the main variables, at different breakdown levels, using the Direct Method (E. Raulin Formula) and the Indirect method (Jackknife).

Annex IV: Not calculated

The unit non-responses

Methods used for taking into account the unit non-response:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: Since reference year 2016 all the non-response is subject to automatic imputation (based on administrative sources and/or information from previous year and/or other kind of imputations) which allows it to be borne in mind as an effective sample.
• Annex II (series 2H, 2I, 2J and 2K): we use available information from previous reference period, auxiliary information sources (companies reports), auxiliary variables calculated from other surveys and the corrector factor in the weighting procedure for imputation.
• Annex IV: Corrector factor in the weighting procedure for imputation.

Measures taken or envisaged for minimizing the unit non-response:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: Data collection Units make reminders by post, by email and by telephone. Different questionnaires for the small enterprises (short questionnaire) are used, in order to reduce the statistical burden for them and favor the response. The enterprises that refuse to answer the questionnaire are fined.
• Annex II (series 2H, 2I, 2J and 2K): Data collection Units make reminders by post, by email and by telephone. The enterprises that refuse to answer the questionnaire are fined.
• Annex IV: Agent special team.

Weighted unit non-response rate

 Characteristics used for weighting non-response:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: Persons employed
• Annex II (series 2H, 2I, 2J and 2K): Number of enterprises  
• Annex IV: Number of enterprises

 The recorded unit non-response rate in the overall context is considered as:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: low for enterprise of 10 or more person employed. On the average for enterprises below 10 persons employed.
• Annex II (series 2H, 2I, 2J and 2K): low
• Annex IV: low.

Bias

The bias resulting from non-response is considered as:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: Small bias (with a limited effect in the overall accuracy of the estimate).
• Annex II (series 2H, 2I, 2J and 2K): Small bias
• Annex IV: Small bias

Evaluation of the impact of imputation on CVs

The impact of imputation on CVs is considered as:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: Not important
• Annex II (series 2H, 2I, 2J and 2K): low
• Annex IV: Not important

Coverage errors- Frame

The impact of imperfection of the relevant business register on the quality of the key statistics is considered as:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: low
• Annex II (series 2H, 2I, 2J and 2K): low
• Annex IV: low

Out-of scope units

The out-of scope units are detected in collection data process itself and in the updating process of the frame. Those units represented less than 1 %  

Annexes I, II (except series 2H, 2I, 2J and 2K), III:  

Data-collection:                       12/2021
Post-collection phase:              06/2022
Dissemination in the country:   06/2022

Annex II (series 2H, 2I, 2J and 2K): 

Data-collection:                        12/2021                       
Post-collection phase:               06/2022
Dissemination in the country:    11/2022

Annex IV:

Data-collection:                        10/2021                       
Post-collection phase:               12/2021
Dissemination in the country:    12/2021

All Spanish SBS series for reference year 2020 has been sent to Eurostat on time (before t+18 months)

The availability of a methodology, a common collecting design and process, filtering, edition and elevation in its entire geographical scope, guarantees comparability of results among the different Autonomous Communities. Moreover, the methodological adjustment to the SBS regulation allows comparability with the rest of European Union countries which disseminate the same information for their respective sectors.

Length of comparable time series

• Annexes I, II, III: 1998-2008; 2008-2017, 2018-2020 (full implementation of the statistical unit Enterprise)
• Annex IV: 2001-2007, 2008-2017, 2018-2020 (full implementation of the statistical unit Enterprise)

Reasons or differences in concepts and methods of measurements for break in time series

2008. Change in NACE classification (from NACE Rev. 1.1 to NACE Rev. 2); 2008 is the double reporting year. 

2018: Full implementation of the statistical unit Enterprise

Comparisons are made with other data sources or statistical domains (business register, business demography, short term statistics, labour statistics, etc.).

There is some inconsistency between BD and SBS regarding the number of enterprises, mainly due to the different treatment for inactive enterprises. There is an ongoing project at the INE to achieve full coherence between both statistical domains.

Regarding other sources, in general the figures are consistent but, due to methodological differences, some discrepancies are possible.

The results of national and satellite accounts, such as environmental accounts, and SBS results may be different because different sources are taken into account in synthesis statistics.

Sometimes the direct comparison of data from different sources is relevant, but in other cases, the analysis of the evolution of the variables is more coherent.

Coherence is an essential matter for the methodological approach of the survey as well as for its entire development process. Coherence among variables is compared in all stages of the statistical process. The aggregates are always consistent with their main sub-aggregates (e.g. the total for different size classes, NACE, NUTS, CPA etc.)

Cost

The estimation of the budgetary credit necessary to finance these statistics, as foreseen in the 2022 Annual Programme, comes to a total of (Thousand euros):

• Annex I: 2888.98
• Annex II (except series 2H, 2I, 2J, and 2K): 1716.27
• Annex II(series 2H, 2I, 2J, and 2K): 435.29
• Annex III: 2053.28
• Annex IV: 544.47

Activities undertaken to reduce the respondents’ burden

• Annexes I, II (except series 2H, 2I, 2J and 2K), III:

- From reference year 2012 onwards, SBS for Industry, Services and Trade sectors are compiled by combination of statistical surveys and administrative data

- Formulas for cooperation are established in order to make the best use of the information available and to avoid unnecessary duplication of statistical operations. In the sample selection process, negative coordination methods are applied, that is to say, minimal overlap possible (covering the survey's objectives) between surveys samples that coexist in the same period.

- Use of different models of questionnaires with different levels of simplification according to the characteristics of each company. Thus, small companies usually fill out a questionnaire with fewer variables and breakdowns. In all the questionnaire models we try to reduce the breadth of the questionnaires to the bare minimum.

- The completion through the internet is enhanced using electronic questionnaires with questions, flows, and validations adapted to the respondents, which facilitates the completion and improves the quality of the data collected. The design of the questionnaires for the Internet are posed in such a way that they can avoid or at least minimize subsequent re-contacts with the respondents

- Some variables related to employment variables are estimated using the administrative data of the Social Security system.

•  Annex II (series 2H, 2I, 2J and 2K): 

The design and selection of the sample have been coordinated with other surveys, so as to reduce as much as possible the overload of certain respondent units caused by the completion of several questionnaires during the same year.
In terms of the data collection, online completion is encouraged as a method that is easy and secure for respondents, and which enables achieving better information quality levels. In any case, respondents have other possible response channels available (print questionnaire by post, fax).
In order to facilitate the joint management of the response of their companies, the enterprise groups contact with a special statistic department to collect the data.

•  Annex IV: Simplification of questionnaires. Cooperation with other organizations. Help in completing

We do not foresee revisions after the sending of final data to Eurostat and the release at national level

Preliminary data versus final data

• Annexes I, II, III y IV: Preliminary data was obtained from a provisional sample of enterprises. The final data was adjusted to the definitive sample of enterprises.

Statistical operations

• Annexes I, II (except series 2H, 2I, 2J and 2K), III:

Until the reference year 2014 included, the structural business statistics produced by the INE (Industrial, Trade and Services Sectors) were elaborated according to different surveys, through separate processes and with different collection and publication schedules.   

Since the reference year 2015 and with the aim to standardize and improve the process of elaboration of these operations, the statistical results and the dissemination of the information, the ‘Integration Project of the Structural Business Statistics’ was put into effect. The project, developed in two consecutive exercises,  has signified a reform of the following statistical operations, which have changed their denomination to emphasize even more the idea of integration:

- Structural Business Statistics: Industrial Sector (previously named 'Industrial Companies Survey')
- Structural Business Statistics: Trade Sector (previously named 'Annual Trade Survey')
- Structural Business Statistics: Services Sector (previously named 'Annual Services Survey')

The reform includes, among other aspects:

1) the use of practically uniform questionnaires for the three sectors under study (keeping some specific issues for the Industrial Sector) and adapting to the regulated models for the Annual Accounting for the juridical persons;
2) more efficient sample design with the purpose of reducing the burden on the reporting parties and obtaining a better use of the administrative registers;
3) a new and homogeneous release plan for each operation that eases the comparison of the statistical results for the different sectors.
4) improvements relating to the treatment of companies that enter or leave the target population during the reference year, as well as the imputation of non-response. Additionally, the estimation of employment variables has been improved by using administrative sources.

       From reference year 2012 onwards, SBS for Industry, Services and Trade sectors are compiled by combination of statistical surveys and administrative data. The proportion of sample units obtained from administrative sources has been increasing progressively. For 2020 reference year, more than 39% of the sample units of industry, commerce and services are obtained from administrative sources, thus avoiding the statistical burden of said units.

• Annex II. Series 2H, 2I, 2J and 2K. Information collected in the Survey on Expenditure on Environmental Protection
• Annex IV. Information collected in the Structural survey on Construction

Data source:

• Annex I: Statistical survey combined with administrative sources
• Annex II (except series 2H, 2I, 2J and 2K): Statistical survey combined with administrative sources
• Annex II (series 2H, 2I, 2J and 2K): Statistical survey
• Annex III: Statistical survey combined with administrative sources   
• Annex IV: Statistical survey combined with administrative sources

Type of sample design:

• Annexes I, II , III, IV: Stratified

Stratification criteria:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III, IV: Activity, Employment size class, Region. 
• Annex II (series 2H, 2I, 2J and 2K): Activity. Employment size class, Groups of regions.

Selection schemes and sampling rates:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III, IV:

            Legal units with 50 or more persons employed are investigated exhaustively.

            For legal units with less than 50 persons employed, stratified sampling with optimal allocation (Neyman) is used, with the following samplig rates.
              - Annex I: 5.2%  
              - Annex II (except series 2H, 2I, 2J, and 2K): 21.4%   
              - Annex III: 8.6%   
              - Annex IV: 4.9% 

• Annex II (series 2H, 2I, 2J, and 2K): sampling rate 24.3% and legal units with 200 or more persons employed are investigated exhaustively.

Threshold values:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III, IV: No thresholds are applied.
• Annex II (series 2H, 2I, 2J and 2K): Establishments with more than 10 employees. In order to meet the total requirements of SBS we use model–based techniques.

The effective sample size and final sampling rates by employment size class (in % of population):

• Annex I: 90,180 of which, based on administrative sources 41.1%

              Sampling rate: Total: 5.4%    By persons employed classes: 0-9: 3.7%   10-49: 32.4%   50 or more: 99.7%   

• Annex II (except series 2H, 2I, 2J, and 2K): 46,328 of which, based on administrative sources 34.1%

              Sampling rate: Total: 23.6%  By persons employed classes: 0-9: 14.4%   10-49: 62.2%   50 or more: 99.7%

• Annex II (series 2H, 2I, 2J, and 2K): 7,948

• Annex III: 64,838 of which, based on administrative sources 40.8%

              Sampling rate: Total: 8.8%   By persons employed classes: 0-9: 6.5%    10-49: 47.4%    50 or more:100%    

• Annex IV: 12,891

              Sampling rate: Total: 3.1%  By persons employed classes: 0-9: 2.1%    10-49: 19.0%    50 or more: 97.3%  

Administrative sources

For SBS the following administrative sources are used:

   - Tax data: Corporation Tax, VAT Tax, Income Tax
   - Social Security
   - Business Register
 

These administrative data are subject to several revisions with (increasing) degree of completeness.
The frequency to which these used administrative data are updated is annual. For the purposes of SBS we assess this frequency as very good.

Characteristics directly available or with a proxy in the administrative source:

   -  Number of employees
   -  Turnover
   -  Production Value
   -  Gross margin on goods for resale
   -  Value Added at Factor Cost
   -  Gross Operating Surplus
   -  Purchases of goods and services purchased for resale in the same condition as received
   -  Wages and Salaries
   -  Social Security Costs
   -  Change in stocks of goods and services for resale
   -  Change in stocks of finished products and work in progress
   -  Gross investment in land
   -  Gross investment in existing buildings and structures
   -  Gross investment in construction and alteration of buildings
   -  Gross investment in machinery and equipment

 
Kind of administrative data accessed: Micro data

Relation between reporting units and the legal units / enterprise (statistical unit):

• Annexes I, II (except series 2H, 2I, 2J and 2K), III, IV:

  Reporting unit = legal unit (in some exceptional cases, the reporting unit is the head of the group)

  Statistical unit = enterprise (the enterprise can be formed by one or several legal units)

• Annex II (series 2H, 2I, 2J and 2K): Reporting unit = local unit (an enterprise or part thereof situated in a geographically identified place )

Frame

   -  The variable used for identifying principal and secondary activities is the Value added. When value added is not available, turnover is used.  
   -  The method used for identifying activities is the Top-down method.
   -  Frequency of updating the unit's principal activity (stability rules): Yearly 
   -  Frequency of updating the business register in our NSI: Yearly

Annual data collection

Data are collected by means of statistical surveys, business register or from various administrative sources.

• Administrative sources:  Information is collected from the administrative sources indicated in the section 18.1

• Statistical surveys: Information is requested by post and the questionnaire is encouraged to be completed online, which does in fact make up for 90% of the total collected questionnaires. In any case, the respondents have other response channels available (fax, postal mail, telephone). The INE Data Collection Units are in charge of the telephone service for solving doubts the respondents may have and are also responsible for the recording and filtering of questionnaires. Anomalous errors and data are subject of revision, this way establishing whether it is necessary to re-contact the respondents by telephone.

Description of data checking and editing:
-  completeness checks (data integrity rules)          
-  validity checks (internal consistency)                        
-  plausibility checks
We check the plausibility of data, namely by contacting directly the units.

The management of the questionnaires (reception, treatment, recording, validation ...) is carried out continuously by the same Data Collection Units establishing the necessary control to guarantee an appropriate quality level for the whole process. This facilitates controlling, already in this phase, the errors that may affect the data obtained from respondent units.

Additionally, during the data collection period, a coverage control of the information is carried out by the Production Unit, in order to guarantee the completeness of the recorded data, detect duplicates and coverage errors, and, at the same time, be able to carry out a first evaluation on the quality of the variables collected in the questionnaires.

During the micro-filtering process, the detection and filtering of errors and inconsistencies in the identification variables of each register as well as the filtering and imputation of content errors are carried out. Once the micro filtering phase has finished, we proceed with the tabulation and dissemination of results, obtaining the analysis tables in order to eliminate errors and inconsistencies detected via macro filtering techniques.

Use of statistical surveys and administrative sources

• Annexes I, II, III: For a subset of the sample of Industry, Trade and Services Sector, characteristics listed in 18.1 are estimated using administrative sources. The other characteristics, not available in administrative sources, are estimated using model based estimation. A ratio imputation is used, based on information from previous reference period, and previous and new administrative data for this subset of the sample. Then, this subset of the sample is integrated with those obtained though the statistical surveys.

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: The estimation of employment variables has been improved by using administrative sources.
• Annex IV: For a small subset of the Construction Sector sample, characteristics are estimated using administrative sources and, if unavailable in administrative sources, they are estimated using a ratio imputation

 Description of imputation methods used for dealing with unit non-response:

• Annexes I, II (except series 2H, 2I, 2J and 2K), III: Since reference year 2016 all the non-response is subject to automatic imputation (based on administrative sources and/or information from previous year and/or other kind of imputations as average per stratum, nearest neighbor...) which allows it to be borne in mind as an effective sample.
• Annex II (series 2H, 2I, 2J and 2K): If no data are available, we use available information from previous reference period. For invalid and inconsistent responses we use different methods of imputation: average per stratum, same distribution as previous year… For the non-response of other enterprises corrector factor in the weighting procedure is used.
• Annex IV: Corrector factor in the weighting procedure.

 Inference (grossing-up)

• Annexes I, II and III: We have applied non-biased estimators in stratified sampling with correction of the population for grossing-up the figures covered by the SBS Regulation.
• Annex IV: Corrector factor in the weighting procedure.

Not applicable

Since reference year 2016, various methodological improvements have been implemented on the Structural Business Statistics for Industrial, Trade and Services Sectors, within the framework of the "Integration Project of the Structural Business Statistics" (see section 18.1). These improvements are related to the treatment of companies that enter or leave the target population during the reference year, as well as the estimation of the non-response units. On the other hand, the estimation of employment variables has also been improved using administrative sources.

From the SBS exercise 2018 onwards, the statistical unit 'enterprise' can be formed by one or several legal units. For enterprises consisting of more than one legal unit, new compilation methods, based on guidelines and recommendations of Eurostat, have been applied in order to measure the economic characteristics of the Enterprise, considering the set of their legal units and eliminating intra-enterprise transactions