Federal Statistical Office of Germany (DESTATIS)

E 21 "Business indices, seasonal adjustment"

Statistisches Bundesamt
Gustav-Stresemann-Ring 11
D-65189 Wiesbaden
Germany

This indicator is part of the overall STS indicator system which supports business cycle observation. This index of persons employed shows the evolution of employment in the enterprises active in “other services“ (according to annex D of the STS-regulation).

NACE Rev. 2 (Statistical Classification of Economic Activities in the European Community)

NACE Rev. 2 Divisions G45 and G46, Sections H, I, J, M_STS (M without 70.1, 72 and 75) and N_STS (N without 77, 81.1 and 81.3)

The objective of the index of persons employed is to show the development of employment. The number of persons employed is defined as the total number of persons who work in the observation unit (inclusive of working proprietors, partners working regularly in the unit and unpaid family workers), as well as persons who work outside the unit who belong to it and are paid by it (e.g. sales representatives, delivery personnel, repair and maintenance teams).

In the respective NACE codes for "other services" mix-models are applied, where different thresholds (depending on NACE code) define whether a unit is covered by primary survey or by administrative data. If the unit is covered by administrative data, the only information available is employees viable for social security (no self-employed, no minor employment). In accordance with the requirements for Business Demography self-employed persons are imputed to complement the administrative data. The thresholds below which administrative date is used are in NACE-division 45: 10 million Euro annual turnover or 100 employees; in NACE division 46: 20 million Euro annual turnover or 100 employees; in NACE section I: 150,000 Euro annual turnover; in all other NACE codes: 15 million Euro annual turnover or 250 employees.

Legal unit

All relevant statistical units in the NACE-sections mentioned under section 3.3 according to the Business Register: ~1.8 million units (2014)

Country as a whole

Labour input outside the geographical territory of Germany is not included in the data.

Since 1980 for Divisions 45 and 46 as well as Section I; since 2003 for the remaining NACE codes of "other services"

2015 = 100

Index based on the evolution of employment in "other services"

Month in Divisions 45 and 46 as well as Section I; calendar quarter in the remaining NACE codes of "other services"

The laws regulating the collection, treatment and dissemination of statistical data are the Council Regulation (EC) No 1165/98 of 19 May 1998 concerning short-term statistics with implementing acts (e.g. Commission Regulation No 1503/2006 of 28 September 2006 concerning the definitions) as well as the Federal Statistics Act (1987), the national Trade Statistics Act (HdlStatG of 2001), the national Services Statistics Act (DlKonjStatG of 2013) and the national Law governing the use of administrative sources for Statistics (VwVDG).

Data sent to Eurostat may contain confidential data to compile the European aggregate.

Confidentiality is guaranteed by section 16 of the Federal Statistics Act. It requires that respondents must be informed about their rights and obligations in providing statistical information. This rule is strictly followed. It also applies to the notification that individual data are only used for statistical purposes and - subject to protection procedures - for scientific purposes.

Furthermore, Chapter V “Statistical Confidentiality” of the REGULATION (EC) No 223/2009 of the European Parliament and of the Council applies.

As regards the confidentiality of data, methods of primary and secondary confidentiality checking have been developed and are being used.

A continuously updated release calendar containing the precise release dates for the reference periods of the following year is published on the website of the FSO always in october.

https://www.destatis.de/SiteGlobals/Forms/Suche/Termine/EN/Terminsuche_Formular.html?nn=43216 

Data are available to the public through Eurostat’s Data-Explorer and the national online-database GENESIS. The public is informed by a press release that is published on the homepage of DESTATIS. Additionally, some main users are informed via email at the same point of time (publication date) and there is a possibility to use RSS (Really Simple Syndication) for information about new data.

Monthly for Divisions 45 and 46 as well as Section I; quarterly for all other NACE codes

Normally 14 days after the latest transmission date to Eurostat (t+60) a press release is published online for NACE division 45 and 46 as well as NACE section I.

publications are linked on the FSO website www.destatis.de

https://www-genesis.destatis.de/genesis/online/data?operation=sprachwechsel&language=en

search for codes 45211, 45214, 45213 and 47414

Not available

The data are sent to Eurostat be released also as national data.

Important methodological papers are published in “Wirtschaft und Statistik” (Economics and Statistics), a monthly publication of DESTATIS, that is published through the homepage of DESTATIS.

Standardized quality reports are published on the website of the FSO

https://www.destatis.de/EN/Methods/Quality/_node.html

The data of German official statistics enjoy a reputation of high reliability in Germany and abroad. The FSO has committed itself to assure and further enhance the level of quality already attained. In order to achieve this task – also in the ESS – the activities of the FSO are based on quality guidelines. To assure quality in the long term, the FSO applies a variety of systematic quality assurance measures, among others the European Statistics Code of Practice.

Detailed descriptions can be found here:

https://www.destatis.de/EN/Methods/Quality/_node.html

The survey is conducted in a decentralized manner. The statistical offices of the Länder are responsible for collecting, checking and processing the respective Länder data. The Federal Statistical Office is entrusted with methodological preparations and the development of the statistics, and it collects the Länder results in order to produce and publish federal results as well as to calculate indices.

National users are especially the Federal Government, the German Central Bank, economic research institutes, universities, trade associations, the media, and also the interested public. The most important European user is Eurostat.

The user satisfaction is measured on different ways:

• Number of downloads/clicks
• Monthly clicks on the topics on Destatis-homepage,
• Quarterly downloads on the on-line database GENESIS,
• Downloads of the Statistical Yearbook sorted by chapters,
• Number of sold printed products or number of downloads of special publications,
• Number of external queries via the costumer management system (CMS) and other analyses of the CMS,
• customer satisfaction surveys (2007 and 2011).

STS requirements according to EU regulations and national laws are fulfilled.

Main sources for errors/potential for bias:

Survey data:

- Unit- and item-non-response

- Over- and under-coverage, incorrect classified units in the sample frame

- Implausible data

- Incorrect signing of the non-response-case

- Inappropriate estimation of missing values

Administrative data:

- Different observation units (e. g. local units)

- Incorrect classification

- Implausible data

- Deviations in the definitions

- Modifications of the legal basis of the administrative data or in the delivery-format

- Lack of timeliness and completeness (dependence)

- No unique identifier (statistical identity number differs from the identity number of the administrative data source)

- Metadata not up-to-date for linking with the Business Register

Divisions 45/46: With the switch from a stratified random sample a mix-model (combination of survey and administrative data) in 2012 the burden, especially for small and medium-sized enterprises, was reduced completely.

Section I: The cut-off limit of 50,000 euros annual turnover was increased to 150,000 euros. About 2,700 companies were relieved from October 2011 of monthly reporting requirements.

Remaining NACE codes of "other services": With the switch from a stratified random sample to a mix-model (combination of survey and administrative data) in 2007 the burden, especially for small and medium-sized enterprises, was reduced completely. Since then, the number of the units in the survey decreased to one-tenth. The quality of the results did not suffer from that. Other positive effects are timelier responses and a better response rate.

Only relevant for section I: The relative standard error for the reporting period September 2009 - September 2011 for the characteristic fulltime-employees was:

Typical non-sampling errors occur, like unit- and item-non-response, wrong classified units, missing units or inactive units in the sample frame (Business Register) and so on. Cases of non-unit and non-item response are not of such importance than in other statistics, because missing values have to be estimated. In the monthly wholesale trade statistics, the proportion of estimates in the primary data collected is decreased by switching to the Mixmodell from 12% to 1%.

Additionally actions to increase the rate of response are used. Those are: written reminders and telephone queries for instance. Besides, there exists a legal obligation for the observation units to respond. Fines can be imposed, if they do not respond.

Results are transmitted to Eurostat 60 days after the end of the reference-quarter.

The monthly publication for Section I is available 45 days after the end of the reference period. The national publication for Section H, J, M and N is approximately 14 days after transmission to Eurostat.

All publication deadlines are met.

Results for federal states (Bundesländer = NUTS 1) are directly comparable to each other.

Since 1994 for Divisions 45 and 46 as well as Section I; Since 2009 for the remaining NACE codes of "other services"

SBS and STS data (change rates) are not completely coherent. The reasons are the different purposes and hence different methods used for the statistics.

All data are checked for internal consistency. The administrative data alone would not meet the European requirements completely. This is the reason for combining administrative data with survey data of the “big enterprises”. It can also be stated that the completeness of data as well as the quality increases the later the data are obtained.

Wholesale trade (G46):

• Costs per year in 1,000 EUR              643

• Number of cases per year                  72,000

• Total time per case in min.                n.a.

• Number of variables per case             3 (turnover, fulltime-employee, parttime-employee)

Source: Standard Cost Model (effective 01/01/2012)

Accomodation and food service activities (Section I)

• Costs per year in 1,000 EUR              1393

• Number of cases per year                  91,200

• Total time per case in min.                n.a.

• Number of variables per case             3 (turnover, fulltime-employee, parttime-employee)

Source: Standard Cost Model (effective 01/01/2012)

remaining NACE codes of "other services"

• Costs per year in 1,000 EUR              643

• Number of cases per year                  20,000

• Total time per case in min.                46

• Number of variables per case             2 (turnover and persons employed)

Source: Standard Cost Model (effective 01/01/2012)

Data for short-term business statistics are revised, when additional information becomes available. If errors are detected, they are corrected immediately. Major revisions and changes in methodology are described in “Wirtschaft und Statistik” (Economics and Statistics).

The revision policy is described on the FSO website.

https://www.destatis.de/EN/Methods/Quality/GeneralRevisionsPolicy.html

The same revision policy applies to data sent to Eurostat as to data that is published nationally.

Data revisions are necessary due to new information, methodological changes or if errors are detected. Detected errors are corrected immediately, while new information is included through regular revisions.

In NACE Divisions 45 and 46 as well as Section I data is corrected every month. Corrections are possible up to 24 months from the reference period. In all other NACE codes the regular revisions affect the two quarters before the current quarter. The impact of these regular revisions is in general relatively low.

Section I: sample survey.

Sections G, H, J, M and N:

The data set is based on a mix-model, that is a combination of a statistical survey and administrative data. Legal units with:

- division 45: at least 100 persons employed and/or at least 10 million annual turnover (in the business register)  

- division 46:at least 100 persons employed and/or at least 20 million annual turnover (in the business register)

- sections H, J, M and N: at least 250 persons employed and/or at least 15 million annual turnover (in the business register)

are surveyed primary. For all the other units administrative data are used. Therefore, it is a full coverage. The units that are primary surveyed are selected once a year.

The administrative data source for persons employed is the number of employees obliged to pay social insurance in the register of the Federal Employment Agency. There are some differences in the definition (e.g. working proprietors, unpaid family workers and civil servants are missing) and in the observation unit, because of the different purposes. The unit is not the enterprise/legal unit but the local unit. Therefore, the number of employees has to be aggregated to the relevant statistical units.

Monthly for Division 45 and 46 as well as Section I; quarterly for the remaining NACE codes of "other services"

Since August 2013 the sample units have to respond on-line. Two methods to report on-line are provided: IDEV and eSTATISTIK.core. IDEV is an internet questionnaire, while eSTATISTIK.core is an application that is based on a special statistical software component. With eSTATISTIK.core enterprises are able to extract statistical raw data automatically from their corporate accounting systems and transfer the compiled data via the Internet to the responsible Statistical Office.

The obligation to provide information is on the owners or directors of the legal unit. Fines can be imposed if they do not response within the period fixed or with the provided reporting systems.

The validation (plausibility checks) is done by the statistical offices of the Länder with a standardised It-Program.

Plausibility checks on micro data level are implemented. There are different checks for survey data and administrative data. For survey data the plausibility checks can be divided in “can-be-errors” (could be an error) and “must-be-errors” (is definitely an error). The “must-be-errors” have to be cleared before the compilation of the indices starts, otherwise the process stops.

Additionally, the results are checked on the macro data level by comparing change rates of unadjusted as well as adjusted indices.

For Division 45 and 46 as well as Section I: Non-response or missing values due to late response are estimated as part of the monthly collection round. The values of enterprises that provide late declarations are estimated as part of the automatic plausibility checks using the relation between the declarations of the preceding three months and the same month one year earlier. These estimates are then replaced by the correct values when the late declaration is received.

For the remaining NACE codes of "other services": Indices are compiled by the change rate to the quarter before on the 4-digit-level of the NACE Rev. 2. The aggregation to higher levels takes place by using SBS-weights.
To calculate the change rate it is important to include only units with data in both quarters (the current and the previous quarter). Therefore, missing values are estimated. If a unit may respond later the estimated value is replaced by the real value within the regular revisions.

Calendar and seasonal adjustment based on X12-ARIMA for Divisions 45 and 46 as well as Section I; Calendar and seasonal adjustment based on X13 in JDemetra+ 2.2 for all other NACE codes

None