STATISTICS AUSTRIA

Directorate „Macro-economic Statistics“, Section „Prices and Purchasing Power Parities“, Project „Producer Price Index for Business Services“

Guglgasse 13, 1110 Wien, Austria

The Services Producer Prices Index (SPPI); (Erzeugerpreisindex für unternehmensnahe Dienstleistungen) is defined as an output price index for the service production of resident producers providing business to business (b2b) services. The export of services is included whereas imported services are out of scope. The surveyed price should exclude VAT but take into account any applicable discounts and surcharges.

NACE Rev. 2.

Apart from Group 50.1 and 50.2 “Sea and coastal water transport - passenger and freight” all SPPI activities mentioned in Council Regulation (EC) No. 1165/98 and the amending Council Regulation (EC) No. 1158/2005 concerning short-term statistics are covered.

• 49.4: Freight transport by road and removal services;
• 51: Air transport;
• 52.1: Warehousing and storage;
• 52.24: Cargo handling;
• 53.1: Postal activities under universal service obligation;
• 53.2: Other postal and courier activities;
• 61: Telecommunications;
• 62: Computer programming, consultancy and related activities;
• 63: Information service activities;
• 69: Legal and accounting activities;
• 70.2: Public relations and communication activities;
• 71: Architectural and engineering activities; Technical testing and analysis;
• 73: Advertising and market research;
• 78: Employment activities;
• 80: Security and investigation activities;
• 81.2: Cleaning activities.

Size classes covered: All size classes of enterprises are included. The branch leaders are covered by selecting the reporting units by cut-off sampling. By drawing an additional simple random sample all other size classes could be part of the survey. Very small enterprises up to a certain treshold are excluded.

The following price determining and service specifying variables are taken into account:

Name of the service: the branch’s or enterprise’s name for the selected service product;

Description of the service: a very detailed description of the service which assures that the respondent is able to re-price the identical service product over time;

Units sold: e.g. in the case of the calculation of a unit value where quantity data are needed for the calculation of a unit price (e.g. minutes or SMS);

Qualification category: when time based pricing methods are used the price may depend on the qualification of the service provider (e.g. partner, senior partner);

Discounts: all discounts should be reported and specified;

Surcharges: all surcharges should be reported and specified.

Reporting unit: prices are provided by enterprises.

Observation unit(s): enterprise.

Transactions of national enterprises concerning business-to-business services for all SPPI activities mentioned in the national Regulation concerning the compilation of price indices in the enterprise sphere (Federal Law GAZETTE II No. 147/2007 as amended by BGBl. II No. 36/2009,  BGBl. II No. 263/2012 and BGBl. II No. 222/2015) in its current version. The statistical population is represented by the NACE codes mentioned under 3.3. The enterprises for the survey are chosen by approved sampling methods (cut-off sampling and if necessary with additional simple random sampling) in the respective service sectors from national business register. If enterprises drop-out during the reporting period, they are replaced by appropriate enterprises regarding turnover and other criteria.

All regions of Austria are covered. Export of services is not covered.

Since the first quarter of 2007 the SPPIs are compiled on a quarterly basis with the respective base/reference years. After the revision in 2015 SPPIs are computed for the period 2015–2020 with the common base/reference year 2015. To compare long-term price developments it is possible to link older index series with the price trends of subsequent index series. Linking coefficients, available on the website, can be used to continue index series for indices of the current period (2015=100) on base/reference year 2010 and 2006.

The current base year for Eurostat is 2021

The current reference year for Eurostat is 2021 (Average 2021=100).

(The current base year for national publications in Austria is 2021. The current reference year for national publications in Austria is 2021 (Average 2021=100).)

SPPIs are nationally published and transmitted to Eurostat as indices. There are no absolute figures published.

As mentioned in the national Regulation concerning the compilation of price indices in the enterprise sphere (Federal Law GAZETTE II No. 147/2007 in its current version) price data refer to the 15th of the second month of each quarter.

• The "Federal Statistics Act 2000" ("Bundesstatistikgesetz 2000", Federal Law Gazette I No. 163/1999) passed in August 1999 is in force since January 2000. The "Bundesstatistikgesetz 2000" was consolidated to the "Bundesstatistikgesetz 2000 - konsolidierte Fassung 2010" (BGBl. I No. 163/1999 as amended by BGBl. I No. 136/2001, BGBl. I No. 71/2003, BGBl. I No. 92/2007, BGBl. I No. 125/2009, BGBl. I No. 111/2010) in its current version.

• (EBS) Regulation (EU) 2019/2152 on 27 November 2019 and the Commission implementing Regulation 2020/1197

• The national Regulation concerning the compilation of price indices in the enterprise sphere (Federal Law GAZETTE II No. 147/2007 in its current version.

Respondents are obliged to respond to the survey by Federal Law GAZETTE II No. 147/2007 in its current version.

SPPIs are not transmitted to any other international organisation apart from Eurostat.

The legal basis is:

• The general obligation to publish statistics and the strict provision on statistical confidentiality is regulated by the “Federal Statistics Act 2000” ("Bundesstatistikgesetz 2000") in the consolidated version of "Bundesstatistikgesetz 2000 - konsolidierte Fassung 2010” (BGBl. I No. 163/1999 as amended by BGBl. I No. 136/2001, BGBl. I No. 71/2003, BGBl. I No. 92/2007, BGBl. I No. 125/2009,  BGBl. I No. 111/2010) in its current version.
• The protection of personal data is covered by the “Data Protection Act 2000” ("Datenschutzgesetz 2000 (DSG 2000)", Austrian Federal Law Gazette I No. 165/1999 as amended by BGBl. I No. 136/2001, BGBl. I No. 13/2005, BGBl. I No. 2/2008, BGBl. I No. 133/2009, BGBl. I No. 135/2009, BGBl. I No. 112/2011, BGBl. I No. 51/2012, BGBl. I No. 57/2013) in its current version.

Aggregations consisting of data of less than 3 establishments are regarded as confidential and therefore not published.

A release calendar is published on the website of STATISTICS AUSTRIA. The release calender is updated at the beginning of a year including the fourth quarter of the former year and the following four quarters of the current year.

https://www.statistik.at/medien/veroeffentlichungskalender

Data are published on the website of Statistics Austria according to the release calendar and no users have prior access to the data. During the same day those data are transmitted to Eurostat.
The “Federal Statistics Act 2000” ("Bundesstatistikgesetz 2000”) states explicitly in §30 (3) the obligation of STATISTICS AUSTRIA to inform without delay the Federal Minister responsible for the subject matter concerned about the results of statistical surveys and to publish them simultaneously.

The preliminary data of the current reference period (t+90) and the final data of the previous period (t+180) are send to Eurostat on the day of the national release via eDAMIS using GESMES files. Data are transmitted on NACE 2-, 3- and 4-digit level and as an aggregate of Division 69 and Group 70.2.

Transmission to Eurostat and national dissemination are carried out both quarterly.

There are no press releases linked to the data.

SPPIs are published annually in the yearbook of Statistics Austria („Statistisches Jahrbuch Österreichs“). This publication is a comprehensive reference work on official statistics and gives concise national information on the demographic, social and economic structure and development of Austria.

Data of SPPIs is available on the website of STATISTICS AUSTRIA in excel-sheets as well as they can be accessed via „StatCube“ – STATISTICS AUSTRIA’s statistical database for free. „StatCube“ provides concerning SPPI data the same quarterly indices as the website of STATISTICS AUSTRIA and is updated according to the release calendar.

Link to SPPI on the website of STATISTICS AUSTRIA:

https://www.statistik.at/en/statistics/national-economy-and-public-finance/prices-and-price-indices/service-producer-price-indices

Link to SPPI on „StatCube“ (Erzeugerpreisindex für unternehmensnahe Dienstleistungen):

https://www.statistik.at/en/databases/statcube-statistical-database      

Micro-data are not disseminated.

On the day of the national publication the respective data are sent to Eurostat according to the data structure for SDMX for EBS STS statistics, either to be used in European aggregates or to be released also as national data.

Methodological papers:
Final reports to Grant Agreements Nr.: 200444401030; 44401.2005.004-2005.306; 44402.2006.004-2006.340; 44402.2007.004-2007.699; 30201.2008.004-2008.283.
"Statistische Nachrichten" (Statistical Bulletin):

• Statistische Nachrichten, Heft 2/2008: Erzeugerpreisindex für unternehmensnahe Dienstleistungen; ein neues Segment der österreichischen Preisstatistik
• "Statistische Nachrichten" (Statistical Bulletin):
• Statistische Nachrichten, Heft 10/2010: Erzeugerpreisindex für unternehmensnahe Dienstleistungen 2007-2010
• Internet:
• Homepage of Statistics Austria: https://www.statistik.at/  

The tables:

• Branch indices (current base/reference year 2015=100)
• Branch indices and rates of change (growth rates on previous period, and same period of the previous year) are calculated and published on a quarterly basis for all the below mentioned service branches (see: Scope of the data; activities covered).

A national quality report of the services producer price index was developed from 2009 to 2011.

A short summary in English could be foud by clicking the + symbol at the topic "Documentation" on the English version of the homepage of Statistics Austria:  https://www.statistik.at/en/statistics/national-economy-and-public-finance/prices-and-price-indices/output-price-index-for-business-services.

The full documentation (German language only) is available on the German version of the homepage of Statistics Austria: https://www.statistik.at/statistiken/volkswirtschaft-und-oeffentliche-finanzen/preise-und-preisindizes/erzeugerpreisindex-dienstleistungen . By clicking the + Symbol at the Topic "Dokumentation" Chapter 3 "Quality" of the report covers national quality management issues.

In the performance of its duties, Statistics Austria must observe a number of principles in the interest of its clients, other users and respondents. The most important of these are:

To maintain objectivity and impartiality in the compilation of statistics;

To apply statistical methods and procedures according to internationally accepted scientific principles and standards and to disclose them;

To continuously control quality improvements;

To maintain a high degree of topicality;

To publish statistics and their underlying concepts, definitions and explanations immediately; main results will be available via the Internet free of charge from 2002 onward;

To endeavour to keep the costs to relevant persons and business enterprises as low as possible and to supply the providers of statistical data with sufficient information;

To keep personal data confidential.

There exists a national quality report of the services producer price index on the homepage of Statistics Austria. Chapter 3 “Quality” of the national quality report covers the following issues:

• Relevance;
• Accuracy (sampling errors, non-sampling-errors, revisions,…);
• Timeliness and punctuality in dissemination;
• Comparability;
• Coherence with other statistics.

A process of validation was installed to check data in questionnaires, control mechanisms are used in index computation and index publication.

There are quality checks on overall quality of the statistical outputs each quarter of index computation. In contrast to some other statistical data services are more specific and differences arise in the price survey among service branches. These are the reasons why in Austria computer based plausibility checks are not installed for the services producer price index.

Users may be Eurostat, institutions, social actors, media, researchers & students, enterprises/businesses, internal users. Services producer price indices guarantee an adequate measurement of service sectors‘ volumes but there are also other areas of application. They can be used as economic indicators, for deflation and economic analysis, indexation or as information.

Eurostat, institutions and internal users use them for deflation and economic analysis. Trade unions need them in wage adjustment negotiations. Enterprises/businesses use them for indexation of contracts and as economic indicators. Researchers and students and media need them for economic analysis.

Austria is producing all services producer price indices according to the relevant EU and Austrian regulations. There is one exception caused by a lack of respondents in Group 50.1 and 50.2 “Sea and coastal water transport - passenger and freight”.

The services producer price index satisfies the principle of accuracy, as it reflects a true picture of the development of services producer prices within a certain economic activity. The enterprises for the survey are chosen with the cut-off method, random sampling is only an additional practice for some service sectors. Therefore sampling errors are not calculated for the services producer price index. There are different processes in use to identify and eliminate non sampling-errors. This applies to the collection phase as well as index computation with subsequent control of index publication. Although the sample size for a single service branch comprises only about 150 – 200 respondents at an average, the coverage in regard to the respective branch turnover exceeds 75% in many cases. Assuming that the price development of a service branch is depending on those covered branch leaders and middle sized enterprises, broadening the sample to small enterprises will not lead to an substantial improvement of reliability.

Sampling errors are not calculated for the services producer price index because the enterprises are chosen predominantly with the cut-off method. The sample size depends on the nature of the specific service sector and ranges from only 1 repondent (NACE 53.10 „Postal activities under universal service obligation“) up to about 250 respondents (NACE 71.11 „Architectural activities“). If possible, it is intended to cover all those enterprises that generate about 75% of the respective service branch turnover. Overall about 1835 respondents provide quarterly information for this price index at the moment.

Coverage Errors: Not relevant. Only enterprises with their major activity (based on business register information) in the respective service branch are chosen for the survey.

Data collection errors: Not relevant. The first plausibility checks start at the time Statistics Austria receives the completed price questionnaire. Collection errors are corrected by inquiries and only accurate price quotations will find their way into the index calculation.

Unit and item non-response (missing data): Non-response predominantly occurs as unit non-response. Response rates are at t+90 days about 99% and at t+180 days over 99%. Due to the fact that the survey is mandatory, missing responses are followed up by two postal reminders and then by a financial fine as a result of a criminal proceeding. Missing prices are imputed by the price development of the respective service aggregate.

Data processing errors: Not relevant. Prices are checked the first time during the price collection. Plausibility checks take place during each step of the index computation in order to eliminate processing errors.

Model errors: Not relevant (rather a model effect than a model error). Due to the fact that currently not all service sectors are covered, one model effect is that no aggregated overall services producer price index can be calculated at the moment.

Prices should be reported till the 15th day of the month following the reference quarter. Provisional data are published 90 days after the end of the reference quarter and will be replaced by final data at the end of the following quarter (t+180).

Concerning the punctuality of releases all of them will be delivered on time according to the release calender as described in point 8. About 18% do not report prices till the 15th day of the month following the reference quarter. About 5% fail to transmit prices till the deadline of the first postal reminder. After the second postal reminder nearly all missing responses are collected and only a few replies are open.

Austria applies the same statistical concepts in its entire national territory. There are no known problems of comparability between Austria and other countries because the concept and computation of services producer price indices are in accordance with international standards (e.g. Eurostat and OECD SPPI Manual, IMF PPI Manual).

Time series for the services producer price indices have been published since the first quarter of 2007. Fully developed quality adjustment methods do not exist for SPPIs which can cause problems in comparability over time.

There is not yet a dissemination of information that support statistical cross-checks and provide assurance of reasonableness. Internal ad hoc cross-checks are made with other surveyed data such as sub-indices on low level aggregates from the CPI.

No inconcistencies of statistical output within the data set.

Burden on respondents amounted to 7170 person-hours. Costs and burden cover the specific STS requirements only.

To reduce respondent’s burden the survey is carried out mainly in an electronic way (via email). Respondents have the possibility to return the questionnaires in any form that is convenient for them (via email, mail, fax, telephone). With collecting data just from a sample of enterprises the burden for the sector as a whole can be reduced. Where possible, data from regulatory authority are used to avoid multiple data collection from the enterprises.

A revision calendar exists on the homepage of Statistics Austria. Users can find the dates of publication for provisional data and final data (see also Concept 8.1). Preliminary data are published 90 days after the end of the reference quarter and will be replaced by final data at the end of the following quarter (t+180). Eurostat gets the respective data on the same day immediately after the national publication. The same revisions policy is applied to national releases and to transmissions to Eurostat.

According to the release calendar revisions are made on a regular basis:

• Provisional data are published 90 days after the end of the reference quarter;
• They are replaced by final data at the end of the next quarter (t+180).

This revision policy is described in footnotes to the relevant tables and in the description of the respective statistics.

Methodological changes are announced at the time of change and published in the monthly bulletin "Statistische Nachrichten". There is at least one quarter advance notice of major changes in methodology by including footnotes in the respective tables.

The type of source is a statistical survey with the frame Business Register (UR – “Unternehmensregister”) on which the source is based. Statistics Austria draws a sample with criteria for stratification on NACE 4- and (national classification) 5-digit headings depending on the various sub indices per branch. Observation units and services within the service branch or the enterprise are selected by using a cut-off sample with full enumeration above the cut-off in order to cover the most representative enterprises and services. It is intended to cover 75% of the turnover of the selected service branch. In the case of a service branch where a turnover cut-off of 75 % could only be reached by selecting thousands of enterprises the cut-off threshold is restricted for example to a certain turnover amount per enterprise (e.g. above 1% of the total turnover of the branch). Additionally a small sample of enterprises is drawn by simple random sampling where appropriate in order to represent not only the price development of market leaders. Sample size: all in all for the complete range of SPPIs about 17500 individual prices are included from a total of about 1835 reporting units that are surveyed. Every 5 years there is an update of the sample.

Data are collected on a quarterly basis.

Prices are collected via online questionnaires. There are no planned changes in the national data collection methodology.

Validation level 0: Format and file structure checks.

Validation level 1, 2 and 3: There are checks on completeness, concistency and changes of each record. In the event of missing prices, substantial price changes and changes of service product descriptions respondents are contacted to explain them. The same procedure takes place when prices are unchanged for a long term.

Validation level 4: Cross-checks with other surveyed data such as sub-indices on low level aggregates from the CPI.

Validation level 5: Cross-checks with data from trade unions.

Treatment of missing prices: using the average price development within a stratum or higher stratum (depending on available information). Estimates for grossing-up to population levels are not available.

All branch indices are fixed based Laspeyres indices with the common base/reference year 2015 (2015=100). Every 5 years the basket of service products and enterprises will be revised. The index measures the change in the prices of a fixed set of business service products provided to enterprises. Weights for all stages of the index aggregation are based on turnover information.
Index compilation:
"Product" SPPIs (aggregation through elementary aggregates or a “family tree” of service products within an industry). This can be explained by the example of NACE Class 81.21 General cleaning of buildings as one of the sub-indices of 81.2.
Stage 1 - for each identified service product category (of which there are at least 7; e.g. office cleaning services, cleaning services of hospitals etc.) several service products (e.g. ongoing cleaning of an office) have been selected and per enterprise a price relative is compiled for the entire time series, setting the average for the base/reference year (2015 = 100).
Stage 2 – an unweighted average (geometric mean) is calculated of these price relatives for all individual service products across all sampled respondents within sub-categories.
Stage 3 – these seven product categories are combined using weights (turnover data) provided by service branch representatives to achieve an average SPPI for general cleaning of buildings.
Stage 4 – the higher level indices “General cleaning of buildings, Chimney cleaning, and Other building and industrial cleaning activities n.e.c” are aggregated using weights (turnover data) from the Structural business survey in order to achieve an average SPPI for NACE Group 81.2.

Depending on the aggregation stage of the index calculation the weights are based on turnover data provided by:

• Branch information (single service level);
• Structural business survey (aggregation of sub-indices on NACE 5 digit level and upwards; NACE 5 digit level → NACE 2 digit level)

There are no planned changes in production methods.

No working day or seasonal adjustment is foreseen. There is only one calculation method used for compiling the indices.

Not applicable.