Annexes to COM(2017)411 - Short term statistics as required by Article 14(2) of Council Regulation (EC) No 1165/1998 - Main contents
Please note
This page contains a limited version of this dossier in the EU Monitor.
| dossier | COM(2017)411 - Short term statistics as required by Article 14(2) of Council Regulation (EC) No 1165/1998. |
|---|---|
| document | COM(2017)411  |
| date | August 2, 2017 |
In March 2016, a new seasonal adjustment software was introduced for European short-term business statistics. Starting with the reference year 2016 the new software JDemetra+ is used. With this change short-term statistics follow recommendations by Eurostat and the ECB for seasonal and calendar adjustment of official statistics in the EU. 5 The new software improves the quality and transparency of the calendar and seasonal adjustment in short-term statistics. The switch to the new seasonal adjustment software has not caused changes in the headline time series, their interpretation or use. The previous seasonal adjustment specifications are also used in JDemetra+. This ensures stability of the seasonal adjustment process and minimises revisions.
3.Scope and compliance with the STS Regulation
Eurostat monitors Member States’ compliance with the STS Regulation in terms of timeliness and completeness. On the basis of the various quality aspects outlined in the European Statistics Code of Practice 6 , twice per year a comprehensive compliance score covering all indicators is calculated for each country. Generally, the results of the monitoring show a high level of compliance with the regulation. As of 1 April 2016, the EU-28 average score was 9.5 (out of 10) with a large majority of Member States fully compliant.
3.1.Accuracy, reliability, coherence and comparability
With Commission Regulation (EC) No 1503/2006 7 common definitions for short-term statistics were introduced in order to ensure the coherence and comparability of data. Eurostat and the national statistical offices work closely together to maintain and improve high accuracy, reliability and coherence of the short-term statistics indicators. The methodological framework established by the STS Regulation is continuously improved by consultations of technical experts and of special thematic task forces.
Despite uniform definitions, statistical data compilation approaches do not have to be identical across Member States. In keeping with the principle of subsidiarity and in order to take account of national differences, e.g. as regards size, economic structure and availability of administrative data, the STS Regulation leaves it to Member States to decide on the most efficient and effective ways of collecting and processing the data.
Eurostat also works together with other international organisations, especially the OECD, to increase the comparability of data and methodology beyond the European Union.
3.2.Timeliness and punctuality
Short-term statistics are among the first official data to provide an indication of the recent economic development. The STS Regulation sets very short deadlines for the delivery of national data to Eurostat 8 . Table 2 shows the current deadlines for data transmission to Eurostat (third column) and the target dates set by the Economic and Financial Committee (EFC) in its most recent report on information requirements in the European Monetary Union. The last column indicates the current deadlines for dissemination (e.g. via special news releases and general online releases).
In general, the timeliness of short-term statistics can be considered very good. Delays usually only occur if the target delivery day falls on a weekend or on a public holiday.
Users are informed about news release publication dates well in advance by means of the news release calendar on the Eurostat website 9 . In past years, all dates announced in the calendar have been met.
Table 2: Timeliness – days between end of reference period and delivery deadline, STS Regulation, EFC target and actual dissemination of EA aggregates (November 2016)
| Indicator | Periodicity | STS Regulation deadlines a) | EFC targets for 2016 b) | Dissemination of EA totals c) |
| Industrial production | Monthly | 40 | 40 | 42 |
| Industrial output prices of the domestic market | Monthly | 35 | 35 | 33 |
| Industrial import prices | Monthly | 45 | 45 | 40 |
| Production in construction | Monthly | 45 | 45 | 49 |
| Building permits | Quarterly | 90 | 90 | 91 |
| Turnover in retail trade | Monthly | 30 | 30 | 34 |
| Turnover in (other) services | Quarterly | 60 | 60 | 63 |
| Service output prices | Quarterly | 90 | 90 | 91 |
a) Deadlines for the transmission of data to Eurostat under the shortterm statistics Regulation; for smaller countries, longer deadlines may apply.
b) Targets set in the Economic and Financial Committee (EFC) 2016 Status Report on information requirements on EMU.
c) Days between the end of reference period and dissemination (e.g. news release, data release).
3.3.Revisions of short-term statistics indicators
The results of short-term indicators have to be based, to some extent, on preliminary, estimated and incomplete data. Following the first data release, survey results are revised when late respondents’ input has been added. But data can also be revised for a whole range of other reasons, including seasonal adjustment, benchmarking, new and/or improved data sources, and corrections of errors or methodological changes. The revisions are generally rather limited in scale, especially at the aggregated EU level and for the euro area.
In order to assess the quality of the first results of the four short-term statistics indicators published monthly in a news release, changes between first and second publications of monthly growth rates were analysed for 2014-16.
Table 3 shows, for the euro-area aggregates of the four STS indicators subject to a news release, the average scale of revisions of growth rates between the first and the second publication one month later (second column). This is zero or close to zero, which means that upwards and downwards revisions of the growth rates cancel each other out, i.e. there appears to be no systematic bias to over- or under-estimating results.
Table 3: Scale of revision for the four principal short-term statistics indicators, 2014-16 a)
| 1 | 2 | 3 | 4 | 5 |
| Average revision b) | Average absolute revision c) | Average absolute growth rate | Relative average revision d) | |
| Industrial production | 0.1 | 0.2 | 0.7 | 0.2 |
| Industrial output prices (domestic market) | 0.0 | 0.0 | 0.4 | 0.1 |
| Production in construction | 0.0 | 0.4 | 0.7 | 0.5 |
| Volume of retail trade | 0.0 | 0.2 | 0.4 | 0.5 |
a) Seasonally adjusted growth rates of indicators for the euro area.
b) Differences between growth rates at second and first publication, 2014-16 average.
c) Absolute differences between growth rates at second and first publication, 2014-16 average.
d) Ratio between absolute revision (3rd column) and absolute growth rate at second publication (4th column), 2014-16 average. Differences are due to rounding errors.
In absolute terms (third column), the average revisions vary between almost zero for industrial output prices and 0.4 percentage points for production in construction. The differences across indicators are explained mainly by different compilation and revision methods. While industrial output prices are often not revised at all, production indicators might even be revised several years after their first publication.
Average growth rates (in absolute terms at second publication) are indicated in column 4. The last column shows the relationship between the revision of the growth rate and the growth rate itself (at the time of the second publication). This number describes more accurately the magnitude of the revisions, since it makes a difference whether a revision of x percentage points is applied to a relatively high or a relatively low growth rate.
3.4.Accessibility, clarity and availability of metadata
In the area of short-term statistics, Eurostat publishes 48 news releases a year, i.e. monthly releases for four principal indicators (industrial production, industrial output prices, production in construction and volume of retail trade). In past years, all news releases have been published according to schedule.
All European short-term statistics data are accessible via the Eurostat website; they are free of charge. The section dedicated to short-term statistics 10 provides access to the complete short-term statistics database but also to a number of pre-defined statistical tables. In addition the dedicated section provides information on the legal basis of short-term statistics and contains numerous methodological publications. Metadata for the EU aggregates as well as national metadata on the various short-term statistics indicators can also be accessed via the dedicated section.
Concise articles on all short-term statistics indicators as well as numerous methodological background articles are published in Statistics Explained, a wiki-style online platform. 11 In the past a large number of articles on short-term statistics had been published in the format of Statistics in focus.
4.Current costs to the statistical system and burden on businesses
The measurement of the costs (for statistical systems, related to collecting, processing and disseminating data) and the burden (on the enterprises providing the data) arising from short-term statistics has proven to be extremely difficult. In line with the subsidiarity principle statistical institutes apply different methods for the data collection and data production. For example data sources vary: some countries use survey data, others rely on secondary administrative sources and others even combine surveys with administrative data. Survey methods and data collection tools also differ. Consequently, any comparison between Member States of cost and burden estimations requires greatest caution. Moreover, when assessing cost and burden related to short-term statistics the benefit side also needs to be considered, as these data provide an important input to other statistics (especially national accounts).
The burden data collected from Member States in the past indicated that an average business has to spend around 20 minutes per month to comply with statistical requests for the collection of production data (industry and construction), while turnover data are easier to provide (5 to 10 minutes per month), and the provision of price data requires around 15 minutes per month. These figures are however only rough indications since differences between Member States are quite large.
Available evidence suggests that the statistical burden has slightly decreased during past years. More and more statistical offices facilitate the provision of data by enterprises, e.g. by an electronic collection instead of paper surveys. Statistical offices also attempt to reduce sample sizes and rely, whenever possible, on administrative data.
5.Future developments in short-term statistics
Since 2000 the share of industry and construction in the value added of all economic activity has dropped from 34.0 % to 30.5 %. Over the same period, the share of market services (excluding banking and insurance) has increased from 41.0 % to 45.6 % 12 . In order to adequately represent this increased importance of the service sector in short-term statistics, Eurostat, in co-operation with Member States and after thorough consultation of key users, has prepared a proposal for a comprehensive enlargement and improvement of the legislative basis for short-term statistics, aiming at better covering the services sectors.
This so-called "STS-package" is part of the proposal for a Framework Regulation Integrating Business Statistics (FRIBS), which the European Commission adopted on 6 March 2017 13 together with an extensive impact assessment. 14
In particular, according to the FRIBS proposal a production (volume) indicator for the service industries (including financial and public services) will be included in the set of short-term statistics indicators. Moreover, the scope of service industries covered by the different short-term statistics indicators will be consolidated. The new indicator of services production (ISP) will have a monthly reference period and should be available 60 days after the reference period. In addition to the improved coverage of the service industries short-term statistics will be streamlined and consolidated in several other respects, e.g. concerning transmission deadlines, the use of statistical units, and the definition of size classes of countries.
The proposed changes will guarantee that short-term statistics continue to fulfil their task of correctly and adequately measuring business cycle developments in a rapidly changing economic environment. The additional service data will also enable Eurostat to publish a new total production indicator.
(1) OJ L 162, 5.6.1998, p. 1.
(2) COM(2008) 340 final, 9.6.2008; COM(2011) 329 final, 8.6.2011; and COM(2014) 381 final, 26.6.2014. There was also a report in 2003: COM(2003) 36 final, 29.1.2003.
(3) The original list of PEEIs was established in 2002 (Communication of the Commission to the European Parliament and the Council on Eurozone Statistics, 27.11.2002, COM(2002) 661 final).
(4) http://ec.europa.eu/eurostat/cache/metadata/en/sts_esms.htm#annex .
(5) Moreover, the ESS guidelines on seasonal adjustment recommend the use of JDemetra+ by Eurostat and by the Members of the European Statistical System in a broader context: http://ec.europa.eu/eurostat/en/web/products-manuals-and-guidelines/-/KS-GQ-15-001
(6) The European Statistics Code of Practice is available at: http://ec.europa.eu/eurostat/web/quality/european-statistics-code-of-practice .
(7) OJ L 281 of 12.10.2006, p. 15.
(8) The original deadlines in Regulation (EC) No 1165/98 were further shortened by Regulation (EC) No 1158/2005 of the European Parliament and of the Council of 6 July 2005 (OJ L 191, 22.7.2005, p. 1).
(9) The calendar is available at: http://ec.europa.eu/eurostat/web/short-term-business-statistics/overview .
(10) http://ec.europa.eu/eurostat/web/short-term-business-statistics/overview
(11) http://ec.europa.eu/eurostat/statistics-explained/index.php/Short-term_business_statistics .
(12) Gross added value chain linked volumes, Eurostat, National Accounts, own calculations.
(13) https://ec.europa.eu/transparency/regdoc/rep/1/2017/EN/COM-2017-114-F1-EN-MAIN-PART-1.PDF .
(14) https://ec.europa.eu/transparency/regdoc/rep/10102/2017/EN/SWD-2017-98-F1-EN-MAIN-PART-1.PDF .