Statbel (Statistics Belgium)

Economy unit

North Gate Boulevard du Roi Albert II 16, 1000 Bruxelles

 Production index (productie-index/indice de la production/produktionindex). This index is designed to be an indicator of the economic cycle.

Nace Rev.2 classification (Statistical Classification of Economic Activities) is used for the production index.

The Prodcom list is used for the Prodcom data from which indices are calculated.

The production indicators cover economic activities listed in sections B to E of Nace rev.2 :

Size classes covered

• for sectors industrial no Prodcom : enterprises with 10 persons employed or more in industrial no Prodcom sectors(1011, slaughterhouse part of 1051, 1910, 1920, 2446)
• for E36 sector : legal units with  principal activity in Nace 3600
• for industrial Prodcom enterprises :
  • all industrial enterprises with 20 persons employed or with a turnover of EUR 4.200.000 or more in the previous reference year;
  • all enterprises with an industrial secondary activity with 20 persons employed or with a turnover of EUR 4.200.000 or more in the previous reference year.

For the Prodcom survey, the threshold is revised every 5 years.

The objective of the production index is to measure changes in the volume of output at close and monthly intervals. It provides a measure of the volume trend in value added over a given reference period. The production index is a theorical measure that must be approximated by practical measures.

List and definition of variables  in Belgium:

• For Prodcom industrial sectors, the production index is based on:
  • deflated value of sold production, 
  • total production

The choice of the type of basic information depends on appropriateness to the activity and also on availability. Deflation is carried out at the Nace 4 level, using the evolution of industrial producer prices (for production measures based on value of sold production). Unit prices are calculated by Prodcom products for measures of production based on total production

• For industrial no prodcom sectors, the production index is based on worked hours
• For E36 sector, the production index is based on available volume of water
• For D351 sector: the net total electricity production in GWh (thermal, hydro, nuclear, geothermal, wind, other of which mainly solar)

Planned changes in information collected: none

Reporting unit

• For industrial Prodcom sector : local unit 
• For industrial no Prodcom sector : local unit
• For E36 sector : legal unit 

Observation unit:

• For industrial Prodcom sector : local unit (with principal or secondary industrial activity)
• For industrial no Prodcom sector : local unit
• For E36 sector : legal unit

• For industrial Prodcom sectors : 3870 local units
• For industrial no Prodcom sectors : ~80 local units
• For E36 sector : ~58 legal units

Indices are published at national and regional levels.

Industrial production indices are available from 2000 for Eurostat.

The series are published also from this period in Belgium.

Before March 2018, the year of reference 2010 was the base year for indices from 2000.

From March 2018, 2015 is the new year of reference and the base year is the year closest to the year of the calculation month.

The following table shows which weights are used for which years:

Indices based on :

• euros for industrial Prodcom sectors
• hours worked for industrial no Prodcom sectors
• m3 for E36 sector
• Gwh for E351 sector

 Absolute figures are published for Prodcom and for the sector E36.

The calendar month is the reference period.

The legal basis for this indicator is the European Business Statistics (EBS) Regulation:

• Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics, repealing 10 legal acts in the field of business statistics (EBS-Regulation
• Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistic (General Implementing Act)

Data are sent to OECD, UN and IMF

Confidential data are treated by suppression.

 Release dates are fixed in advance for a calendar year

Direct link: https://statbel.fgov.be/en/calendar

To access from the main page www.statbel.fgov.be: choose your language, click on the link 'calendar' on 'release schedule' on the bottom right of the page.

• Simultaneous release to all interested parties : Data are transmitted to Eurostat when they are published in Belgium. For certain reference months, the national bank of Belgium has a prior access to the date for internal use only.
• Identification of internal government access to data before release : The news are sent to the president of our ministry for information and to our general director 1 hour before publication
• Transmission to Eurostat and further use of the statistics :Data are transmitted to Eurostat when they are published in Belgium. The transmission is done by use of EDAMIS using GESMES. The level of detail corresponds to the 2-digit level of NACE Rev. 2, the totals and the MIG's

The data are transmitted monthly to Eurostat and disseminated monthly to national users.

There are news on this indicator available on Dutch, French, English and German versions. In these news, the seasonally adjusted production index for the reference month and for the total of industry is compared to the same month of year before and the calendar adjusted index is compared to the month before.

The latest version of these news is available on https://statbel.fgov.be/en/themes/indicators/production/production-industry

There are no other publications.

The news are a web document and are accompanied by an excel table containing :

• gross indexes
• working day adjusted indexes
• trends
• seasonally adjusted indexes

For Belgium and the 3 regions, for Nace 2 digits, Nace 3 digits , MIG's, sections B, C, D, E and totals

Microdata are not disseminated.

Indices are transmitted to Eurostat to be used in calculating European aggregates and to be released as national data.

There is no available methodology document.

There is no available quality document

A great effort was made to optimise data collection and calculation methods (large amount of respondents and basic data, sophisticated estimation procedures, calculations at 8-digit product level, etc.). For each step of the production chain, the data are controlled. At present no explicit quality control checks are carried out, but in the near future the stability of revisions will be tracked.

Once the figures produced, they are validated by our internal validation service before publication.

Checks carried out along the production chain of the indices and these are finally validated by our validation service before diffusion

Belgian national bank, Eurostat, Federations, Professional associations, Universities, Regional public services, Other users

No measure of user satisfaction but we receive feedback of users by mail, phone,...

All of the required series are available

The sampling error is not relevant.

The non-sampling error is small but this error is more important for less detailed aggregates.

The first estimates are biased.

The principal source of errors is the non-response error.

The sampling error is not relevant since the main data for the calculation of the production index (Prodcom) are based on a cut-off sample.

Are questioned in the prodcom survey :

• All industrial enterprises with 20 persons employed or with a turnover of EUR 4.200.000 or more in the previous reference year;
• All enterprises with an industrial secondary activity with 20 persons employed or with a turnover of EUR 4.200.000 or more in the previous reference year.

The main source of non-sampling error is the response rate

The indexes are revised 3 times and are considered as stable 3 months and 25 days after the reference period.

The survey used for the production index is compulsory but for the first round of calculation, the response rate is about 80-85% for the total industry and estimations are made at the individual level based on VAT turnover, history of the responding unit or based on the sector evolution.

However the variability of certain declarations is such that it is impossible to estimate correctly. This is a problem if the responding unit is in an important sector.

Therefore, to avoid to have big biases between 2 revisions, it is imperative to receive such declarations with important impacts on the indexes in time.

A list of important responding units was created with which we have privileged contacts to get their data for the first round of calculation.

To increase the response rate, postal reminders are used and enterprises are also contacted by phone.

Three months after the reference period, the response rate is then approximately 95%.

The production index is released within 1 month and 10 days after the reference period. Then, there is revised twice : two months and 10 days after the reference period and 3 months and 10 days after the reference period (in principle, the third release is final).

Questionnaires are sent to the reporting unit around the 15th of the reference month asking for a reply by the 15th of the following month.

Generally, there is no problem of punctuality with these indices.

The definitions are equally applied to the 3 regions of Belgium.

No problem of over time comparability

The Prodcom data before 2008 (based on nace rev.1) were converted into prodcom data based on nace rev.2 with a conversion table and indices were recalculated.

No regular confrontation is made, but comparisons are made on an ad-hoc basis, especially when results seem contrary to expectations.

Data of production index are often compared with VAT data.

No problem of internal coherence

Burden on respondents

The Prodcom survey was simplified in 2008.

Cost associated with the collection and production of the industrial production index

~10 persons for ~15000 hours/year

The same revision policy is applied to the data released nationally and transmitted to Eurostat. The general Belgian revision policy is available on our internet site following the link: https://statbel.fgov.be/en/about-statbel/quality/revision-policy or under the 'quality' tab after selecting 'About Statbel' at the top right of the home page (https://statbel.fgov.be).

Routine revisions : For the industrial production indices, we recalculated the indices of the last 3 months. The indices are considered as stable 3 months and 10 days after the reference period. Each month, revisions rates are calculated for the different aggregates. If this rate exceeds a determined threshold (generally 5%), differences in the microdata are investigated.

Major revisions : If important changes are made in the production chain, we inform the users in the news tab on the page concerning the industrial production indices

(https://statbel.fgov.be/en/themes/indicators/production/production-industry)

Non-scheduled revisions may occur due to errors discovered in the input data after results are considered definitive. In this case, explanations are also given in the published news and to Eurostat.

For the industrial production indices, there is no revision calendar.

The same revision policy applied to data released nationally and transmitted to Eurostat to assure coherence.

The regular source of revision is the replacement of estimates by the actual values reported. For the industrial production indices, the routine revisions rates depend on the level of the aggregation. Generally, this rate is weak for the high levels aggregates. The revision rates are not published

For June 2017, the revision rate was of -0.4% for the second release (in September 2017) for the aggregate B_C for the index in reference year=year-1 and -0.1% for the third release (in October 2017). 

Rates of difference between 2 versions of an identical reference month are calculated at Nace 2 digits. If these rates are greater than 5%, then investigations are made to find the reasons of the differences.

All of the versions and data are stored in a database with a transaction date. The last version of the indices is stored with '31DEC9999' as transaction date.

Data used to calculate the indices are also stored with a transaction date. We be able to find the data used for the different calculations.

Benchmarking is not carried out with other statistics.

The change in the reference year in March 2018

Changes have occurred in the production process in March 2018 to best respect Eurostat recommendations (update of weights and chain linking with annual overlap).
The base year (year of the weighting scheme) for the indices previously calculated was 2010. At the base change of 2010, we had recalculated the indices since 2000 with the weights of 2010.

Now, the indices have been recalculated as follows :
for indices before 2013 : the recalculation was done based on the nace4  digits indices stored on our database
for indices from 2013 : recalculation was done from microdata

The weights used are the values added at the cost factors from the structural business survey.

if year <=2009 then weight=VA_2008;
if year=2010 then weight=VA_2009;
if year=2011 then weight=VA_2010;
if year=2012 then weight=VA_2011;
if year=2013 then weight=VA_2012;
if year=2014 then weight=VA_2013;
if year=2015 then weight=VA_2014;
if year=2016 then weight=VA_2015;
if year=2017 then weight=VA_2015;
if year=2018 then weight=VA_2015;
if year=2019 then weight=VA_2015;
if year=2020 then weight=VA_2015;
if year=2021 then weight=VA_2015;
if year=2022 then weight=VA_2015;

In the future, we could update the weights to best fit the reference year (but the way of proceeding still needs to be worked out).

The recalculation of the indices directly from the microdata from 2013 follows the discovery of an error in the pharmaceutical sector.
All series since 2000 were then referenced compared to 2015(2015=100).

We also took advantage of change of the reference year to adjust the series with Jdemetra+ (before, we used Demetra+). The models were also adapted.
All of these changes result in revisions to the series previously published.

• MR = 0,18%
• MAR = 0,5748

Sources : cut-of sample survey for industrial Prodcom and no Prodcom sectors

•  For sectors industrial no Prodcom : enterprises with 10 persons employed or more in sectors industrial no Prodcom (1011, slaughterhouse part of 1051, 1910, 1920, 2446)

~80 enterprises are surveyed

• For industrial prodcom enterprises :
  • All industrial enterprises in Nace Prodcom with 20 persons employed or with a turnover of EUR 4.200.000 million or more in the previous reference year;
  • All enterprises in with an industrial secondary activity with 20 persons employed or with a turnover of EUR 4.200.000 million or more in the previous reference year.

If enterprises meet these conditions, their local units are examinated. Local units in industrial sectors are surveyed.

• exhaustive survey for E36 sector : ~58 legal units are surveyed

The frame on which the source is based is DBRIS (Belgian business register).

Data are collected monthly

Questionnaires used in the survey:

• Prodcom survey :
  • Variables by Prodcom code:
    • Total production for T codes
    • Value and quantities of sold production
    • Value of contract work if the principal is in a foreign country
  • Variables by local unit :
    • Commercial value
    • Value of treatment and industrial services
    • Value of sold products made in a foreign country
    • Value of contract work if the principal is in Belgium
• 60EP survey
  • Variables : turnover and hours worked
• Water survey
  • Variables entering in the composition of water available volume

The data of these questionnaires are mostly collected by a web application and few of these are collected by postal questionnaires. The management of the questionnaires is carried out in our operating system : when a questionnaire is validated by the responding unit, it is granted a temporary status. An agent in our collect center must then validate the questionnaire before having a status ok. The validation is made based on the past values of the responding unit and a comparison with the VAT data.

Checks are carried out at the time of data collection and throughout the production chain indices. These are also validated by another service before distribution.

Collected data are daily transferred from the operating system in a datawarehouse. An error reporting with several information (included months for the different survey, included estimations, ...)  is sent to all users everyday.

Then, a SAS program allows to check the variation of the different variables for a statistical unit and allows to identify problems. If a problem is detected, the collect center is contacted. In function of the response of our collect center, the declaration is maintained, re-estimated or removed for the index calculation. Comparisons with VAT data are also made in this case.

The gross indexes based on year-1 are calculated for a reference month. A SAS program allows then to calculate the number of standard errors between the reference month and the 12 months before by nace 4 digits and by regions. This program allows also to compute contributions of the statistical units to the variations and to identify possibly the problematic values. If a problem is detected, the collect center is contacted. In function of the response and the remaining time before the publication, the declaration is corrected or removed for the calculation.

Final tables and press releases are validated by our validation service : format, content, missing information, layout of press release,...are verified.

If results appear abnormal, the validation service recontacts the responsible for verification.

Estimates for non-response

Non-response is treated using several imputation techniques, depending on specific cases. If the correlation between data and VAT data is good then VAT data are used.

Estimates for grossing-up to population levels: Not relevant.

Type of index Fixed base year Laspeyres index

Method of weighting and chaining :The weights used are the values added at the cost factors from the structural business survey.

if year <=2009 then weight=VA_2008;
if year=2010 then weight=VA_2009;
if year=2011 then weight=VA_2010;
if year=2012 then weight=VA_2011;
if year=2013 then weight=VA_2012;
if year=2014 then weight=VA_2013;
if year=2015 then weight=VA_2014;
if year=2016 then weight=VA_2015;
if year=2017 then weight=VA_2015;
if year=2018 then weight=VA_2015;
if year=2019 then weight=VA_2015;
if year=2020 then weight=VA_2015;
if year=2021 then weight=VA_2015;
if year=2022 then weight=VA_2015;

In the future, we could update the weights to best fit the reference year (but the way of proceeding still needs to be worked out).

Adjustments made for :

•  Changes in price levels (deflation): for the production measures based on sold value, a deflation is conducted using the evolution of the producer price index by Nace 4 level

• Quality and productivity: quality and productivity are not adjusted

Seasonal and calendar adjustments

Since March 2018, all of our series are adjusted using the automatic module of JDemetra+ (version 2.2.0)

The following series are published in Belgium: MIG_CAG, MIG_COG, MIG_DCOG, MIG_ING, MIG_NDCOG, MIG_NRG and B_C_D

These series are sent to Eurostat but not published: MIG_ING_CAG, MIG_NRG_X_E, MIG_COG_X_FOOD

The following table shows some informations about different kinds of adjustment and the seasonal adjustment decomposition for MIG-s and total series.

 More details about seasonal and calendar adjustments are in annex.

Not requested.