The Central Statistical Office (CSO)

Business Statistics Industry (DCU) & Business Statistics RAP

Skehard Road, Cork T12 X00E, Ireland

The Index of Industrial Production and Turnover is an electronic and postal survey that collects information on industrial production and industrial turnover. 

The monthly index of industrial production monitors current trends in the volume of production of industrial local units with three or more persons engaged. The primary purpose of the index is to measure changes in the value added at constant prices in the Mining, Manufacturing and Energy sectors. The production index is regarded as one of the most important measures of economic activity. The main advantages of the production index compared with other indicators are its combination of fast availability and at the same time its detailed breakdown.

The monthly index of turnover measures the change in the level of sales each month of industrial products, whether manufactured in the month in question or in previous periods (excluding non-industrial activity where practicable) in the Mining and Manufacturing sectors. The turnover indices are exclusive of VAT. The turnover index is an important indicator of activity in general, namely in terms of the demand for industrial output. It is the objective of the turnover index to show the development of the market for industrial goods and services, and it is an important short-term indicator.

The principal difference between the production and turnover indices is that the production index traces the quantitative volume of production (i.e. effect of price changes excluded) each month whether sold or not, whereas the turnover index reflects both the volume of sales and the price level.

Statistical Classification of Economic Activities in the European Community, Rev. 2 (2008) (i.e. NACE Rev. 2).

Activity Coverage:

Mining and Qarrying - Naces 05 - 09

Manufacturing - Naces 10 - 33

Electricity, Gas, Steam and Air Conditioning Supply - Nace 35

Water Supply; Sewerage, Waste Management and Remediation Activities - Naces 36 - 39

Size Coverage: The monthly survey which provides data for calculating the indices covers the larger local units which between them account for the bulk of industrial output – in general those which in the base year 2015 had twenty or more persons engaged. For certain sectors the threshold of 20 persons engaged is lowered where this is considered necessary to ensure representativeness or where data for a sector would have to be kept confidential if the 20 persons engaged threshold were applied.

Variable(s): The IPT measures production and turnover for the industrial sector in Ireland for a specific period. The index also provides information on year-on year, quarter-on-quarter and month-on-month changes in industrial production and turnover. The turnover indices are exclusive of VAT.  

Reporting unit/respondents:

Reporting unit - 4 digit NACE level reporting for industrial local units with twenty or more persons engaged (threshold is lowered for some NACE groups).

Respondents – industrial local units with twenty or more persons engaged (threshold is lowered for some NACE groups).

Observation units/collection units/units of presentation: The survey is made up of a census of industrial local units with twenty or more persons engaged in the base year (2015) - between NACE codes 05 and 39 inclusive. For certain sectors the threshold of 20 persons engaged is lowered. The survey is a census of approximately 1400 local units which account for 97 % of production and turnover values of local units with three or more persons engaged (as measured by Census Industrial Production 2015).

The data is published to one place of decimal by NACE at state level.

The survey population is made up of all industrial local units with 20+ persons engaged in NACE 05-39 in the base year (year 2015). For certain sectors the threshold of 20 persons engaged is lowered. The survey is a census of approximately 1400 local units which account for 97 % of production and turnover values of local units with three or more persons engaged (as measured by Census Industrial Production 2015).

While the ultimate source of the IPT survey frame is the CSO’s Central Business Register the survey frame is drawn from Census of Industrial Production (CIP) in the base year (which draws from the Central Business Register).

The Republic of Ireland.

A quarterly survey of the level of industrial production was initiated in respect of the third quarter of 1942 covering Transportable Industries. This was changed to a monthly periodicity from the period July 1975 onwards.

The monthly industrial production and turnover indices, base year 2015 = 100, are available from January 2015 onwards. The monthly industrial production indices, base year 2010 = 100, are available from January 1980 onwards.

The monthly industrial turnover indices, base year 2005 = 100, are available from January 2000 onwards.

2015

The following units are used:

• Index (unitless);
• Percentage change on the previous month (rates);
• Percentage change on the previous quarter (rates);
• Percentage change on the same period of the previous year (rates).

The monthly data covers the whole calendar month.

National: The Monthly Industrial Inquiry is a statutory survey and is conducted under Statutory Instrument (S.I.) No. 261 of 2019 - Statistics (Monthly Industrial Inquiry) Order 2019. All statistics are collected under the Statistics Act, 1993.

European: – This survey is carried out to satisfy EU requirements on Short-Term Statistics Regulation- namely 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) and the Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to the EBS Regulation (General Implementing Act).

The Industrial Production and Turnover Inquiry provides some of the information needed by National Accounts in the compilation of quarterly GDP- Regulation EC number 2223/96.

Section 34 of the 1993 Statistics Act states:

The office may provide, for statistical purposes only, information obtained in any way under this Act or the repealed enactments, in such form that it cannot be directly or indirectly related to an identifiable person or undertaking, to such persons and subject to such charges, conditions and restrictions as the Director General may determine.

The data collected is treated as strictly confidential in accordance with the Statistics Act, 1993. The provision of statistical confidentiality is regulated by Sections 32 and 33 of the 1993 Statistics Act. These state:

Section 32: All information furnished by a person, undertaking or public authority under this Act shall be used only for statistical compilation and analysis purposes.

Section 33:

1. No information obtained in any way under this Act or the repealed enactments which can be related to an identifiable person or undertaking shall, except with the written consent of that person or undertaking or the personal representative or next-of-kin of a deceased person, be disseminated, shown or communicated to any person or body except as follows:

- for the purpose of a prosecution for an offence under this Act;

- to officers of statistics in the course of their duties under this Act;

- for the purposes of recording such information solely for the use of the office in such form and manner as is provided for by a contract in writing made by the Director General which protects its confidentiality to his satisfaction

2. The office may, for statistical purpose only, assign codes derived from information collected under this Act classifying undertakings listed in the administrative systems of other public authorities by economic activity and size (persons engaged) categories.

3. The Taoiseach (the head of government of Ireland) may by order prescribe such further prohibitions on the disclosure of identifiable records or information obtained under this Act or the repealed enactments for such periods as may be prescribed.

4. Nothing in the act shall be construed to require any person or undertaking to provide information in relation to a matter on which information was sought in circumstances that would entitle the person or undertaking to decline to give the information in a civil proceeding in any court or on grounds of privilege.

All data are treated as strictly confidential in accordance with Part V of the Statistics Act, 1993. In order to ensure confidentiality, NACE Groups are amalgamated where companies are identifiable due to their size. Data cannot be accessed under the terms of the Freedom of Information Act, 1997. Data are not disclosed by the CSO to any other Government department or outside body.

A four month advance calendar showing release dates on a 'no later than' basis is publicly available and published at the end of each month on the CSO website. In addition, a weekly release calendar which contains a schedule of precise release dates for those statistics that are to be published by the CSO during the following week is issued each Thursday by e-mail to the media and all other interested parties and is also publically available on the CSO website.

Both the four month advance calendar and the weekly release calendar can be accessed on the CSO website: https://www.cso.ie/en/csolatestnews/releasecalendar/

The CSO disseminates the IPT publication on the CSO website at 11am (local time) on the day of publication (see point 8.1).

No early data is provided to any users. No prior release is provided to any users. Data required under STS regulations are transmitted to Eurostat the same day following national release.

Monthly.

No CSO releases are accompanied by commentary from any ministerial office.

Aggregated indices are published in the IPT monthly publication on the CSO website. Each aggregated index is published as a seasonally adjusted value index in the publication.

The data that is published in the IPT monthly publication is also made available on Px Stat (CSO Main Data Dissemination Service).

Microdata are not publicly available.

Data are sent to Eurostat to be used in European aggregates and/or to be released as national data. There are no planned future changes in national dissemination methods.

Information on methodology is available within the https://www.cso.ie/en/methods/surveybackgroundnotes/industrialproductionturnover/ (see Background Notes section of publication).

Description of standard tables:

Table 1: Volume Indices of Production (seasonally adjusted)

Table 1a: Volume Indices of Production (unadjusted)

Table 2: Volume Indices of Production for “Modern” and “Traditional” Sectors.

Table 3a - 3c: Volume Indices of Production for Industrial Sectors.

Table 4: Industrial Turnover Indices (seasonally adjusted).

Table 4a: Industrial Turnover Indices (unadjusted).

Table 5a-5c: Turnover Indices for Industrial Sectors.

Indicators are published mainly at the 2-digit level of NACE Rev. 2 with, in certain cases, 3 digit breakdowns for major sectors. Broader aggregates are also published including capital, intermediate and consumer goods indices. No regional breakdown exists.

Not available.

Response rates are compiled and analysed each month.

Data for the IPT are benchmarked against the latest Census of Industrial Production (CIP).

The ultimate source of the IPT survey frame is the CSO’s Central Business Register. The survey frame is drawn from Census of Industrial Production (CIP) in the base year (which draws from the Central Business Register). Duplicates, births and deaths are checked for and removed/added each month as necessary. Coverage is restricted to industrial local units operating in NACE sectors 05- 39. Quality checks and validation of data are done throughout the whole compilation process. Details are available in the Standard Report on Methods and Quality for Monthly Industrial Inquiry Index. A summary of main points is provided below:

Sampling Effect, Representativity

Standard error is not relevant as it is a census. A census is taken of all NACE groups in 05-39 with twenty or more persons engaged to ensure that the statistics produced are properly representative of the trends of these groups (the threshold is lowered for some NACE groups).

Non‐Sampling Effects

Quality of the Data Sources used (other than survey register)

Sources of data external to the survey- data from Trade Section (ultimate source of this data is Revenue Commissioners) and Balance Of Payments (BOP) are used to provide estimates. Ongoing quality issues are addressed by the relevant survey managers in the CSO.

Register Coverage

Census is taken of local units that had twenty or more persons engaged in base year (2015). New births are added to the survey frame (birthed after 1 January 2016). Local units are ceased as appropriate. No other data sources are used.

As this survey is a census of local units with twenty or more persons engaged in the base year, all local units in the relevant NACE groups are surveyed.

Coverage errors – there may be some multiple listings (duplicates) when the census is first taken. These are removed from the census when discovered. Coverage is restricted to industrial local units operating in NACE sectors 05- 39.

Non‐response (Unit and Item)

High quality data is ensured due to high value response rate 55% (96% of production value and 96% of turnover value for final results). For non-responding large local units, data is estimated using a variety of methods such as seasonal factors, three-month average, previous month’s data and other CSO data.

Unit Non‐Response: The response rates for the final results are typically 55% in terms of the local units surveyed covering 96% of the value of production and 96% of the value of turnover. These response rates are based on averages for a 12 month period.

The main reasons for unit non-response are as follows:

1. Large burden on respondents- companies in industrial sector also need to complete other CSO surveys.
2. Lack of resources in smaller companies.
3. Frequency of survey may reduce response.
4. Due to the short turnaround the data may not be available.

Item Non‐Response is not measured in the survey. There is no imputation for item non-response.

Where a part of the form is not completed one of the following is done:

1. The company is contacted by telephone to obtain the data.
2. Data is estimated for a variable if no other data are available.

Measurement Errors

Measurement Error is not formally calculated for the survey. The following should be noted:

1. Questionnaire – the questionnaire is clear with definitions provided of ‘persons engaged’,’ and ‘production’. The design of the questionnaire is monitored and changes made if necessary.
2. Data collection method – the collection method is e-form and by post. This method is unlikely of itself to be a source of measurement error.
3. Interviewer- this technique not used by the survey.
4. Respondent – survey data can contain sensitive and confidential information due to the fact that it is a short-term survey and may be requested prior to annual accounts being completed. The survey publishes aggregated indices at 2 digit NACE (may aggregate more that one-two digit NACE group – due to confidentiality).

 Processing Errors

1. Data capture errors are likely to be low risk as postal survey forms are scanned and verified.
2. Forms are coded in advance of issue and no manual coding is done.
3. The form is scrutinised before scanning and verification.
4. Data editing- edits are run which test the returns against a range of values;

• Currency entered changed since previous period.
• Turnover is equal to zero.
• Turnover is unusually high.
• Large change in turnover since previous month/year (range varies between +/- 20% and +/- 40% depending on size of company).
• Large change in production since previous month/year (range varies between +/- 20% and +/- 40% depending on size of company).
• Production is unusually high.
• Total persons engaged are equal to zero.
• Large change in number of persons engaged since previous month.

 Contact is made with company to confirm returned figures. Data are manually changed if required.

Coherence

Consistency checks are done on a regular basis with Balance of Payments (quarterly basis) and Trade exports data (monthly basis) for the larger local units. Differences are verified and this leads to improved data quality.

In certain instances comparison with other data sources such as Census of Industrial Production (CIP) are conducted. These are investigated where necessary and this has led to better quality data.

Not available.

Industrial Production and Turnover Index is an important short term indicator used both nationally and by the EU. It provides some of the information needed by National Accounts in the compilation of quarterly GDP- Regulation EC number 2223/96.

Users include Eurostat; National Accounts, CSO; Government Departments and Agencies; Economists; Brokers; Researchers and Academics; Professional Bodies; General Public.

Not available.

The index is fully compliant with requirements of the 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) and the Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to the EBS Regulation (General Implementing Act).

Each month the provisional data from the previous month is revised on the receipt of late returns or amendments of existing returns.

The current month’s data is always provisional and only becomes final when the next month is published. The size of the revision varies from month to month but would typically be in the order of +/- 2% for ‘All Industries’ indices.

As the production and turnover indices are seasonally adjusted, the series are revised annually due to updated seasonal factors (updated once a year). 

Other revisions are flagged as such in the Release. Updating methodologies are not usually announced in advance but simultaneously with implementation. A significant change (e.g. rebasing of the indices) will be notified in advance.

Standard error is not relevant as it is a census. A census is taken of all NACE groups in 05-39 with twenty or more persons engaged to ensure that the statistics produced are properly representative of the trends of these groups (the threshold is lowered for some NACE groups).

Quality of the Data Sources used (other than survey register)

Sources of data external to the survey- data from Trade Section (ultimate source of this data is Revenue Commissioners) and Balance Of Payments (BOP) are used to provide estimates. Ongoing quality issues are addressed by the relevant survey managers in the CSO.

Register Coverage

Census is taken of local units that had twenty or more persons engaged in base year (2015). New births are added to the survey frame (birthed after 1 January 2016). Local units are ceased as appropriate. No other data sources are used.

As this survey is a census of local units with twenty or more persons engaged in the base year, all local units in the relevant NACE groups are surveyed.

Coverage errors – there may be some multiple listings (duplicates) when the census is first taken. These are removed from the census when discovered. Coverage is restricted to industrial local units operating in NACE sectors 05- 39.

Non‐response (Unit and Item)

High quality data is ensured due to high value response rate 65% (96% of production value and 96% of turnover value for final results). For non-responding large local units, data is estimated using a variety of methods such as seasonal factors, three-month average, previous month’s data, and other CSO data.

Unit Non‐Response: The response rates for the final results are typically 65% in terms of the local units surveyed covering 96% of the value of production and 96% of the value of turnover. These response rates are based on averages for a 12 month period.

The main reasons for unit non-response are as follows:

1. Large burden on respondents- companies in industrial sector also need to complete other CSO surveys.
2. Lack of resources in smaller companies.
3. Frequency of survey may reduce response.
4. Due to the short turn around the data may not be available.

Item Non‐Response Item non‐response is not measured in the survey. There is no imputation for item non-response.

Where a part of the form is not completed one of the following is done:

1. The company is contacted by telephone to obtain the data
2. Data is estimated for a variable if no other data are available.

Measurement Errors

Measurement Error is not formally calculated for the survey. The following should be noted:

1. Questionnaire – the questionnaire is clear with definitions provided of ‘persons engaged’ and ‘production’. The design of the questionnaire is monitored and changes made if necessary.
2. Data collection method – the collection method is e-form and by post. This method is unlikely of itself to be a source of measurement error.
3. Interviewer- this technique not used by the survey.
4. Respondent – survey data can contain sensitive and confidential information due to the fact that it is a short-term survey and may be requested prior to annual accounts being completed. The survey publishes aggregated indices at 2 digit NACE (may aggregate more that one-two digit NACE group – due to confidentiality).

Processing Errors

1. Data capture errors are likely to be low risk as postal survey forms are scanned and verified.
2. Forms are coded in advance of issue and no manual coding is done.
3. The form is scrutinised before scanning and verification.
4. Data editing- edits are run which test the returns against a range of values;

• Currency entered changed since previous period.
• Turnover is equal to zero.
• Turnover is unusually high.
• Large change in turnover since previous month/year (range varies between +/- 20% and +/- 40% depending on size of company).
• Large change in production since previous month/year (range varies between +/- 20% and +/- 40% depending on size of company).
• Production is unusually high.
• Total persons engaged are equal to zero.
• Large change in number of persons engaged since previous month.

 Contact is made with company to confirm returned figures. Data are manually changed if required.

The target date for detailed provisional data is 6 weeks after the end of the reference month. Final detailed results are published around 10 weeks after the reference period. Questionnaires are sent out immediately after the end of the reference month.

Not available.

Not available.

Not available.

SBS, PRODCOM, Balance of payments and External trade data are used for cross-checks to ensure data is consistent for enterprises and same variables. Where data is queried/questioned, we contact the respondent enterprises for clarification/confirmation/correction of the data. Such interactions with respondents is recorded for validation purposes.

Not available.

Not available.

The current month’s data is always provisional and only becomes final when the next month is published. The seasonally adjusted production and turnover indices are revised each month as the new data is included. Any major revision, including rebasing would be flagged in advance. Any revision due to errors or coherence are flagged as such in the release. Any national revisions are transmitted to Eurostat at the earliest opportunity. This usually is on the day of publication or the next transmission date.

Each month the provisional data from the previous month is revised on the receipt of late returns or amendments of existing returns.

The size of the revision varies from month to month but would typically be in the order of +/- 2% for ‘All Industries’ indices.

The seasonal models used in the seasonal adjustment are typically updated once a year if required.

Other revisions are flagged as such in the Release. Updating methodologies are not usually announced in advance but simultaneously with implementation. A significant change (e.g. rebasing of the indices) will be notified in advance.

To improve coherence across surveys once a year a major revision of the previous year’s data is undertaken. This is usually carried out in June which corresponds with annual data being collected from enterprises.

The MAR/MR values below are calculated for total of NACE (B_C): 

Mean Revision (MR) = 1.13

Mean Absolute Revision (MAR) = 3.42

Standard Deviations of Revision (STDR) = 4.67

Range of Revisions (RR) = 13.5

Type of source: Statistical survey based on a census.

Frame on which the source is based: Central Business Register. There are approximately 1,400 local units in the reference population.

Frequency of updating the sample: Sample is taken every five years (base year). New enterprises are added to the sample each month.

Data are collected monthly.

Industrial Production Turnover - questionnaires are customised to suit each NACE Rev. 2 Class. Data are collected by means of postal questionnaires.

Not available.

The following methods are used for non-response:
— Imputation; the approach adopted is to use the non-respondent’s last value carried forward and to adjust that by monthly trends of respondents in that stratum;
— Non-responding large companies are estimated each month (rather than imputed). These estimates can be based on data for Trade exports, Balance of payments, SBS (prior year), previous month, average of previous three months, sectoral trends, and other methods.

Type of index: The production index is a Laspeyres index.

Method of weighting and chaining: The basic indices are weighted by the gross value added at factor cost for local units with 3 or more persons employed.

CALENDAR ADJUSTMENT

A national regressor is used to include our national bank holidays.

SEASONAL ADJUSTMENT

Win X 13 is the software used, it is an adaptation of the US Bureau of the Census X-13 Seasonal Adjustment Program.

For most of the series automatic models are used but some manual models are also included depending on the adequacy of the automatic model. If a manual model is used the same model is used for both the production and turnover index for a series.

The models and the respective parameters are looked at on an annual basis and revised if appropriate.

The series are revised from January 2015 onwards (beginning of base year).

Each series is independently seasonal adjustment some are additive and some are multiplicative as appropriate.

The automatic outliers detected are included. Outliers that are known to us from the data are also included eg large shutdown in a major enterprise in a particular sector.

The filter length is automatically chosen.

There are no seasonal breaks in the series, from new data collection method or methodology.

Residual seasonality is checked at index level once a year when the models are looked at.

The different levels of breakdown are treated individually. 

Not available.