United Nations Statistics Division

ISIC and CPC implementationISIC and CPC implementation

2

Expert GroupExpert Group

• Expert Group discussed a proposal for the Expert Group discussed a proposal for the implementation of ISIC and CPC at its implementation of ISIC and CPC at its meeting in April 2007meeting in April 2007

• Main recommendations relate to:Main recommendations relate to:• TimetableTimetable• Documentation and toolsDocumentation and tools• Regional workshopsRegional workshops• Technical cooperationTechnical cooperation

3

TimetableTimetable

• EG recommended two milestones:EG recommended two milestones:• Countries should adapt their national classification to Countries should adapt their national classification to

ISIC Rev.4 by 2009ISIC Rev.4 by 2009• To meet population census target of 2010To meet population census target of 2010

• International data reporting in ISIC Rev.4 format International data reporting in ISIC Rev.4 format should begin in 2012should begin in 2012

• Some countries recommended earlier goals, but Some countries recommended earlier goals, but that may be unrealistic for majority of countriesthat may be unrealistic for majority of countries

4

TimetableTimetable

• No specific dates have been recommended for No specific dates have been recommended for completion of:completion of:• Adaptation of business registerAdaptation of business register

• Sample designs and weights for surveysSample designs and weights for surveys

• BackcastingBackcasting

• Reasons:Reasons:• The timing and organization of surveys varies by The timing and organization of surveys varies by

countrycountry

• Scope (length) of backcasting differs by countryScope (length) of backcasting differs by country

5

Documentation and toolsDocumentation and tools

• UNSD will develop:UNSD will develop:• A Companion Guide to ISIC and CPC that A Companion Guide to ISIC and CPC that

helps to understand:helps to understand:• the concepts and structure of the classificationsthe concepts and structure of the classifications• Application of the classificationApplication of the classification

• Correspondence tablesCorrespondence tables• ISIC and CPC indexesISIC and CPC indexes

• UNSD will also set up an open discussion UNSD will also set up an open discussion forum and expand the classifications website forum and expand the classifications website into a larger knowledge platforminto a larger knowledge platform

6

Documentation and toolsDocumentation and tools

• UNSD will develop documents, based on UNSD will develop documents, based on European and other experiences, for:European and other experiences, for:• Establishing a national classificationEstablishing a national classification• Recoding of business registersRecoding of business registers• Sampling design and weight estimationSampling design and weight estimation• Backcasting Backcasting

7

Regional workshopsRegional workshops• Regional workshops have been supported in two Regional workshops have been supported in two

phases:phases:• Workshops on final versions of ISIC and CPC in 2007/8Workshops on final versions of ISIC and CPC in 2007/8• Workshops on specific elements of implementation*, Workshops on specific elements of implementation*,

such as:such as:• Adaptation of business registersAdaptation of business registers• Statistical collection programmesStatistical collection programmes• Use of the classification for administrative dataUse of the classification for administrative data

* (carried out most likely in connection with industry * (carried out most likely in connection with industry workshops)workshops)

8

Technical cooperationTechnical cooperation

• EG supported regional partnership for technical EG supported regional partnership for technical cooperationcooperation

• Proposal:Proposal:• US, Canada, selected S.American countries support US, Canada, selected S.American countries support

ECLACECLAC• EU members support ECA, ECE, ESCWA (where EU members support ECA, ECE, ESCWA (where

relationships already exist)relationships already exist)• Australia, Japan, India support ESCAPAustralia, Japan, India support ESCAP• UNSD will serve specific subregionsUNSD will serve specific subregions

• First steps of cooperation will be organized First steps of cooperation will be organized through the open discussion forumthrough the open discussion forum

9

Implementation processImplementation process

10

Implementation of ISIC Rev.4Implementation of ISIC Rev.4• Major tasks and challenges:Major tasks and challenges:

• Defining a new classification and associated toolsDefining a new classification and associated tools• Reclassification of all units on the business register Reclassification of all units on the business register

according to the revised classificationaccording to the revised classification• Maintaining two classifications for an interim periodMaintaining two classifications for an interim period• Sampling and weighting under the new classificationSampling and weighting under the new classification• Simultaneous estimation and results assessment under Simultaneous estimation and results assessment under

both new and old classificationboth new and old classification• Construction of industry weights for short term Construction of industry weights for short term

statisticsstatistics• Construction of back series in terms of the revised Construction of back series in terms of the revised

classificationclassification• Handling of the national accounts move to the revised Handling of the national accounts move to the revised

classificationclassification

11

Overall implementation processOverall implementation process

• NSIC setup processNSIC setup process• creation of a new national version of national creation of a new national version of national

classification (NSIC) according to ISIC Rev.4 classification (NSIC) according to ISIC Rev.4 for national usefor national use

• Register update processRegister update process• switchover in the Business Register to new switchover in the Business Register to new

NSICNSIC

• Statistics update processStatistics update process• co-ordinate simultaneous switchover in co-ordinate simultaneous switchover in

statistics to new NSIC and the control of the statistics to new NSIC and the control of the quality of time seriesquality of time series

12

NSIC setup processNSIC setup process

13

Register update processRegister update process

14

Statistics update processStatistics update process

15

Coordination of workCoordination of work

• Groups involved in the three processes need Groups involved in the three processes need to be coordinatedto be coordinated• Need for a programme managerNeed for a programme manager

• Communication with users of the Communication with users of the classification (potential data providers)classification (potential data providers)• Need to get other users involved in setup and Need to get other users involved in setup and

keep them updated on work and impact keep them updated on work and impact according to a fixed scheduleaccording to a fixed schedule

16

Criteria for developing NSICCriteria for developing NSIC

• Discussed in previous presentationDiscussed in previous presentation

• Should involve other stakeholders / users/ Should involve other stakeholders / users/ data providersdata providers

• If sufficient, ISIC Rev.4 can be used as isIf sufficient, ISIC Rev.4 can be used as is• In this case, tools (correspondence tables etc.) In this case, tools (correspondence tables etc.)

are readyare ready

17

Recoding the business registerRecoding the business register

• Based on known correspondence table, can Based on known correspondence table, can use different methods:use different methods:• Direct (straight) recodingDirect (straight) recoding• Use of additional info from NSO or external Use of additional info from NSO or external

registersregisters• SurveysSurveys• ProfilingProfiling• Probabilistic models Probabilistic models

18

Recoding the business registerRecoding the business register

• Dual coding should be kept for several Dual coding should be kept for several yearsyears• Problem: some units may cease operationProblem: some units may cease operation

• Use of reference dates may helpUse of reference dates may help

• Can be implemented by using a transition codeCan be implemented by using a transition code• Reflect correspondence tableReflect correspondence table

• Can computer-assisted coding help?Can computer-assisted coding help?

19

Future statistical workFuture statistical work• Sampling procedures have to be updated for Sampling procedures have to be updated for

the new classificationthe new classification• Scope of categories may have changedScope of categories may have changed• Should (at least for the transition period) Should (at least for the transition period)

produce estimates for old and new produce estimates for old and new classificationclassification

• Overall sample size may increaseOverall sample size may increase• To increase reliability of dual-coded resultsTo increase reliability of dual-coded results• To compensate for startup problems of the new To compensate for startup problems of the new

classification (misclassification of units)classification (misclassification of units)• To allow for increase of detail required in new To allow for increase of detail required in new

classificationclassification• To better evaluate coherence between old and new To better evaluate coherence between old and new

classificationclassification

20

Sample timetable for transition (1)Sample timetable for transition (1)

Jan Jan 20092009

BR recoded to old and new classificationBR recoded to old and new classification

20092009 Continue sampling according to old NSIC but use Continue sampling according to old NSIC but use Register information to tabulate the existing sample Register information to tabulate the existing sample against the new strata in new NSIC. Use this scheme against the new strata in new NSIC. Use this scheme as the first attempt at the sample on the new basis. as the first attempt at the sample on the new basis. For strata that are weak, in terms of their sample size, For strata that are weak, in terms of their sample size, estimate the numbers required for acceptable results estimate the numbers required for acceptable results on the new NSIC and seek approval to carry out this on the new NSIC and seek approval to carry out this top-up of the sampletop-up of the sample

21

Sample timetable for transition (2)Sample timetable for transition (2)

Jan Jan 20102010

Draw the sample on new NSIC (initially: old sample Draw the sample on new NSIC (initially: old sample tabulated against the new industries and new strata, tabulated against the new industries and new strata, with weakest strata under the new NSIC boosted by a with weakest strata under the new NSIC boosted by a top-up of the sample)top-up of the sample)

20102010 Calculate variance of the elements in the sample in Calculate variance of the elements in the sample in each of the strata based on returned data. Use sample each of the strata based on returned data. Use sample variances as approximations of the population variances as approximations of the population variances and use Neyman allocation to calculate variances and use Neyman allocation to calculate new samples.new samples.

Jan Jan 20112011

Re-allocate the sample based on reported data (in a Re-allocate the sample based on reported data (in a more efficient way now). The top-up sample can more efficient way now). The top-up sample can cease. Survey can be conducted according to new cease. Survey can be conducted according to new NSIC.NSIC.

22

Additional guidanceAdditional guidance

• More detailed guidance for developing new More detailed guidance for developing new national classification, recoding of business national classification, recoding of business register, sampling design and weight register, sampling design and weight estimation, backcasting is being prepared in estimation, backcasting is being prepared in the implementation materialsthe implementation materials