Evolution of Census Statistics on Enterprises in Italy 1996-2006: from the Traditional Census to a Register of Local Units

Monica Consalvi, Luigi Costanzo, Danila FilipponiISTAT, Division for Administrative Archives, Data and Statistical Registers

IAOS Conference 2008 – “Reshaping Official Statistics”Session 9: Use of Administrative Data in CensusShanghai, October 14-16, 2008

Shanghai, October, 14-16,

2008

IAOS Conference

20081. The starting point

EU Requirements

EEC No 2186/93 of 22 July 1993 on Community coordination in drawing up business registers for statistical purposes

Evolution of the Economic Structure

Problems of previous Censuses: survey technique Under-coverage

A.S.I.A. Project

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006 Archivio (Register)

Statistico (Statistical)

Imprese (Enterprises)

Attive (Active)

Shanghai, October, 14-16,

2008

IAOS Conference

2008

EU Requirements

EEC No 2186/93 of 22 July 1993 on Community coordination in drawing up business registers for statistical purposes

Evolution of the Economic Structure

Problems of previous Censuses: survey technique Under-coverage

A.S.I.A. ProjectStatistical Register of Active Enterprises

Objectives:

To reduce costs To reduce statistical burden To improve timeliness

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

1. The starting point

Shanghai, October, 14-16,

2008

IAOS Conference

2008 2. The experimental phase of the BR

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Definition of a metadata framework;

Study of the main available administrative archives (definition of units and characters, classifications used, coverage, maintenance and updating procedures);

Development of a “metadata translator”, i.e. a set of rules to convert the administrative data into statistical information, by identifying the statistically-relevant units among the legally-relevant ones;

Set up of a robust methodology to estimate/validate the characters of the identified statistical units.

First release of the Business Register A.S.I.A. (reference year = 1996)

The BR was validated through a field survey, with a special Intermediate Census.

Shanghai, October, 14-16,

2008

IAOS Conference

2008 3. The 1996 Intermediate Census

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Double aim:

To supply territorial information on the economic structure

General check of the information about the active units recorded in the BR

Comparison with the traditional Census:

A higher coverage rate (about 95%), thanks to the use of administrative sources;

Lesser costs and lesser burden for the respondents, thanks to the innovation of the survey technique (the units surveyed were only a 15% of the entire universe, and they had to answer less questions);

A shortening of the data processing and, then, a better timeliness in the data dissemination (one year)

New survey technique:

Personalised questionnaires with pre-printed information drawn from the BR

Shanghai, October, 14-16,

2008

IAOS Conference

2008 4. The 1996 Intermediate Census

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

A coverage error:

over-coverage, generally due to errors in coding the economic activity (especially for self-employers and enterprises without employees, as they are not covered by all the available sources;

under-coverage of some economic sectors, such as construction, transportation and trading intermediation.

But, in terms of quality of the characters:

high concordance between the characters estimated in the Register and those resulting from the survey

Evaluation of the first release of the Business Register through the Intermediate Census

Shanghai, October, 14-16,

2008

IAOS Conference

2008 5. The main features of the Business Register A.S.I.A.

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

A.S.I.A. records all the active enterprises of industry and services and their structural characters by integrating information coming from both sources

Statistical sourcesowned by Istat

Administrativesourcesmanagedby publicagenciesor private companies

Shanghai, October, 14-16,

2008

IAOS Conference

2008 6. The main features of the Business Register A.S.I.A.

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

SOURCES (OWNER) PERSONS OBLIGED TO REGISTRATION

OBSERVED UNIT TYPE BR UNITS DERIVABLE

Enterprise Register of Enterprises (Chambers of Commerce)

Entrepreneurs (excluding self-

employers) Local unit

Local unit

Natural person VAT payers

Legal person Enterprise Tax Register (Ministry

of Economy and Finances) Legal persons exempt

from VAT Legal person Enterprise / Institution

Social Security Archives (INPS)

Employers Social security position Enterprise

Yellow Pages (SEAT) Telephone customers Business consumers Local unit

Synoptic table of units recorded in the major administrative sources of the BR

InputSources

Intra-archivelink

2

Standardization

1Conceptualintegration

PhysicalIntegration

Attributesanalysis Estimation

of attributes

5

Resultsevaluation

Inter-archives link

3

StatisticalUnits

ConceptualSteps

Phases

Analysis ofClusters

4

Products (DB Tables)

Results

Input sources Standard.

1

IncludedClusters

4

Activeenterprises

5

Asia.accoppiamenti.ap

Groups of records

2

Clusters of records

3

List of active enterprises from administrative source

Active enterprises year t

Asia.imprese

6Control

correction

Identification

Macro-phase 1

Macro-phase 3

Macro-phase 2

Conceptual steps

Products(DB tables)

ResultsPhases

The BR provides a list of enterprises but does not provide information regarding territorial distribution of local units/employees

To supply territorial information on the economic

structure

To create the basis for the development – within the BR – of a register of local units, able to produce territorial data in order to achieve the major objective of a complete replacement of the Census itself

A turning point in the reshaping process of Italian business statistics: The Industry and Services’ Census of 2001

Shanghai, October, 14-16,

2008

IAOS Conference

2008 8. The 2001 Census

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

For the first time, enumerators were supplied with lists of the enumeration units located in their districts, drawn by the BR.

Some days before the survey, all the listed units received by mail (except in minor municipalities) a personalised questionnaire to be withdrawn by enumerators, partly filled in with information drawn from the BR.

Enumerators were also provided with blank (non-personalised) questionnaires; to be used only for non-listed units or in substitution of personalised questionnaires got lost or damaged.

The “register-assisted” survey:

Some elements of a classic direct door-to-door survey combined with a survey by list

Shanghai, October, 14-16,

2008

IAOS Conference

2008 9. The 2001 Census

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

The synergy between the BR and the Census

made possible a micro-level coverage analysis, based on the comparison of the data collection with an image of the BR referring to the same date

This comparison confirmed – even with the innovation of the survey technique – the inadequacy of the traditional Census survey in covering the whole observation field (about 700 thousands units out of 4 million were missed by enumerators)

Despite of its poor effectiveness, the field survey remains so far the only reliable source of statistical information about the localization of economic activities.

Shanghai, October, 14-16,

2008

IAOS Conference

2008 10. The 2001 Census

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Shanghai, October, 14-16,

2008

IAOS Conference

2008 11. The Register of Local Units

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

New EU RequirementsEC N.177/2008 of the European Parliament and of the Council of 20 February 2008 establishing a common framework for business registers for statistical purposes and repealing EEC N.2186/93

The under-coverage error of the Census survey

A.S.I.A. Local Units

ASIA-UL provides every year (with a two-year delay) the information on local units, that before was available only from the Census, that was normally taken every ten years

Shanghai, October, 14-16,

2008

IAOS Conference

2008 12. The Survey on Local Units of large Enterprises (IULGI)

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Just like the BR of enterprises, ASIA-UL is the result of the integration of administrative and statistical sources, in part already used for the construction of ASIA-Enterprises

The administrative archives available in Italy don’t provide reliable and complete information on local units, especially with regards to the territorial distribution of employees.

To fill this gap a new direct survey on local units of multi-location enterprises of large dimensions (IULGI) was necessary

In order to verify on the field the state of activity and the other characteristics of the local units.

Shanghai, October, 14-16,

2008

IAOS Conference

2008 13. The Survey on Local Units of large Enterprises (IULGI)

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

IULGI observation field

>249 employees all enterprises

100 - 249 employees a panel that rotate every 2 years

50-99 employees a panel that rotate every 3 years

IULGI is a yearly survey:in three years it is possible to cover the entire field of enterprises with more then 50 employees

All innovations introduced in the IULGI survey have made possible the achievement of a very high response rates

Shanghai, October, 14-16,

2008

IAOS Conference

2008 14. The Register of Local Units

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

L O C A L

U N I T S

ADMIN.SOURCES

IULGI

ESTIMATIONMODEL

(Large enterprises)

(Small / mediumenterprises)

Shanghai, October, 14-16,

2008

IAOS Conference

2008 15. The Register of Local Units – Estimation models

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

The activity status of the local unitsA generalized linear mixed model (GLMM) is used to model the probability that a local unit is active using administrative/statistical data as predictors, and taking into account the available longitudinal information;

The number of employees of the active local unitsIt is estimated applying a rank function to the available sources, and then solving an optimization problem with linear constraints to estimate the employees for the local units defined as active (to make the total number of employees for the local units of each enterprise consistent with the BR).

The economic activity codeWhen there are several possible codes, the choice of this attribute is based on a probabilistic procedure, which selects the most likely among the different values provided by administrative sources, making use of appropriate quality indicators, derived from the data themselves.

Shanghai, October, 14-16,

2008

IN THE PAST

to define the statistical units

for the preparation and co-ordination of surveys and for grossing up survey results, by providing the frame necessary to prevent coverage errors in data collection

IAOS Conference

2008

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

16. The role of the BR

THE NEW ROLE:

A coordinating role

cooperation among departments coherence

“core” of a coherent system of economic statistics (physical and logical connector centre)

A dissemination tool

the BR is a “live entity”, which is constantly revised and updated

Necessity of a balance between two opposite needs:

Consistency continuity/stability in time series (SBS/STS)

Accuracy/transparency in statistical information (BR) (Italian choice)

Future issue a possible reconstruction of consistent time series and the estimation of the optimum time lag after which the BR data can be considered stable

Shanghai, October, 14-16,

2008

An overview of the evolution process

IAOS Conference

2008

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006- - -

2011

17. CONCLUSIONS

Traditional enumeration of economic activities

An integrated system of statistical production

The 2011 Census will be replaced by a (much less expensive) sample survey to assess

the quality of the BR

The Business register as a continuous Census:

It provides every year statistical information on the territorial distribution of economic activities

and the employment, so far available throughthe Census every decade

Shanghai, October, 14-16,

2008

Thanks foryour attention!

IAOS Conference

2008