You are here

Statistical Business Register


The Statistical Business Register (SBR): Management and Quality


1 Presentation of the Statistical Business Register (SBR)

The TBR was created according to the Decree No. 94-780 of April 4, 1994 which consists that the INS is charged of the initialization of the Business Register, the updating, the management and the dissemination of statistics. The main partners of the INS in the management of the Business Register are the Finance Ministry (DGI) and the Social Security Fund (CNSS). Indeed the Business Register is based on the collection of enterprises’ information that can be held by different administrations as part of their own activities. Note that the individual level of the SBR is subject to the confidentiality provisions outlined in Article 5 of the Law on the National System of Statistics.

1.1 Objectives of the Business Register

The Statistical Business Register is a mainly tool for statistical coordination that allows:

  • Collecting and updating data regarding enterprises: identification, location, activities, size, dates,...
  • Harmonization and standardization of variables according to standards and statistical concepts: activity, address, size,...
  • Dissemination of business statistics
  • Monitoring of business demography
  • The use of the Business Register as a sampling and estimation frame for business surveys.

1.2 Sources of update

The TBR was created mainly on the base of two administrative sources: the DGI and the CNSS. The DGI file contains information about all legal units; therefore, it constitutes the most exhaustive source identifying enterprises. Similarly, the CNSS files (Employers and Independents) are considered as two of the largest bases identifying enterprises and their employees.

To improve the Business Register quality, INS integrated:

  • Other administrative sources: the files of industrial enterprises managed by the Agency for the Promotion of Industry and Innovation (APII) and customers data (imports and exports).
  • Individual responses of economic surveys, collected by the INS.

1.3 The contents of the Business Register

The Business Register covers four types of units: private enterprises (for profit), administrations, associations and public enterprises. Private enterprises includes three kinds of information:

  • Identification information :
  • Identifiers: (Statistical Identifier, DGI Identifier and CNSS Identifier);
  • Name of the enterprise;
  • Identification of the responsible for Independents (National Identity Card, gender, date of birth...)
  • Address (number in the street, street name, location, postal code, delegation and governorate);
  • Information about the characteristics of the enterprise:
  • Legal form of the company (PP, SA Ltd., SUARL, ...);
  • Regime (Onshore, Offshore)
  • Nationality of the enterprise (Tunisian, Foreign);
  • Activities (principal and second activities);
  • Date of creation, date of cessation and date of resumption;
  • Statistical information relating to the activity of the company:
  • Quarterly Number of salaries and their wages
  • Turnover
  • Values ​​and weight of imports and exports

1.4 Steps of update

The diagram below describes the process for managing the SBR. It consists essentially on four steps:

- The acquisition of source files: verification, standardization and attribution of statistical identifier;

- The matching of administrative files (primarily to determine the number of employees);

- The codification of the main activity: The coding of the main activity is made on the base of the enterprises declarations in the administrative files and will be fixed or approved by the surveys. The coding is done according to the official classification of activities: Tunisia Activities Classification (NAT).

- The update of the TBR: integration of new creations, modifications, ...

2 Quality Improvement of the SBR

The quality of the Business Register is very important for all users. It is therefore important to determine the required level of quality, and have a method to measure and improve the quality of the register.

To improve the quality of the Business Register content, the INS, first, diversified sources of update and, second, developed statistical models to estimate the key variables: the status of activity, address, economic activity, number of employees and turnover.

2.1 False active

The problem of false active is partly explained by the gap between the occurrence of an event and its registration in the administrative files and then in the business register. Note that for example the registration of cessation is not made during the same year of cessation: 25% of cessation cases make declaration after more than two years of their real cessation.

Indeed we can find enterprises which are really inactive but still active in the Business Register. To resolve this problem, the INS developed a deterministic model based on activity signals (administrative declarations) that assigns to each unit the probability to be active. This model shows, for example the case of 2009, that false active are mainly concentrated in the class of individual enterprises (independents). The rate of false active in this class is around 11%.

In the absence of data needed to estimate the status of activity of the business register for the period 1996-2012, the statistics of this publication does not take account the statistical status of activity but rather its administrative status of activity. In other words, the number of individual enterprises is probably overestimated by 11%.

2.2 Accuracy of the address

According to the recommendations of the International Guidelines on SBRs, the addresses must be registered at the most detailed level (governorate, delegation, postal code, street name, number in the street, phone number, web address, etc. site).

The address defects are due to:

  • Luck of detail in the declaration made by the enterprise.
  • Change of address not declared by the enterprise.

The rate of administrative information which presents problems of address is around 13.9% for micro-enterprises and 8% for enterprises employing six or more salaries (PMGE).

To resolve this problem, the INS developed a deterministic model based on a priority order between different sources (statistical and administrative). In the absence of additional sources, the correction of address for micro-enterprises remains limited. Note that we have a good quality on the level of the "governorate" for all units.

2.3 Accuracy of activity

The main activity of an enterprise is the activity that contributes the most to its added value. In our case, in the absence of this information by activity, we codify the main activity according to the enterprises declarations registered on administrative files. These sources don’t use the official classification of activities used by the INS. And the information about the activity presents some problems:

  • Sometimes the declarations are just intentions of activity
  • Administrative files are not always updated when there is a change in activity
  • And do not always have the necessary detail for a NAT code.

To improve the quality of the activity, the INS diversified sources of update (APII, customers’ files and survey results). The quality of the activity for micro-enterprises remains relative to the reliability of the DGI files.

2.4 Accuracy of the number of employees

Actually, the CNSS is the only source to update data about employees. This source contains the quarterly declarations of the employer. The size of the company is derived from the average of the four quarters.

A comparison was made between the employment covered by the Business Register and the employment covered by the social survey (which contains formal and informal employment) shows that the coverage of total employment in the Business Register is around 77% in private sector excluding agriculture and around 91% excluding construction.