Abstract
The methodological soundness of statistics in Tunisia is consistent in several areas with outdated international statistical standards. The Central Bank of Tunisia (BCT) is responsible for the collection, processing, and dissemination of monetary and balance of payments statistics. The accuracy and reliability of statistics in Tunisia are broadly adequate. Adequate national accounts data and some brief methodological notes are disseminated, albeit with no analyses accompanying the data. However, there is now a need to proceed quickly with the migration to the new standards.
I. Overall Assessment
1. The quality of macroeconomic statistics in Tunisia has improved over the past decade, and the statistics have been broadly adequate for macroeconomic analysis and policy design and monitoring. The Tunisian authorities have demonstrated their commitment to the quality of macroeconomic statistics by their subscription to the Special Data Dissemination Standard (SDDS) and the related significant steps they have taken, as well as their support to the preparation of this Report on the Observance of Standards and Codes (ROSC). The authorities are currently creating the conditions for accelerating the development of the financial system and moving towards full capital account convertibility. At the same time, they are focusing on managing the domestic macroeconomic environment with a view to achieving sustainable growth and employment gains. In this context, there is a need to pursue plans to further improve the quality of the country’s statistics through, inter alia, a closer alignment with internationally accepted standards and good practices. Some of these plans have already been clearly articulated and are in the process of being implemented, while others need to be fleshed out in the period ahead. To achieve the above mentioned macroeconomic objectives, there is also a need to give greater weight to the accessibility of statistics.
2. Tunisia has a broadly effective legal and institutional framework, as well as a culture of good practices, that together support the quality of macroeconomic statistics. The staff resources available for statistical work in all areas are not fully adequate, and especially for government finance statistics (GFS). The work of all statistical agencies is based on professionalism, transparency, and adherence to ethical standards. Official statistics follow methodologies, which in most areas need to be updated. In particular, there is a need to complete the adoption of 1993 SNA for the national accounts, implement the plan to adopt the recommendations of the Monetary and Financial Statistics Manual (MFSM), and put forward a plan to adopt the methodology recommended in the Government Finance Statistics Manual 2001 (GFSM 2001). Tunisia’s macroeconomic statistics are, to a large extent, reconcilable within each dataset and with other datasets. There are opportunities to improve the dissemination of statistics as well as information on methodology and assistance available to users in most assessed areas, and particularly in GFS.
3. Tunisia subscribed to the SDDS on June 20, 2001 and started posting its metadata on the Dissemination Standards Bulletin Board (DSBB) on the same date. Tunisia is in observance of the SDDS, meeting the specifications for coverage, periodicity, timeliness, as well as for the dissemination of advance release calendars. However, coverage of international reserves and foreign currency liquidity only partly meets SDDS requirements. Tunisia uses the flexibility options on the timeliness of the national accounts and on the periodicity and timeliness of employment and unemployment data. The metadata for these indicators do not specify any plans for improvement to meet the relevant SDDS requirements. While Tunisia has posted summary methodology pages for most data categories, preparatory work on several other data categories (employment, unemployment, interest rates, stock market share price index, exchange rate, and population) could be accelerated. Tunisia regularly updates its advance release calendar for all data categories, which is also posted on the DSBB. Appendix I provides an overview of Tunisia’s dissemination practices compared to the SDDS.
4. Section II of this ROSC—Data Module provides a summary assessment by agency and dataset based on a four-part scale. This is followed by staff recommendations in Section III. Practices compared to the SDDS are summarized in Appendix I. The authorities’ response to this report and a volume of detailed assessments are presented in separate documents.
5. In applying the IMF’s Data Quality Assessment Framework (DQAF July 2003), the remainder of this section presents the mission’s main conclusions. The presentation is done at the level of the DQAF’s quality dimensions, by agency for the first two dimensions and by dataset for the remaining four.
Prerequisites of Quality and Assurances of Integrity
6. Tunisia’s law on the National Statistical System (SNS) provides a proper environment for the production of official statistics at the national level, where the National Statistics Institute (INS) is the central body. The institutional coordination and planning of statistical activities is ensured by the National Council on Statistics (CNS), which is an advisory body whose membership comprises both producers and users of statistics. As the central executive body of the SNS, the INS ensures the technical coordination of statistical activities. There is an adequate practice of data exchange among the INS and the other government agencies. The law establishes the compulsory and confidential character of data reporting to the INS. Resources allocated to the INS for the conduct of surveys, as well as for equipment, are broadly sufficient, but staff resources are too limited. Users’ feedback is sought by the INS in a wide variety of ways. The INS strives to improve the quality of its statistics and to meet international standards—all in the framework of its work program that is part of a five-year national statistical plan. Legal framework and actual practices support the independence of the INS in its work and the professionalism of its staff, who are subject to the statute of civil servants. However, the terms and conditions under which statistics are collected, processed, and disseminated, as well as the relevant metadata, are not as fully available to the domestic public as they could be.
7. Tunisia’s legal framework clearly assigns the responsibility of collecting, processing, and disseminating GFS to the Ministry of Finance (MF). The MF compiles data on budget execution and debt in accordance with the Organic Budget Law. As mentioned above, there is firm legal basis for the coordination of agencies in the SNS by means of the CNS. Actual data-sharing and coordination between the MF and other agencies are satisfactory. The Directorate in charge of budget statistics is seriously understaffed but otherwise adequately equipped. The MF values the quality of its statistics, but improvements in GFS on the basis of consultation with users (including via MF’s participation in the CNS) could be sought more actively. MF staff are professional and generally well-trained, and subject to the statute of civil servants. The MF does not adequately disseminate methodological information or publicize the terms and conditions under which GFS are produced and disseminated, hampered inter alia by personnel shortages and the absence of a MF website.
8. The Central Bank of Tunisia (BCT) is responsible for the collection, processing, and dissemination of monetary and balance of payments statistics. As noted above, the law provides for the coordination of BCT with other agencies, and actual data-sharing between the agencies is satisfactory. The compulsory aspect of data reporting to the BCT is provided for in the legal framework. The relevance of BCT’s statistics is safeguarded by consultations within the context of the CNS as well as with other data users. BCT closely monitors and assesses its own statistical work against standards. Staff—although well trained and professional—and computer software are not fully adequate. The central bank law contains provisions that ensure the overall independence of the BCT and thus of the statistical work of the institution. That work is found to follow strict technical criteria and to be free of any interference. Staff are subject to and aware of the staff code of conduct. Data are disseminated adequately to all interested parties through the BCT’s web site and publications. However, the terms and conditions under which statistics are collected, processed, and disseminated, as well as some of the metadata, could be made more accessible to the public. Moreover, no advance notice of major changes in methodology, source data, and statistical technique is provided.
9. The methodological soundness of statistics in Tunisia is consistent in several areas with outdated international statistical standards. However, there is now a need to proceed quickly with the migration to the new standards. In the national accounts, work is well advanced to implement fully the 1993 SNA, and it remains to finalize and schedule publication of the new series that have been compiled. Work has also begun in accordance with a plan to adopt the MFSM for compiling monetary statistics. As for the GFS, despite an expressed interest, Tunisia has not yet developed a transition plan to GFSM 2001. Balance of payments statistics (as provided in the DSBB) are already broadly consistent with the methodology of BPM5. However, the statistics of the data template on international reserves and international liquidity are not fully consistent with the requirements of the SDDS as they do not cover foreign currency deposits of resident banks with the BCT under “pre-determined short-term net drains on foreign currency assets.” In addition, the debt service of public companies is included under this heading, without this being specifically mentioned in the country notes, contrary to the recommendations of the International Reserves and Foreign Currency Liquidity Guidelines for a Data Template. With regard to national accounts statistics, the classification and valuation of currently published data series do not conform to international standards; however, new, still unpublished, statistics are consistent with international standards. Regarding GFS, coverage of published data outside those reported in the DSBB is limited to budgetary central government. The operations of the numerous public administrative units (such as schools) have only begun to be included in the budget. Moreover, GFS classification, while very close to GFSM 1986 standards, could be improved. The coverage and scope of balance of payments statistics is good, but classification of the important item of workers’ remittances is inappropriate. There are also departures from the standard regarding recording on an accrual basis and, in national publications, regarding the breakdowns of the financial account. In monetary statistics, there are departures from the MFSM inter alia regarding disaggregation of credits by sector, valuation, and classification of assets and liabilities. For example, banks’ credit to the rest of the domestic sector (other than central government and financial corporations) is not disaggregated into subsectors; monetary gold is not valued at market price, as recommended in the MFSM.
10. The accuracy and reliability of statistics in Tunisia are broadly adequate. In the national accounts, source data are sufficient to compile reliable annual estimates, but could be further diversified to improve indicator coverage for quarterly GDP estimates. Data sources for compiling GFS are adequately detailed and suitable for deriving the published aggregates. However the scope of data sources is rather narrow. In the context of current strict foreign exchange regulations, the data sources for balance of payments statistics are fully adequate. While very comprehensive in coverage and timely, source data for monetary statistics do not contain sufficient detail for classification of assets and liabilities as recommended in the MFSM. In all areas of statistics assessed, there is careful processing and validation of source data (following up actively with reporting units) as well as of intermediate data. An outdated base year (1990) is used for the current series and there is a need to refine the statistical techniques of the quarterly national accounts estimates. Adequate statistical techniques are used in all assessed sectors. Explicit studies of—the otherwise well-thought-out—revisions are not undertaken, but experience gained in revisions is used to inform subsequent work.
11. Serviceability of the assessed macroeconomic statistics is broadly satisfactory, as confirmed by the results of the user survey conducted in the context of this assessment. Data in all assessed areas are compiled and disseminated with the periodicity and timeliness prescribed by the SDDS with the exception of the national accounts data (whose timeliness lags behind the SDDS requirement by one month for the first quarter). Statistics are largely reconcilable within each dataset and over time, as well as with other datasets. However, the reconciliation of fiscal and monetary data cannot be readily confirmed using solely published information, as budget domestic financing is not presented by debt holder. Moreover, monetary statistics data on transactions between the BCT and deposit money banks, as well as among the latter, cannot be reconciled from published data as presentations are not consistently classified. In all assessed areas, there are established revision practices that agencies follow—albeit do not publicize—and preliminary and revised data are clearly identified in most publications, except in the monthly GFS printed publication.
12. There is room to improve accessibility to statistics. Adequate national accounts data and some brief methodological notes are disseminated, albeit with no analyses accompanying the data. Regularly updated metadata are provided on the DSBB, but no link is provided from the INS’ website. National accounts statistics are released simultaneously to all users outside the government. Monthly GFS data are also disseminated without analysis or adequate publicizing of a preannounced release schedule, and their presentation could be improved. Moreover, the MF does not have a website. Methodological notes are only available on the DSBB website (in English). Monetary statistics and related methodological material are readily available to the public. However, information on how these statistics are compiled is not publicized. Accessibility of balance of payments statistics is also broadly adequate, but accessibility to metadata could be improved by posting them on the BCT website. Adequate assistance to users is provided for national accounts and monetary data. However much greater assistance can be provided to GFS users. It is also possible to further improve assistance to users of balance of payments statistics.
II. Assessment by Agency and Dataset
13. Assessment of the quality of four macroeconomic datasets—national accounts, government finance, monetary, and balance of payments statistics—were conducted using the DQAF July 2003. In this section, the results are presented at the level of the DQAF elements and using a four-point rating scale (Table 1). Assessments of the prerequisites of data quality and the assurances of integrity (Dimensions “0” and “1” of the DQAF) are presented in Tables 2a–c. For each dataset, the assessment of methodological soundness, accuracy and reliability, serviceability, and accessibility (Dimensions “2” to “5” of the DQAF) are shown in Tables 3a–d.
Tunisia: Data Quality Assessment Framework July 2003—Summary Results
Tunisia: Data Quality Assessment Framework July 2003—Summary Results
| Key to symbols: O = Practice Observed; LO = Practice Largely Observed; LNO =Practice Largely Not Observed; NO = Practice Not Observed; NA = Not Applicable | ||||
|---|---|---|---|---|
| Datasets | National Accounts | Government Finance Statistics | Monetary Statistics | Balance of Payments Statistics |
| Dimensions/Elements | ||||
| 0. Prerequisites of quality | ||||
| 0.1 Legal and institutional environment | O | O | O | O |
| 0.2 Resources | LO | LNO | LO | LO |
| 0.3 Relevance | O | LO | O | O |
| 0.4 Other quality management | O | O | O | O |
| 1. Assurances of integrity | ||||
| 1.1 Professionalism | O | O | O | O |
| 1.2 Transparency | LO | LO | O | O |
| 1.3 Ethical standards | O | O | O | O |
| 2. Methodological soundness | ||||
| 2.1 Concepts and definitions | LO | LO | LO | LO |
| 2.2 Scope | O | LO | O | O |
| 2.3 Classification/sectorization | LNO | LO | LO | LO |
| 2.4 Basis for recording | LNO | O | LO | LO |
| 3. Accuracy and reliability | ||||
| 3.1 Source data | LO | O | LO | O |
| 3.2 Assessment of source data | O | O | O | O |
| 3.3 Statistical techniques | LO | O | O | O |
| 3.4 Assessment and validation of intermediate data and statistical outputs | O | O | O | O |
| 3.5 Revision studies | LNO | LO | LO | LO |
| 4. Serviceability | ||||
| 4.1 Periodicity and timeliness | LO | O | O | O |
| 4.2 Consistency | O | LO | LO | O |
| 4.3 Revision policy and practice | LO | LO | LO | LO |
| 5. Accessibility | ||||
| 5.1 Data accessibility | O | LO | O | O |
| 5.2 Metadata accessibility | LO | LO | LO | LO |
| 5.3 Assistance to users | O | LNO | O | LO |
Tunisia: Data Quality Assessment Framework July 2003—Summary Results
| Key to symbols: O = Practice Observed; LO = Practice Largely Observed; LNO =Practice Largely Not Observed; NO = Practice Not Observed; NA = Not Applicable | ||||
|---|---|---|---|---|
| Datasets | National Accounts | Government Finance Statistics | Monetary Statistics | Balance of Payments Statistics |
| Dimensions/Elements | ||||
| 0. Prerequisites of quality | ||||
| 0.1 Legal and institutional environment | O | O | O | O |
| 0.2 Resources | LO | LNO | LO | LO |
| 0.3 Relevance | O | LO | O | O |
| 0.4 Other quality management | O | O | O | O |
| 1. Assurances of integrity | ||||
| 1.1 Professionalism | O | O | O | O |
| 1.2 Transparency | LO | LO | O | O |
| 1.3 Ethical standards | O | O | O | O |
| 2. Methodological soundness | ||||
| 2.1 Concepts and definitions | LO | LO | LO | LO |
| 2.2 Scope | O | LO | O | O |
| 2.3 Classification/sectorization | LNO | LO | LO | LO |
| 2.4 Basis for recording | LNO | O | LO | LO |
| 3. Accuracy and reliability | ||||
| 3.1 Source data | LO | O | LO | O |
| 3.2 Assessment of source data | O | O | O | O |
| 3.3 Statistical techniques | LO | O | O | O |
| 3.4 Assessment and validation of intermediate data and statistical outputs | O | O | O | O |
| 3.5 Revision studies | LNO | LO | LO | LO |
| 4. Serviceability | ||||
| 4.1 Periodicity and timeliness | LO | O | O | O |
| 4.2 Consistency | O | LO | LO | O |
| 4.3 Revision policy and practice | LO | LO | LO | LO |
| 5. Accessibility | ||||
| 5.1 Data accessibility | O | LO | O | O |
| 5.2 Metadata accessibility | LO | LO | LO | LO |
| 5.3 Assistance to users | O | LNO | O | LO |
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—National Statistics Institute
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—National Statistics Institute
| 0. Prerequisites of quality | 1. Assurances of integrity |
|---|---|
| Legal and institutional environment The terms and conditions regulating the statistical activities are specified in Law No. 99-32 of April 13, 1999 on the SNS. The INS, as the central body of the SNS, is responsible, in collaboration with the other public agencies, for the compilation, production, analysis, and dissemination of statistics. Although production of the national accounts statistics is not explicitly spelled out in the Statistical Law, in undisputed practice it is carried out by the INS by tradition since the INS’s creation. As an advisory body, the CNS ensures adequate coordination of statistical activities thus avoiding duplication of efforts at the national level. The main coordinating mechanism is the five year statistical program, which is developed by the CNS based on the work program proposals of government agencies and submitted for the government’s approval. Data exchange among agencies is adequate. Confidentiality of individual information is legally guaranteed; although the penalties for noncompliance are insignificant, compliance is by-and-large observed in practice. Resources Budget allocations for surveys as well as computer and material equipment are broadly sufficient. However, staff resources are too tight for the current work program despite a large recruitment of qualified staff in recent years. Relevance Users’ feedback is sought in various ways, notably through the meetings of working groups of the CNS, during the elaboration of the work program, or through well-established mechanisms for collecting the views of users. Other quality management Managers strive to ensure the quality of statistics with regard to trade-offs between improvements and available resources. Improvements are introduced to address problems in data production processes and adopt international standards. Performance is routinely monitored in relation to the work program and its commitments in the national statistical program. | Professionalism The Statistical Law provides the INS with scientific and professional independence. Statistics are compiled on an impartial basis. INS is independent in its choice of appropriate source data and statistical techniques. Efforts are made to improve underlying statistical methodologies, consistent with international standards. Recruitment and promotion of staff are based on professional qualifications and technical competence. The general director is appointed and dismissed by the President of the Republic. The INS is entitled to comment on the misuse or misinterpretation of its statistical products. Transparency The Statistical Law was published in the JORT, and copies can be found either at the INS or by accessing the website of the CNS. However, the INS publications and website do not quote the Statistical Law. The government’s internal access to data prior to their release is not adequately publicized, except in the metadata posted on the IMF’s DSBB website, to which there is no hyperlink in the official websites. The INS clearly identifies its statistical products and requests attribution when its statistics are reproduced by others. Advance notice of major changes in statistical methodology or availability of data is given to a limited group of users. Ethical standards The INS is a member of the CNS, which oversees ethical standards in statistics (Article 14 of Law 99-32). INS staff operate under the statute of civil servants, which includes rules of ethics. The guidelines on ethics are made known to the staff upon employment. |
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—National Statistics Institute
| 0. Prerequisites of quality | 1. Assurances of integrity |
|---|---|
| Legal and institutional environment The terms and conditions regulating the statistical activities are specified in Law No. 99-32 of April 13, 1999 on the SNS. The INS, as the central body of the SNS, is responsible, in collaboration with the other public agencies, for the compilation, production, analysis, and dissemination of statistics. Although production of the national accounts statistics is not explicitly spelled out in the Statistical Law, in undisputed practice it is carried out by the INS by tradition since the INS’s creation. As an advisory body, the CNS ensures adequate coordination of statistical activities thus avoiding duplication of efforts at the national level. The main coordinating mechanism is the five year statistical program, which is developed by the CNS based on the work program proposals of government agencies and submitted for the government’s approval. Data exchange among agencies is adequate. Confidentiality of individual information is legally guaranteed; although the penalties for noncompliance are insignificant, compliance is by-and-large observed in practice. Resources Budget allocations for surveys as well as computer and material equipment are broadly sufficient. However, staff resources are too tight for the current work program despite a large recruitment of qualified staff in recent years. Relevance Users’ feedback is sought in various ways, notably through the meetings of working groups of the CNS, during the elaboration of the work program, or through well-established mechanisms for collecting the views of users. Other quality management Managers strive to ensure the quality of statistics with regard to trade-offs between improvements and available resources. Improvements are introduced to address problems in data production processes and adopt international standards. Performance is routinely monitored in relation to the work program and its commitments in the national statistical program. | Professionalism The Statistical Law provides the INS with scientific and professional independence. Statistics are compiled on an impartial basis. INS is independent in its choice of appropriate source data and statistical techniques. Efforts are made to improve underlying statistical methodologies, consistent with international standards. Recruitment and promotion of staff are based on professional qualifications and technical competence. The general director is appointed and dismissed by the President of the Republic. The INS is entitled to comment on the misuse or misinterpretation of its statistical products. Transparency The Statistical Law was published in the JORT, and copies can be found either at the INS or by accessing the website of the CNS. However, the INS publications and website do not quote the Statistical Law. The government’s internal access to data prior to their release is not adequately publicized, except in the metadata posted on the IMF’s DSBB website, to which there is no hyperlink in the official websites. The INS clearly identifies its statistical products and requests attribution when its statistics are reproduced by others. Advance notice of major changes in statistical methodology or availability of data is given to a limited group of users. Ethical standards The INS is a member of the CNS, which oversees ethical standards in statistics (Article 14 of Law 99-32). INS staff operate under the statute of civil servants, which includes rules of ethics. The guidelines on ethics are made known to the staff upon employment. |
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—Ministry of Finance
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—Ministry of Finance
| 0. Prerequisites of quality | 1. Assurances of integrity |
|---|---|
| Legal and institutional environment Decree No. 2005-492 of March 1, 2005 assigns the responsibility of collecting, processing, and disseminating fiscal statistics to the MF’s General Directorate of Resources and Balance (DGRE). The above decree assigns the compilation of public and publicly-guaranteed debt to the MF’s General Directorate of Debt Management. The MF compiles data on budget execution and debt in accordance with the Organic Budget Law (Law No. 2004-42 of May 13, 2004). The legal underpinnings for coordination between the MF and the BCT on financing and debt statistics is Law No. 99-32 of April 13, 1999 on the SNS, which establishes the CNS, a statistical coordination body. Actual foreign debt data-sharing (through the SIADE system) and coordination between the MF and BCT are said by the DGGD to be satisfactory. Resources The DGRE is seriously understaffed, with only two people in charge of compiling statistics on government accounts. The DGRE and the DGGD are otherwise adequately equipped. With such staff resource constraints, DGRE staff are tightly managed through redeployments, as needed. A staff evaluation system, common to all public agencies, is in place. Costs of producing GFS are not quantified. Relevance Most DGRE staff are well-trained, and one of the two compilers is very experienced in GFS methodology. However, staff knowledge does not translate into proactive support to users. Moreover, users are not systematically consulted on aspects of GFS data improvement. In this respect, the role that the CNS could play as a discussion forum between data producers and users has not been fully explored yet. Other quality management Articles 3 and 4 of Law 99-32 emphasize statistical quality, and this issue is explicitly inculcated in the staff. The staff is well aware of the trade-off between timeliness and coverage needed to achieve high quality data. The priority in GFS work is to minimize errors. | Professionalism Objectivity and independence in data compilation are outlined in Article 3 of Law 99-32. Professional qualifications and competence of GFS compilers are high, and the data they process (mostly carrying out methodologically-mandated reclassifications) are derived from public accounting records. The DGRE and DGGD have the full authority to comment through press releases in case of erroneous interpretations of GFS statistics. Transparency While the Organic Budget Law, various laws relating to the final budget accounts, national accounting standards, and budget nomenclature are all published in the JORT in both Arabic and French, the MF’s GFS publications do not provide the terms and conditions under which GFS are produced and disseminated. There is no MF website. In the GFS tables posted on Tunisia’s SDDS website (DSBB), the products of specific units within the MF are not clearly identified as such. Outside the MF, no government bodies have access to the data prior to their release. Prior notice is not always given before a change in the database. For example, the change in VAT classification was announced simultaneously with the first release of the new data. Ethical standards The MF is a member of the CNS, which oversees ethical standards in statistics (Article 14 of Law 99-32). The MF staff also operates under the statute of civil servants, which includes rules of ethics. The guidelines on ethics are made known to the staff upon employment. |
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—Ministry of Finance
| 0. Prerequisites of quality | 1. Assurances of integrity |
|---|---|
| Legal and institutional environment Decree No. 2005-492 of March 1, 2005 assigns the responsibility of collecting, processing, and disseminating fiscal statistics to the MF’s General Directorate of Resources and Balance (DGRE). The above decree assigns the compilation of public and publicly-guaranteed debt to the MF’s General Directorate of Debt Management. The MF compiles data on budget execution and debt in accordance with the Organic Budget Law (Law No. 2004-42 of May 13, 2004). The legal underpinnings for coordination between the MF and the BCT on financing and debt statistics is Law No. 99-32 of April 13, 1999 on the SNS, which establishes the CNS, a statistical coordination body. Actual foreign debt data-sharing (through the SIADE system) and coordination between the MF and BCT are said by the DGGD to be satisfactory. Resources The DGRE is seriously understaffed, with only two people in charge of compiling statistics on government accounts. The DGRE and the DGGD are otherwise adequately equipped. With such staff resource constraints, DGRE staff are tightly managed through redeployments, as needed. A staff evaluation system, common to all public agencies, is in place. Costs of producing GFS are not quantified. Relevance Most DGRE staff are well-trained, and one of the two compilers is very experienced in GFS methodology. However, staff knowledge does not translate into proactive support to users. Moreover, users are not systematically consulted on aspects of GFS data improvement. In this respect, the role that the CNS could play as a discussion forum between data producers and users has not been fully explored yet. Other quality management Articles 3 and 4 of Law 99-32 emphasize statistical quality, and this issue is explicitly inculcated in the staff. The staff is well aware of the trade-off between timeliness and coverage needed to achieve high quality data. The priority in GFS work is to minimize errors. | Professionalism Objectivity and independence in data compilation are outlined in Article 3 of Law 99-32. Professional qualifications and competence of GFS compilers are high, and the data they process (mostly carrying out methodologically-mandated reclassifications) are derived from public accounting records. The DGRE and DGGD have the full authority to comment through press releases in case of erroneous interpretations of GFS statistics. Transparency While the Organic Budget Law, various laws relating to the final budget accounts, national accounting standards, and budget nomenclature are all published in the JORT in both Arabic and French, the MF’s GFS publications do not provide the terms and conditions under which GFS are produced and disseminated. There is no MF website. In the GFS tables posted on Tunisia’s SDDS website (DSBB), the products of specific units within the MF are not clearly identified as such. Outside the MF, no government bodies have access to the data prior to their release. Prior notice is not always given before a change in the database. For example, the change in VAT classification was announced simultaneously with the first release of the new data. Ethical standards The MF is a member of the CNS, which oversees ethical standards in statistics (Article 14 of Law 99-32). The MF staff also operates under the statute of civil servants, which includes rules of ethics. The guidelines on ethics are made known to the staff upon employment. |
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—Central Bank of Tunisia
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—Central Bank of Tunisia
| 0. Prerequisites of quality | 1. Assurances of integrity |
|---|---|
| Legal and institutional environment. The Law 58-90 of 9/19/1958 on the creation and organization of the BCT (amended by laws 88-119 et 2000-37) and several BCT circulars (e.g., circulars No. 93-08 of 7/10/93 and 96-11 of 9/24/96) constitute the basic regulatory framework for the data collection, processing, and dissemination. Although the provisions in the Law are not explicit, there is a long established tradition and well established practices regarding BCT’s statistical work. Article 34 (4. (new) (5)) of Law 58-90 empowers the BCT to collect data for the purpose of monitoring trends in credit and short-term economic conditions, while circulars No. 93-08 and 96-11 set out requirements for banks to prepare and report data to the BCT. Article 71 (new) allows the BCT to publish any documents of an economic, monetary, or banking nature. The confidentiality of the reported data is guaranteed by the BCT’s staff code of conduct. Resources. Staff and computer resources are not fully adequate. There was recently high turnover of staff, and currently some new staff are under training and are not yet fully involved in the statistical production process. The number of staff could be further increased. The software developed by the BCT’s own staff for monetary statistics is sufficient for the current compilation of monetary statistics, but will need to be updated for implementing the MFSM. The software developed for balance of payments statistics is fully satisfactory. Staff salaries are considered competitive with the public sector. No separate budgeting process takes place for the statistical work. Relevance. The relevance of monetary and balance of payments statistics is safeguarded by consultations with the CNS’s members. Interaction with other BCT departments and other data users outside the BCT (e.g., with the wide membership of the CNS) also takes place and helps identify new data requirements. Other quality management. Computerized procedures are available for data checking and validation alongside staff expertise. Work on statistics is scheduled to meet the deadlines of data dissemination as indicated in the advance release calendar published on the BCT’s website. The entire work is closely monitored and assessed against the standards set out in the BCT’s Circular 93-08 and the advance release calendar. Staff training is always a priority for the BCT. | Professionalism. The Law 58-90 of 9/19/1958 contains provisions that promote the independence of the BCT. For example, Articles 2 and 3 state that “The BCT is a domestic public entity with legal personality and financial autonomy … [it] shall not be subject to the laws and regulations governing the public accounts of the state.” Articles 8 (new) and 9 (new) further stipulate that the Governor is appointed by decree for a renewable six-year term of office and can only be removed from office by decree. These provisions appear to be sufficient to support the statistical work of the BCT, which is found to follow strict technical criteria. There has been no political interference in the BCT’s statistical work. Professionalism is underpinned by a transparent and competitive recruitment process. The recruitment is announced in newspapers and the selection is based exclusively on merit. Staff promotion is based on professional competence that is determined by the annual staff performance review. Staff are subject to rules contained in the staff code of conduct, which contribute to safeguarding and strengthening staff professionalism. Transparency. Data are disseminated simultaneously to all interested parties through the BCT’s website and in the BCT’s QFSB. A section on methodology in the BCT’s QFSB explains any changes to the data series. No officials outside the BCT have access to the data prior to their release. The Law 58-90 and selected circulars that support the statistical work are posted on the BCT’s website. The other laws and circulars are available from the BCT upon request. While no advance notice is provided, information on changes to the data series are announced when the data are released. Ethical standards. Staff members are subject to rules contained in the staff code of conduct, which provides guidelines for staff behavior. New staff are made aware of this code of conduct when they join the BCT and these standards are reinforced through briefing. Staff are reminded of these ethical standards when behavior is found to depart from these standards. |
Tunisia: Assessment of Data Quality—Dimensions 0 and 1—Central Bank of Tunisia
| 0. Prerequisites of quality | 1. Assurances of integrity |
|---|---|
| Legal and institutional environment. The Law 58-90 of 9/19/1958 on the creation and organization of the BCT (amended by laws 88-119 et 2000-37) and several BCT circulars (e.g., circulars No. 93-08 of 7/10/93 and 96-11 of 9/24/96) constitute the basic regulatory framework for the data collection, processing, and dissemination. Although the provisions in the Law are not explicit, there is a long established tradition and well established practices regarding BCT’s statistical work. Article 34 (4. (new) (5)) of Law 58-90 empowers the BCT to collect data for the purpose of monitoring trends in credit and short-term economic conditions, while circulars No. 93-08 and 96-11 set out requirements for banks to prepare and report data to the BCT. Article 71 (new) allows the BCT to publish any documents of an economic, monetary, or banking nature. The confidentiality of the reported data is guaranteed by the BCT’s staff code of conduct. Resources. Staff and computer resources are not fully adequate. There was recently high turnover of staff, and currently some new staff are under training and are not yet fully involved in the statistical production process. The number of staff could be further increased. The software developed by the BCT’s own staff for monetary statistics is sufficient for the current compilation of monetary statistics, but will need to be updated for implementing the MFSM. The software developed for balance of payments statistics is fully satisfactory. Staff salaries are considered competitive with the public sector. No separate budgeting process takes place for the statistical work. Relevance. The relevance of monetary and balance of payments statistics is safeguarded by consultations with the CNS’s members. Interaction with other BCT departments and other data users outside the BCT (e.g., with the wide membership of the CNS) also takes place and helps identify new data requirements. Other quality management. Computerized procedures are available for data checking and validation alongside staff expertise. Work on statistics is scheduled to meet the deadlines of data dissemination as indicated in the advance release calendar published on the BCT’s website. The entire work is closely monitored and assessed against the standards set out in the BCT’s Circular 93-08 and the advance release calendar. Staff training is always a priority for the BCT. | Professionalism. The Law 58-90 of 9/19/1958 contains provisions that promote the independence of the BCT. For example, Articles 2 and 3 state that “The BCT is a domestic public entity with legal personality and financial autonomy … [it] shall not be subject to the laws and regulations governing the public accounts of the state.” Articles 8 (new) and 9 (new) further stipulate that the Governor is appointed by decree for a renewable six-year term of office and can only be removed from office by decree. These provisions appear to be sufficient to support the statistical work of the BCT, which is found to follow strict technical criteria. There has been no political interference in the BCT’s statistical work. Professionalism is underpinned by a transparent and competitive recruitment process. The recruitment is announced in newspapers and the selection is based exclusively on merit. Staff promotion is based on professional competence that is determined by the annual staff performance review. Staff are subject to rules contained in the staff code of conduct, which contribute to safeguarding and strengthening staff professionalism. Transparency. Data are disseminated simultaneously to all interested parties through the BCT’s website and in the BCT’s QFSB. A section on methodology in the BCT’s QFSB explains any changes to the data series. No officials outside the BCT have access to the data prior to their release. The Law 58-90 and selected circulars that support the statistical work are posted on the BCT’s website. The other laws and circulars are available from the BCT upon request. While no advance notice is provided, information on changes to the data series are announced when the data are released. Ethical standards. Staff members are subject to rules contained in the staff code of conduct, which provides guidelines for staff behavior. New staff are made aware of this code of conduct when they join the BCT and these standards are reinforced through briefing. Staff are reminded of these ethical standards when behavior is found to depart from these standards. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—National Accounts
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—National Accounts
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions Although the System of National Accounts 1968 (1968 SNA) is broadly followed, an important number of conceptual elements of the System of National Accounts 1993 (1993 SNA) are implemented (institutional sector accounts, estimates of unrecorded economy, etc.). Work is advanced to fully adopt the 1993 SNA, and data series are produced according to it, but are not yet finalized to be published. Scope The range of tables and accounts currently produced and published by the INS exceeds the minimum requirements of the Inter- Secretariat Working Group on National Accounts (ISWGNA) for the implementation of the 1993 SNA. The scope of production and assets boundaries is broadly in accord with international standards. Classification/sectorization In the currently published series, the classification of units, transactions and sectors is largely in accordance with the 1993 SNA. However, classifications of activities, products, government functions, and individual consumption used in the currently published series do not correspond to international standards. On the contrary, all classifications adopted in the new series of accounts (which are yet to be finalized) are derived from the international standards. Basis for recording Valuation of flows and stocks broadly follows the 1968 SNA. Recording is on cash basis but differences with accrual recording are noted in government and construction areas only. | Source data Source data are broadly sufficient to compile reliable annual estimates, and they could be further diversified to improve indicator coverage for quarterly GDP estimates (agriculture, construction, trade and other services), particularly expenditure-based ones and volume measures. International classifications used by source data require “bridges” to conform to those used by national accounts. Timeliness is broadly adequate. Assessment of source data Survey results and administrative data are regularly checked for accuracy. When outliers are detected in source data, reporting units/agencies are contacted to verify the data. Statistical techniques Adjustments are done to source data. Statistical techniques are adequate for the annual accounts but need some refinement for the quarterly estimates. An outdated base year (1990) is used for the current series. 2000 is used as base year for the new series that will be released at a date to be decided. Assessment and validation of intermediate data and statistical outputs Intermediate data and statistical outputs are routinely assessed. Estimates are validated against other information. Revision studies Magnitude of regular revisions is taken into account to inform ongoing work but no explicit studies are undertaken to identify trends or systematic biases. Methodological revisions for the new series are internally documented. | Periodicity and timeliness Periodicity follows the SDDS requirements. Timeliness lags behind the SDDS requirements by about one month for the first quarter. Consistency Statistics are broadly consistent in their aggregation and within time series. Data can be reconciled across data sets. Eventual small differences inside the aggregates are due to different concepts or classifications used, but totals are consistent. Revision policy and practice There is a regular schedule and practice of revisions, but these are not adequately publicized. Preliminary and revised data are clearly identified in publications. | Data accessibility National accounts are disseminated in a clear manner and at different levels of disaggregation in hard copy (monthly and annual publications), as well as via the internet. However, no analyses accompany the data. Statistics are released simultaneously to all users outside government. Metadata accessibility Brief methodological notes are disseminated. A more detailed description of sources and methods was published in 1992 but no update followed, and reference to it is not publicized. Updated and more detailed information can be provided on demand. No metadata for quarterly accounts are published. The metadata provided in DSBB are updated regularly, but no link is provided from the INS’ website. Assistance to users Support to users is provided at the INS. In addition, sufficient contact information is available on the website and in the publications. A catalog of publications is available on the website or in hard copy. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—National Accounts
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions Although the System of National Accounts 1968 (1968 SNA) is broadly followed, an important number of conceptual elements of the System of National Accounts 1993 (1993 SNA) are implemented (institutional sector accounts, estimates of unrecorded economy, etc.). Work is advanced to fully adopt the 1993 SNA, and data series are produced according to it, but are not yet finalized to be published. Scope The range of tables and accounts currently produced and published by the INS exceeds the minimum requirements of the Inter- Secretariat Working Group on National Accounts (ISWGNA) for the implementation of the 1993 SNA. The scope of production and assets boundaries is broadly in accord with international standards. Classification/sectorization In the currently published series, the classification of units, transactions and sectors is largely in accordance with the 1993 SNA. However, classifications of activities, products, government functions, and individual consumption used in the currently published series do not correspond to international standards. On the contrary, all classifications adopted in the new series of accounts (which are yet to be finalized) are derived from the international standards. Basis for recording Valuation of flows and stocks broadly follows the 1968 SNA. Recording is on cash basis but differences with accrual recording are noted in government and construction areas only. | Source data Source data are broadly sufficient to compile reliable annual estimates, and they could be further diversified to improve indicator coverage for quarterly GDP estimates (agriculture, construction, trade and other services), particularly expenditure-based ones and volume measures. International classifications used by source data require “bridges” to conform to those used by national accounts. Timeliness is broadly adequate. Assessment of source data Survey results and administrative data are regularly checked for accuracy. When outliers are detected in source data, reporting units/agencies are contacted to verify the data. Statistical techniques Adjustments are done to source data. Statistical techniques are adequate for the annual accounts but need some refinement for the quarterly estimates. An outdated base year (1990) is used for the current series. 2000 is used as base year for the new series that will be released at a date to be decided. Assessment and validation of intermediate data and statistical outputs Intermediate data and statistical outputs are routinely assessed. Estimates are validated against other information. Revision studies Magnitude of regular revisions is taken into account to inform ongoing work but no explicit studies are undertaken to identify trends or systematic biases. Methodological revisions for the new series are internally documented. | Periodicity and timeliness Periodicity follows the SDDS requirements. Timeliness lags behind the SDDS requirements by about one month for the first quarter. Consistency Statistics are broadly consistent in their aggregation and within time series. Data can be reconciled across data sets. Eventual small differences inside the aggregates are due to different concepts or classifications used, but totals are consistent. Revision policy and practice There is a regular schedule and practice of revisions, but these are not adequately publicized. Preliminary and revised data are clearly identified in publications. | Data accessibility National accounts are disseminated in a clear manner and at different levels of disaggregation in hard copy (monthly and annual publications), as well as via the internet. However, no analyses accompany the data. Statistics are released simultaneously to all users outside government. Metadata accessibility Brief methodological notes are disseminated. A more detailed description of sources and methods was published in 1992 but no update followed, and reference to it is not publicized. Updated and more detailed information can be provided on demand. No metadata for quarterly accounts are published. The metadata provided in DSBB are updated regularly, but no link is provided from the INS’ website. Assistance to users Support to users is provided at the INS. In addition, sufficient contact information is available on the website and in the publications. A catalog of publications is available on the website or in hard copy. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Government Finance Statistics
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Government Finance Statistics
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions The overall structure, concepts, definitions are generally consistent with GFSM 1986. However, Tunisia does not have in place a transition plan to GFSM 2001. Scope The coverage of published monthly and annual data is limited to budgetary central government. Over 90 percent of public administrative units (EPA) such as hospitals and schools, are not consolidated, neither are the social security funds or local governments. However, for the purposes of SDDS, summary annual data on general government (still excluding EPA) are prepared for posting on the DSBB. Classification/sectorization Domestic financing by type of debt holder (bank/nonbank) is not published. Remaining classifications are broadly consistent with GFSM 1986. Basis for recording Government operations are recorded on a cash basis. Gross debt is valued at face value. Foreign currency flows are converted at the prevailing market exchange rate, and foreign currency stocks are converted at the rate prevailing at the end of the relevant period. | Source data Detailed monthly budget reports prepared by the MF are the main data source for compiling GFS. Data sources are adequately detailed and suitable for deriving the published aggregates, but their budgetary scope is rather narrow. Assessment of source data Assessment of source data is done systematically but no computerized checking system is used. Accuracy and reliability are assessed by checking against direct sources, such as major taxpayers for tax revenue. Statistical techniques Apart from simple reclassifications to conform generally to GFSM 1986, statistical techniques are rather limited. Assessment and validation of intermediate data and statistical outputs The main intermediate data are validated against alternative information whenever possible. Revision studies Revisions are taken into account by staff in their work. Provisional data are archived but no explicit revision studies are carried out. | Periodicity and timeliness Both periodicity and timeliness of GFS data meet SDDS requirements. Consistency GFS data are internally consistent. Domestic financing is reconciled with changes in debt. Net foreign financing is consistent between MF and BOP data. The analytically important presentation of domestic financing by type of debt holder is not currently published, thus impeding reconciliation of GFS and monetary data. Revision policy and practice Preliminary annual data are identified, while preliminary monthly data are not. There is a well established revision practice but users are not informed of the revision cycle. | Data accessibility Monthly GFS publications include charts, graphs, and tables, and the main aggregates are adequately detailed. However the presentation, which does not include comments, is not analytically oriented; it mixes two distinct presentations: the derivation of the balancing items of the presentation close to GFSM 1986 is not explicitly described, and the same descriptor is confusingly used for different aggregates. Domestic debt data are presented net of deposits of the government with the banking system. The release of data, which is done simultaneously to all users outside the MF, does not follow a preannounced schedule. Metadata accessibility The MF does not have a website. Methodological notes are not available to support published data, but the IMF’s DSBB includes information (in English) on scope, methodology, classification, data sources, and techniques. Some updating of this information is needed. Assistance to users GFS publications do not list contact names or numbers, or publicize an information desk at the DGRE to help the public. No documents are provided to familiarize interested users with GFS. Some of this information is provided on the DSBB (in English, and with no link from a MF website). Publications are available (free) but only upon request. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Government Finance Statistics
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions The overall structure, concepts, definitions are generally consistent with GFSM 1986. However, Tunisia does not have in place a transition plan to GFSM 2001. Scope The coverage of published monthly and annual data is limited to budgetary central government. Over 90 percent of public administrative units (EPA) such as hospitals and schools, are not consolidated, neither are the social security funds or local governments. However, for the purposes of SDDS, summary annual data on general government (still excluding EPA) are prepared for posting on the DSBB. Classification/sectorization Domestic financing by type of debt holder (bank/nonbank) is not published. Remaining classifications are broadly consistent with GFSM 1986. Basis for recording Government operations are recorded on a cash basis. Gross debt is valued at face value. Foreign currency flows are converted at the prevailing market exchange rate, and foreign currency stocks are converted at the rate prevailing at the end of the relevant period. | Source data Detailed monthly budget reports prepared by the MF are the main data source for compiling GFS. Data sources are adequately detailed and suitable for deriving the published aggregates, but their budgetary scope is rather narrow. Assessment of source data Assessment of source data is done systematically but no computerized checking system is used. Accuracy and reliability are assessed by checking against direct sources, such as major taxpayers for tax revenue. Statistical techniques Apart from simple reclassifications to conform generally to GFSM 1986, statistical techniques are rather limited. Assessment and validation of intermediate data and statistical outputs The main intermediate data are validated against alternative information whenever possible. Revision studies Revisions are taken into account by staff in their work. Provisional data are archived but no explicit revision studies are carried out. | Periodicity and timeliness Both periodicity and timeliness of GFS data meet SDDS requirements. Consistency GFS data are internally consistent. Domestic financing is reconciled with changes in debt. Net foreign financing is consistent between MF and BOP data. The analytically important presentation of domestic financing by type of debt holder is not currently published, thus impeding reconciliation of GFS and monetary data. Revision policy and practice Preliminary annual data are identified, while preliminary monthly data are not. There is a well established revision practice but users are not informed of the revision cycle. | Data accessibility Monthly GFS publications include charts, graphs, and tables, and the main aggregates are adequately detailed. However the presentation, which does not include comments, is not analytically oriented; it mixes two distinct presentations: the derivation of the balancing items of the presentation close to GFSM 1986 is not explicitly described, and the same descriptor is confusingly used for different aggregates. Domestic debt data are presented net of deposits of the government with the banking system. The release of data, which is done simultaneously to all users outside the MF, does not follow a preannounced schedule. Metadata accessibility The MF does not have a website. Methodological notes are not available to support published data, but the IMF’s DSBB includes information (in English) on scope, methodology, classification, data sources, and techniques. Some updating of this information is needed. Assistance to users GFS publications do not list contact names or numbers, or publicize an information desk at the DGRE to help the public. No documents are provided to familiarize interested users with GFS. Some of this information is provided on the DSBB (in English, and with no link from a MF website). Publications are available (free) but only upon request. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Monetary Statistics
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Monetary Statistics
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions. Monetary statistics compiled by BCT are broadly based on the guidelines of the 1984 Guide and there are several areas where they do not conform to the recommendations of the MFSM. BCT has initiated a plan to migrate to the methodology recommended in the MFSM. Scope. The monetary system’s monthly statement is the analytical presentation of the consolidated accounts of most resident other depository corporations (ODCs) and the BCT, covering their headquarters and all resident branches. The monthly statement of the overall financial system, also compiled by the BCT, is similar in scope to the depository corporations survey, which consolidates the accounts of ODCs and the central bank, as defined in the MFSM. Classification/sectorization. The principles used are largely in line with good practices. However, there are some differences from the MFSM. For example, the rest of the domestic sector (other than central government and financial corporations) is not disaggregated into the recommended subsectors; classification of offshore banks is ambiguous. Basis for recording. Valuation of assets is not fully consistent with the MFSM. For example, monetary gold is valued at a fixed price; accrued interest on deposits is not included in the value of these deposits. | Source data. A comprehensive bank register is maintained and used by BCT as a core data source, thus ensuring the complete institutional and geographic coverage of monetary statistics. Timely data reporting is enforced by BCT’s Circular No. 93-08. However, the reported data do not contain sufficient detail for classification of assets and liabilities as recommended in the MFSM. Assessment of source data. Data on BCT are available electronically, while data on ODCs are reported in hard copy. Except for data collection from ODCs, the entire data processing, including data checking, validation, and production, is computerized. Problems with source data are resolved immediately with reporting banks. Statistical techniques. Monetary statistics are derived electronically thus minimizing errors. Estimation is used only to process provisional data for publication, which normally lasts for only a few days and is subsequently replaced by final data. Generally, source data are not adjusted. Assessment and validation of intermediate data and statistical outputs Cross-checks are carried out against data available from other BCT departments. For example, cross-checks are made between the banks’ monthly statements and (i) transactions in the financial markets and (ii) the banks’ accounts with BCT. Revision studies. Provisional data are first published, and then replaced by final data. However, no revision studies are undertaken. | Periodicity and timeliness Tunisia’s monthly monetary statistics meet SDDS requirements and are disseminated according to an advanced release calendar, which is posted on the DSBB. Consistency. Consistent time series are available for an adequate period of time. Changes to data series are explained in the methodological pages of QFSB. Data on foreign assets and liabilities in monetary statistics and in balance of payments data are consistent. However, fiscal data and monetary data on government accounts with banking system are not reconcilable from published data due to lack of breakdown by type of debt holder in the published budget financing data. Data on transactions between the BCT and ODCs, as well as among ODCs, in monetary statistics are not consistent due to classification differences. Revision policy and practice Changes to data are explained in the methodological pages of QFSB. Provisional data are posted on BCT’s website with footnotes to inform users. They are replaced by final data, based on a predetermined schedule. These revision practices are not publicized. | Data accessibility. The monetary statistics are published in BCT’s website and the QFSB, available in Arabic, French, and English. Publication also contains tables of major categories of assets and liabilities with further breakdown by detailed components. BCT’s annual report covers major aggregates of the banking system with commentaries highlighting the important developments. Data are disseminated to all users outside the BCT at the same time based on the advanced release calendar, which is posted on BCT’s website. Metadata accessibility. SDDS metadata and summary methodologies are updated on the DSBB on a regular basis. QFSB contains a section on methodology that provides information on the structure of the financial system, as well as notes on data and changes to data series that are useful to data users. However, information on the methodology used to compile monetary statistics is not available in the QFSB. Assistance to users. Data users can reach the contact person for monetary statistics using the address and e-mail provided on the publication and website, and by direct contact with the BCT. A subscription form containing a list of publications and prices is available. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Monetary Statistics
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions. Monetary statistics compiled by BCT are broadly based on the guidelines of the 1984 Guide and there are several areas where they do not conform to the recommendations of the MFSM. BCT has initiated a plan to migrate to the methodology recommended in the MFSM. Scope. The monetary system’s monthly statement is the analytical presentation of the consolidated accounts of most resident other depository corporations (ODCs) and the BCT, covering their headquarters and all resident branches. The monthly statement of the overall financial system, also compiled by the BCT, is similar in scope to the depository corporations survey, which consolidates the accounts of ODCs and the central bank, as defined in the MFSM. Classification/sectorization. The principles used are largely in line with good practices. However, there are some differences from the MFSM. For example, the rest of the domestic sector (other than central government and financial corporations) is not disaggregated into the recommended subsectors; classification of offshore banks is ambiguous. Basis for recording. Valuation of assets is not fully consistent with the MFSM. For example, monetary gold is valued at a fixed price; accrued interest on deposits is not included in the value of these deposits. | Source data. A comprehensive bank register is maintained and used by BCT as a core data source, thus ensuring the complete institutional and geographic coverage of monetary statistics. Timely data reporting is enforced by BCT’s Circular No. 93-08. However, the reported data do not contain sufficient detail for classification of assets and liabilities as recommended in the MFSM. Assessment of source data. Data on BCT are available electronically, while data on ODCs are reported in hard copy. Except for data collection from ODCs, the entire data processing, including data checking, validation, and production, is computerized. Problems with source data are resolved immediately with reporting banks. Statistical techniques. Monetary statistics are derived electronically thus minimizing errors. Estimation is used only to process provisional data for publication, which normally lasts for only a few days and is subsequently replaced by final data. Generally, source data are not adjusted. Assessment and validation of intermediate data and statistical outputs Cross-checks are carried out against data available from other BCT departments. For example, cross-checks are made between the banks’ monthly statements and (i) transactions in the financial markets and (ii) the banks’ accounts with BCT. Revision studies. Provisional data are first published, and then replaced by final data. However, no revision studies are undertaken. | Periodicity and timeliness Tunisia’s monthly monetary statistics meet SDDS requirements and are disseminated according to an advanced release calendar, which is posted on the DSBB. Consistency. Consistent time series are available for an adequate period of time. Changes to data series are explained in the methodological pages of QFSB. Data on foreign assets and liabilities in monetary statistics and in balance of payments data are consistent. However, fiscal data and monetary data on government accounts with banking system are not reconcilable from published data due to lack of breakdown by type of debt holder in the published budget financing data. Data on transactions between the BCT and ODCs, as well as among ODCs, in monetary statistics are not consistent due to classification differences. Revision policy and practice Changes to data are explained in the methodological pages of QFSB. Provisional data are posted on BCT’s website with footnotes to inform users. They are replaced by final data, based on a predetermined schedule. These revision practices are not publicized. | Data accessibility. The monetary statistics are published in BCT’s website and the QFSB, available in Arabic, French, and English. Publication also contains tables of major categories of assets and liabilities with further breakdown by detailed components. BCT’s annual report covers major aggregates of the banking system with commentaries highlighting the important developments. Data are disseminated to all users outside the BCT at the same time based on the advanced release calendar, which is posted on BCT’s website. Metadata accessibility. SDDS metadata and summary methodologies are updated on the DSBB on a regular basis. QFSB contains a section on methodology that provides information on the structure of the financial system, as well as notes on data and changes to data series that are useful to data users. However, information on the methodology used to compile monetary statistics is not available in the QFSB. Assistance to users. Data users can reach the contact person for monetary statistics using the address and e-mail provided on the publication and website, and by direct contact with the BCT. A subscription form containing a list of publications and prices is available. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Balance of Payments Statistics
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Balance of Payments Statistics
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions Balance of payments statistics are broadly consistent with the methodology of BPM5. The data of the international reserves template are not fully consistent with SDDS requirements. Scope The BCT compiles monthly and quarterly balance of payments statistics that cover in principle all transactions between residents and nonresidents. Classification/sectorization Revenues earned by Tunisian workers abroad are classified under income instead of under current transfers. In the financial account, the sectorization recommended by BPM5 is only applied to the data transmitted to the IMF, not to the data published in Tunisia. Positions/transactions on bonds are recorded under other investment instead of under portfolio investment. Basis for recording Time of recording is broadly in line with BPM5 recommendations as trade in goods is recorded based on Customs data. However, interest is recorded on a cash basis and not on an accrual basis. The data are recorded at market price; they are converted into local currency using the exchange rate of the day of the transaction. Consistency between flows and stocks needs to be improved. | Source data The data collection system is an ITRS, complemented by other sources: Customs data adjusted by the INS, administrative sources, targeted surveys, stock market information (for portfolio investment) and monetary statistics (for banks’ external assets and liabilities and reserve assets). Assessment of source data Against strict foreign exchange regulations, the combination of these various sources results in a good coverage of transactions. Numerous checks and controls further enhance the comprehensiveness of the collection process. Statistical techniques Imports are adjusted from c.i.f. to f.o.b. value using a coefficient reviewed periodically on the basis of invoicing information available to the banks. Assessment and validation of intermediate data and statistical outputs The data collected are subject to systematic plausibility checks and respondents are frequently requested to provide additional information to ensure the accuracy of the balance of payments. Revision studies Quarterly data are provisional when first released. Final data are published in the Annual Report of the BCT and in the publication La Balance des Paiements de la Tunisie. No explicit study is made on revisions, but information on revisions is used in carrying out the statistical work. | Periodicity and timeliness Periodicity and timeliness are in line with the SDDS requirements. Consistency Balance of payments data are consistent with the rest-of- the-world account of the national accounts. BOP/IIP and monetary statistics are also consistent, as are BOP/IIP and external debt data. There is also consistency between budget financing from GFS and balance of payments external public borrowing data. Revision policy and practice Changes in methodology are publicized in the methodology section in the BCT’s QFSB. Revision practice is to update monthly data of the current year on a continuous basis, and to freeze the data when the year is complete until the publication of the Annual Report and then again until publication of the annual brochure on the BOP. Revision practice is not publicized. | Data accessibility The balance of payments and IIP statistics are published on the website of the BCT and in the QFSB, which is also posted on the website. The BCT also publishes an annual analysis of Tunisia’s balance of payments, showing in particular summary bilateral balance of payments with partner countries and a currency composition of the settlements recorded in the ITRS. Metadata accessibility Documentation on the methodology and sources used in compiling the balance of payments is available upon request in French at the BCT via the Documentation and Public Relations Department. Existence of metadata on the IMF’s DSBB is not mentioned on the BCT’s website. Assistance to users Information on the contact person for BOP/IIP statistics is available on the DSBB. This is also the case for the other categories of the SDDS (external debt, exchange rates). A generic e-mail address is indicated on the BCT’s home page that can be used for requesting additional information. |
Tunisia: Assessment of Data Quality—Dimensions 2 to 5—Balance of Payments Statistics
| 2. Methodological soundness | 3. Accuracy and reliability | 4. Serviceability | 5. Accessibility |
|---|---|---|---|
| Concepts and definitions Balance of payments statistics are broadly consistent with the methodology of BPM5. The data of the international reserves template are not fully consistent with SDDS requirements. Scope The BCT compiles monthly and quarterly balance of payments statistics that cover in principle all transactions between residents and nonresidents. Classification/sectorization Revenues earned by Tunisian workers abroad are classified under income instead of under current transfers. In the financial account, the sectorization recommended by BPM5 is only applied to the data transmitted to the IMF, not to the data published in Tunisia. Positions/transactions on bonds are recorded under other investment instead of under portfolio investment. Basis for recording Time of recording is broadly in line with BPM5 recommendations as trade in goods is recorded based on Customs data. However, interest is recorded on a cash basis and not on an accrual basis. The data are recorded at market price; they are converted into local currency using the exchange rate of the day of the transaction. Consistency between flows and stocks needs to be improved. | Source data The data collection system is an ITRS, complemented by other sources: Customs data adjusted by the INS, administrative sources, targeted surveys, stock market information (for portfolio investment) and monetary statistics (for banks’ external assets and liabilities and reserve assets). Assessment of source data Against strict foreign exchange regulations, the combination of these various sources results in a good coverage of transactions. Numerous checks and controls further enhance the comprehensiveness of the collection process. Statistical techniques Imports are adjusted from c.i.f. to f.o.b. value using a coefficient reviewed periodically on the basis of invoicing information available to the banks. Assessment and validation of intermediate data and statistical outputs The data collected are subject to systematic plausibility checks and respondents are frequently requested to provide additional information to ensure the accuracy of the balance of payments. Revision studies Quarterly data are provisional when first released. Final data are published in the Annual Report of the BCT and in the publication La Balance des Paiements de la Tunisie. No explicit study is made on revisions, but information on revisions is used in carrying out the statistical work. | Periodicity and timeliness Periodicity and timeliness are in line with the SDDS requirements. Consistency Balance of payments data are consistent with the rest-of- the-world account of the national accounts. BOP/IIP and monetary statistics are also consistent, as are BOP/IIP and external debt data. There is also consistency between budget financing from GFS and balance of payments external public borrowing data. Revision policy and practice Changes in methodology are publicized in the methodology section in the BCT’s QFSB. Revision practice is to update monthly data of the current year on a continuous basis, and to freeze the data when the year is complete until the publication of the Annual Report and then again until publication of the annual brochure on the BOP. Revision practice is not publicized. | Data accessibility The balance of payments and IIP statistics are published on the website of the BCT and in the QFSB, which is also posted on the website. The BCT also publishes an annual analysis of Tunisia’s balance of payments, showing in particular summary bilateral balance of payments with partner countries and a currency composition of the settlements recorded in the ITRS. Metadata accessibility Documentation on the methodology and sources used in compiling the balance of payments is available upon request in French at the BCT via the Documentation and Public Relations Department. Existence of metadata on the IMF’s DSBB is not mentioned on the BCT’s website. Assistance to users Information on the contact person for BOP/IIP statistics is available on the DSBB. This is also the case for the other categories of the SDDS (external debt, exchange rates). A generic e-mail address is indicated on the BCT’s home page that can be used for requesting additional information. |
14. To complement the Fund staff’s assessment of the quality of macroeconomic statistics produced in Tunisia, an informal survey of key users of such statistics was conducted. The authorities sent questionnaires to 14 users; responses were received from 12 of them. Respondents included international and domestic banks, international and regional organizations, the financial media, think tanks, employer and producer associations, the trade unions, and some users of data in the public administration. Users were asked to evaluate the coverage, periodicity, timeliness, dissemination practices, accessibility, and overall quality of the official statistics assessed in this report.
15. Users overall expressed satisfaction with the quality of the statistics produced in Tunisia. The results of the survey indicated that a majority of users are satisfied with the coverage, periodicity, level of detail, methodological soundness, and timeliness of official statistics. On a five point scale (1 meaning poor and 5 excellent), respondents’ average rating for the overall quality of the official statistics for all assessed sectors was a 3.8, with the lowest score (3.5) given to GFS and the highest (4.1) to monetary statistics. Most users regarded Tunisia’s official macroeconomic statistics as being of better quality than the statistics of countries at similar levels of development in the region.
16. The main uses of official statistics are, in order of importance, for general economic background, analysis of current developments, analysis of trends, comparisons with developments in other countries, and economic research. The macroeconomic statistics most frequently used are the monetary and balance of payments statistics, closely followed by national accounts statistics. GFS are used less frequently. The main source of official statistics is, according to most users, official publications and press releases. However, similar numbers of users indicated they rely on publications of international organizations about Tunisia. Over 40 percent of users do not consult official descriptions of the sources and methods used to compile the official statistics.
17. Several comments and suggestions for improving official statistics were provided in the responses to the questionnaires. These concern mostly the national accounts (diversification of dissemination means for quarterly national accounts; dissemination of data with greater level of detail, including information on the informal sector). Some respondents also commented that the timeliness of some statistics (national accounts, GFS, and balance of payments) could be improved.
III. Staff’s Recommendations
18. Based on the review of Tunisia’s statistical practices, discussions with the data producing agencies, and responses from data users (see Appendix III of the Detailed Assessments volume), the mission has a set of recommendations. They are designed to increase further Tunisia’s adherence to internationally accepted statistical practices and would, in the mission’s view, enhance the analytical usefulness of Tunisia’s statistics. Some additional technical suggestions are included in the Detailed Assessments volume.
Cross-cutting Recommendations
Pursue rigorously plans to further improve the quality of the country’s official statistics by increasing the alignment of the current practices with internationally accepted standards and good practices.
Promote in all data-producing agencies a culture of dissemination oriented toward the public at large.
Dissemination of information on the conditions of statistical production, as well as of data, metadata, and contact information should be made fully and consistently for all assessed areas of statistics and through all available media.
Provide more staff resources to statistical production in all the areas assessed in this report. This is particularly important in the area of GFS.
Ensure that the adoption of international standards is translated into changes of domestically available data, beyond the data made available through the IMF’s DSBB.
Undertake revision studies and disseminate them as well as the revision policies and practices.
National Accounts
High Priority
Complete work on the new series of annual accounts (starting with 1997), which is in line with the 1993 SNA and introduces a new base year, and disseminate it shortly. Revise backward the annual and quarterly account series to ensure consistency as far back as possible.
Develop new infra-annual source data or improve existing ones that are particularly needed for quarterly accounts (e.g., agriculture, construction, and certain services); make use of all available administrative data (e.g., balance sheets submitted at the MF) as information becomes transmissible.
Prepare and disseminate “source and methods” documents that would include detailed methodological notes on the new series of annual and quarterly accounts; update these documents regularly.
Other
Provide a link from the INS website to the metadata posted on the DSBB.
Undertake basic analyses of data to accompany the dissemination of data.
Government Finance Statistics
High priority
Adopt a migration plan to GFSM 2001. Start implementation of the plan by producing, concurrently with the current GFSM 1986-based GFS data, the same cash data based on the GFSM 2001 classification.
Expand the coverage of government in published annual data to include all EPA, social security funds, and local governments. Ensure that the legal basis for the relevant reporting is in place.
Adapt fully the current presentation of the monthly/annual data to the GFSM 1986 framework by performing the appropriate changes in descriptors and classifications.
Publish a breakdown of domestic financing by type of debt holder (BCT, banking sector excluding BCT, and non-banking sector).
Improve the accessibility of GFS data and methodology: identify all primary data sources in the main GFS tables, prepare methodological notes on GFS compilation, and include their summaries in monthly publications (and the upcoming website). Identify contact persons and their telephone, fax, and email in all publications and in the upcoming website.
Other
Accelerate the introduction of the MF’s own website; post on it all available detailed GFS data and expanded coverage data as they become available; provide a hyperlink to the DSBB.
The MF should engage in dialogue with current and prospective GFS users through its participation in the CNS.
Streamline the presentation of tables in the monthly publication, with those following the GFSM 1986 at the beginning of the publication, and those following the Organic Law presentation at the end.
Implement changes to the accounting system to move to double-entry accounting and identify and collect accrual data for the relevant main aggregates.
Monetary Statistics
High priority
Develop and implement a timetable of actions to adopt the recommendations of the MFSM in the following areas: (i) sectorization of the economy; (ii) classification of financial assets and liabilities (e.g., classification of domestic claims); (iii) valuation of financial assets and liabilities.
Improve the report forms and the accompanying instructions (Circular 93–08) used to collect data for the compilation of monetary statistics with emphasis on the detailed breakdowns as recommended in the MFSM.
Compile and disseminate the central bank survey, the other depository corporations survey, and the depository corporations survey according to the framework recommended in the MFSM.
Review and improve the classification of data on interbank transactions in the compilation of monetary statistics to ensure consistency in these data.
Balance of Payments Statistics
High priority
Reclassify workers’ remittances from income to current transfers.
Review the contents of the international reserves template to make it compliant with the SDDS’s Guidelines for a data template.
Move to the standard presentation of BPM5 for national publications of balance of payments and international investment position statistics and use the appropriate breakdowns in the financial account for instruments and sectors.
Implement an annual collection of data on stocks of foreign direct investment and record reinvested earnings in the balance of payments.
Post on the DSBB the summary methodology for exchange rates.
Ensure consistency between flows and stocks for short-term trade credits.
Other
Study the possibility of recording interest on an accrual basis.
Review the recording of technical assistants’ remittances and of pensions.
Provide a link to the DSBB website on the BCT’s website page dedicated to external statistics.
Add reference in publications to the legal framework within which the BCT compiles and disseminates balance of payments, international investment position, and external debt statistics.
APPENDIX I
Tunisia: Practices Compared to the SDDS Coverage, Periodicity, and Timeliness of Data
Does not cover foreign currency deposits of resident banks with the BCT under “pre-determined short-term net drains on foreign currency assets,” but covers debt service of public companies, without it being flagged.
Given that the data are broadly disseminated by private means, the timeliness with which official data are disseminated is not time critical.
Tunisia: Practices Compared to the SDDS Coverage, Periodicity, and Timeliness of Data
| SDDS Data Category | Coverage (Meets SDDS) | Periodicity | Timeliness | Comments | ||
|---|---|---|---|---|---|---|
| SDDS | Tunisia | SDDS | Tunisia | |||
| Real Sector | ||||||
| National accounts | Yes | Q | Q | 1Q | 1Q (4M for the first quarter) | Use flexibility option |
| Production index | Yes | M | M | 6W (1M encouraged) | 6w | |
| Labor market: Employment (Atvinna) | Yes | Q | A | 1Q | 7M | Use Periodicity and timeliness flexibility. Missing SM page for UEM |
| Labor market: Unemployment (Registered unemployment) | Yes | Q | A | 1Q | 7M | |
| Labor market: Wages/earnings | Yes | Q | Q | 1Q | 3M | Missing SM page. |
| Price index: Consumer prices | Yes | M | M | 1M | 15D | |
| Price index: Producer prices | Yes | M | M | 1M | 1M | |
| Fiscal Sector | ||||||
| General government or public sector operations | Yes | A | A | 2Q | 6M | |
| Central government operations (revenue and expenditure, capital transactions, borrowing, and overall cash balance) | Yes | M | M | 1M | 1M | |
| Central government debt Yes | Yes | Q | Q | 1Q | 2M | |
| Financial Sector | ||||||
| Analytical accounts of the banking sector | Yes | M | M | 1M | 1M | |
| Analytical accounts of the central bank | Yes | M (W recommended) | M | 2W (1W encouraged) | 2W | |
| Interest rates | Yes | D | D (all for tenders rate and the 7-day allowance uptake rate) | D | Missing SM page | |
| Stock market: Share price index | Yes | D | D | D | Missing SM page | |
| External Sector | ||||||
| Balance of payments | Yes | Q | Q | 1Q | NLT 2M | |
| International reserves and foreign currency liquidity | Partly 1/ | M (W recommended) | M | 1W for official reserve assets 1M (1W encouraged) for the reserves template | 7D for international reserves and the reserves template data | |
| Merchandise trade | Yes | M | M | 8W (4-6W encouraged) | NLT 20D | |
| International investment position | Yes | A (Q recommended) | A | 3Q (1Q encouraged) | 6M | |
| External debt | Yes | Q | Q | 1Q | 3M | |
| Exchange rates | Yes | D | D | 2/ | D | Missing SM page. |
| Socio-demographic data | ||||||
| Population | Yes | A | A | … | 4M | Missing SM page. |
Does not cover foreign currency deposits of resident banks with the BCT under “pre-determined short-term net drains on foreign currency assets,” but covers debt service of public companies, without it being flagged.
Given that the data are broadly disseminated by private means, the timeliness with which official data are disseminated is not time critical.
Tunisia: Practices Compared to the SDDS Coverage, Periodicity, and Timeliness of Data
| SDDS Data Category | Coverage (Meets SDDS) | Periodicity | Timeliness | Comments | ||
|---|---|---|---|---|---|---|
| SDDS | Tunisia | SDDS | Tunisia | |||
| Real Sector | ||||||
| National accounts | Yes | Q | Q | 1Q | 1Q (4M for the first quarter) | Use flexibility option |
| Production index | Yes | M | M | 6W (1M encouraged) | 6w | |
| Labor market: Employment (Atvinna) | Yes | Q | A | 1Q | 7M | Use Periodicity and timeliness flexibility. Missing SM page for UEM |
| Labor market: Unemployment (Registered unemployment) | Yes | Q | A | 1Q | 7M | |
| Labor market: Wages/earnings | Yes | Q | Q | 1Q | 3M | Missing SM page. |
| Price index: Consumer prices | Yes | M | M | 1M | 15D | |
| Price index: Producer prices | Yes | M | M | 1M | 1M | |
| Fiscal Sector | ||||||
| General government or public sector operations | Yes | A | A | 2Q | 6M | |
| Central government operations (revenue and expenditure, capital transactions, borrowing, and overall cash balance) | Yes | M | M | 1M | 1M | |
| Central government debt Yes | Yes | Q | Q | 1Q | 2M | |
| Financial Sector | ||||||
| Analytical accounts of the banking sector | Yes | M | M | 1M | 1M | |
| Analytical accounts of the central bank | Yes | M (W recommended) | M | 2W (1W encouraged) | 2W | |
| Interest rates | Yes | D | D (all for tenders rate and the 7-day allowance uptake rate) | D | Missing SM page | |
| Stock market: Share price index | Yes | D | D | D | Missing SM page | |
| External Sector | ||||||
| Balance of payments | Yes | Q | Q | 1Q | NLT 2M | |
| International reserves and foreign currency liquidity | Partly 1/ | M (W recommended) | M | 1W for official reserve assets 1M (1W encouraged) for the reserves template | 7D for international reserves and the reserves template data | |
| Merchandise trade | Yes | M | M | 8W (4-6W encouraged) | NLT 20D | |
| International investment position | Yes | A (Q recommended) | A | 3Q (1Q encouraged) | 6M | |
| External debt | Yes | Q | Q | 1Q | 3M | |
| Exchange rates | Yes | D | D | 2/ | D | Missing SM page. |
| Socio-demographic data | ||||||
| Population | Yes | A | A | … | 4M | Missing SM page. |
Does not cover foreign currency deposits of resident banks with the BCT under “pre-determined short-term net drains on foreign currency assets,” but covers debt service of public companies, without it being flagged.
Given that the data are broadly disseminated by private means, the timeliness with which official data are disseminated is not time critical.