0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

Law 13/PR/99 of June 15, 1999 regulates statistical activity in Chad, defining the institutional framework and the broad principles governing the work of the agencies producing official statistics. This framework, however, has been implemented only partially, and the expectations of change it has elicited are materializing only very slowly.

According to Article 2 of Law 13/PR/99, the National Statistical System (SSN, for Système statistique national) comprises the Higher Council for Statistics (CSS, for Conseil supérieur de la statistique), the Statistics Program Committee (CPS, for Comité des programmes statistiques), the National Institute of Statistics and Economic and Demographic Studies (INSEED, for Institut national de la statistique, des études économiques et démographiques),¹ and the statistics offices and units in ministries, in the central bank, and in various other government agencies. The CSS and the CPS are the central pillars of the system (idem, Article 3).

According to Article 6 of the law, the CSS should provide guidance and issue directives on the production and dissemination of official statistics, with due regard to international standards. It should approve the national statistics program prepared by the CPS. It is also supposed to see that the components of the NSS have the necessary human, technical, material, and financial resources.

The CPS should coordinate the production of official statistics within the national statistics program and also oversee the execution of the decisions of the CSS. The national statistics program comprises a long-term program and an annual program of activities (idem, Article 8), covering periodic, as well as ad hoc, operations (idem, Article 9).

Article 9 of the law provides no details on multiyear projects, confining itself to stating the objectives and the approximate implementation dates: it thus constitutes an expression of the country’s statistics strategy rather than a precise work plan. The annual program of statistical activities defines the approaches for executing selected operations (idem, Article 10).

Article 12 requires the central statistics body, INSEED, to coordinate the SSN and to serve as the secretariat of the CSS.

Article 13 provides that statistical work must observe the fundamental principles of official statistics declared on April 14, 1994 by the Statistical Commission of the United Nations Economic and Social Council, including those relating to impartiality and transparency.

Decree 417/PR/MPED/2000 sets out the responsibilities, composition, and operations of the CSS and the CPS, confirming the CSS as the designer of national statistics policy. The decree (Chapter 2, Article 2) requires the CSS (chaired by the Ministry of Economic Development) to provide a framework document every four to six years defining medium-term objectives for the multiyear statistics program. The decree confirms the multiyear program as a forward planning tool, for which the CSS is to decide the main operations and the approximate dates for carrying them out. The CSS also establishes the main targets for the coming two years and provides an opinion on emergency operations not included in the annual program.

To date, however, neither the CPS nor the CSS has ever met, and consequently these bodies are not fulfilling the functions assigned to them. The INSEED (whose director is by law the secretary of the CSS and the president of the CPS) must therefore try to fulfill these roles and is generally recognized as the lead authority. In this capacity, the director has organized a series of coordination meetings and seminars that have exchanged views on concepts, classifications, and methods, focusing particularly on the national accounts. The most recent initiative involves coordinating the three-year statistics programs: however, the responsible managers have presented programs for only 11 of the 21 fields of activity for which INSEED has requested a program.

INSEED is responsible for preparing the national accounts (Law 13/PR/99, Article 12). Other agencies, however, also use, and sometimes disseminate, estimates of the major aggregates, which may differ from the figures produced by INSEED. Thus, INSEED has frequently had to intervene with those agencies to correct the situation. It may, in fact, have contributed to the situation because of the architecture of its estimating model. (The INSEED tends to compile successive versions of national accounts aggregates as additional information becomes available, but it does not always control their dissemination.)²

0.1.2 Data sharing and coordination among data-producing agencies are adequate

Although the CSS is still not in place, arrangements for data sharing and coordination among statistics-producing agencies are slowly taking shape.

The country’s statistical system is decentralized, heightening the importance of arrangements between INSEED, the Bank of Central African States (BEAC, for Banque des États de l’Afrique centrale), and the Ministry of Economy and Finance (MEF), among others.

INSEED is responsible for synthesizing information and, in particular, for preparing the national accounts. However, it does not generally produce the source data it needs. Thus, its coordination with other official statistics and data-producing agencies is crucial to its work. One of its objectives in its three-year plan is to coordinate the statistical activities of the ministries. In this regard, it has taken the initiative to organize coordination meetings (see 0.1.1) and introductory seminars on the national accounts for its data compilers and ministerial offices. Clearly, it has made progress in furthering the understanding of the purpose of the national accounts, their concepts, classifications, and, hence, the INSEED’s needs regarding source data.

Coordination has been particularly difficult when it comes to the BEAC because it provides only limited data on businesses’ activities, collected in the course of its quarterly survey. Moreover, its foreign trade figures sometimes differ widely from those of INSEED. However, because INSEED is a member of the Balance of Payments Committee (CBP, for Comité de balance des paiements), it has an active say about the foreign trade figures presented there.

Finally, a benefit exists in the membership of the MEF, the BEAC, and INSEED in the Macroeconomic Framework Committee (CCM, for Comité de cadrage macroéconomique), established in May 2004 by MEF Decree 167/MEF/SG/DGB/DEP/04. The CCM is responsible for “collecting, centralizing, processing, and interpreting the macroeconomic data in order to make short-term forecasts and to prepare economic budgets.”

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

Article 17 of Law 13/PR/99 specifies the statistical secrecy with which agencies must treat compulsory surveys. No exceptions are to be allowed when dealing with data relating to private matters concerning personal and family life. In the case of economic and financial data, the law notes that agencies may not always find it possible to respect the principle of nonidentification (even indirectly) of individual units. The law provides, however, that interested parties must be advised of such situations and that tables including such data may not be released without the approval of the CSS. Finally, the law stipulates that in no case may these data be used for purposes of fiscal or economic controls. Article 20 of the law provides for penalties against agents or individuals who violate the statistical secrecy rules, although the nature of those penalties is not specified.

In practice, for some industries like cotton, water, and electricity distribution, oil production, beer, sugar, etc., individual data are in fact disclosed by INSEED in its quarterly economic bulletins. In these cases the interested parties are not specifically advised, and the table including the data is released without prior approval of the CSS, because, to date, the latter has never been convened.

The only device in place for protecting the confidentiality of information collected for the national accounts is a password installed on staff computers.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

Article 14 of Law 13/PR/99 establishes the reporting obligation of natural and legal persons and the level of fines for noncompliance. Nevertheless, the penalty procedures, which the CSS was supposed to establish, have never been put in place. Some businesses simply refuse to report information, and no measures, taken against them to date, have been effective.

The law also refers to public service administrations and to legal persons under private law charged with managing a public service, and it sets the corresponding penalties.

Since no data surveys exist, the question of the response burden implied by surveys does not arise.

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

Although the financial situation at INSEED has improved somewhat since the May 2000 visit by the IMF multisectoral statistics mission, the funding earmarked for statistical programs still falls far short of needs. It lacks the assured continuity required for long-term programs. INSEED therefore has to rely heavily on external financing.

Regarding staffing, senior INSEED officers all embody recognized competence in the field of economic statistics and are graduates of the region’s statistics institutes. INSEED pays real attention to staff training: National accounts training sessions have been held, and two officers were sent for practical training to Cameroon to learn good practices in the field.

Since 2002, more staff members have been devoted to the national accounts. The responsible division now has four persons: two economists/statisticians and two technical staff trained at institutes in Abidjan and Yaoundé. They have attended regular training seminars in recent years, financed by Afristat, French cooperation, or the African Development Bank. These seminars have helped keep their skills up to date.

Nonetheless, how long the staff will remain in their present positions is unclear. In fact, three out of four officers were paid from the budget of the World Bank’s capacity building project on “Management of the Petroleum Economy” (GEEP, for Gestion de l’économie à l’ère pétrolière), which has just been wrapped up. A plan exists to convert these officers into INSEED’s permanent staff, but the pay schedule offered by INSEED is not competitive, even within the public sector (including in comparison with academic pay rates). INSEED has prepared draft personnel bylaws, including a pay schedule, but the agency’s Board of Directors has not yet considered them.

Computing resources seem adequate, and each officer has a computer. Those staff who use the existing software consider it adequate for their current tasks, but the hardware may prove to be underdimensioned if new dissemination tasks are to be taken on. (For example, there is no software for converting pdf files.) INSEED also faces Internet access problems: The only two direct points of access are located in the offices of the Director General and the Afristat representative. While all staff members are allowed to use these work stations as needed, the arrangements make it difficult for them to participate actively in sharing methodological information as well as data that are constantly accumulating on the network.

The revised versions of the national accounts series are backed up every six months on a CD and stored in a safe place.

INSEED’s office facilities have been upgraded with funds from the GEEP program. While space is still cramped, working conditions have substantially improved, especially with the installation of air conditioning.

An effort is underway to compile and centralize documentation that was scattered among various government offices, in libraries, and even in cardboard boxes, but this project is not yet completed. INSEED is assembling a documentation center for employees and for public use. The Agence française de développement (the French development agency) has financed the necessary materials (cabinets and shelving), and two people have received special training in document management. A reception desk has been set up for users, and requested information is supplied if available.

A multiyear program for 2002–07 was adopted when the statistics system was reformed, with help from the United Nations Development Program. Although the program has been partially implemented, the CPS, which is to monitor it, is not yet in place. INSEED has meanwhile adopted a 2005–07 statistics program of its own, based on this initial plan, but has yet to submit it to the Board of Directors. INSEED has also taken steps to coordinate the statistics programs of other government agencies, but coordination with the BEAC still presents serious problems. Moreover, too many changeovers have occurred in the Director General position at INSEED, and progress has been further slowed by that position’s extended vacancy. A Department of Coordination and Dissemination has been created—responsible for planning and programming the agency’s activities—but the lack of human resources means that it is not yet really operational.

INSEED’s financial resources are not commensurate with the ambitious intentions of Law 13/PR/99. Article 21 confirms a statistics user fee among other sources of funding for the statistics system, but it does not further specify the sources or proportions of such funding, or the agencies to which they are to be allocated. Actually, a statistics user fee was instituted as early as 1965, and its rate was raised in 1996. A portion of its revenue was supposed to be returned to INSEED, but this has never happened.

Consequently, INSEED continues to operate as a directorate of its supervisory ministry, as far as the payment of salaries and current operating costs is concerned. Although it does receive a substantial annual grant from the government budget, the grant bears no relationship to the statistics fee income. Moreover, the grant is paid only in part and is glaringly inadequate for carrying out program activities. This has led INSEED to seek funding from donors, especially for major statistical operations, training, and modernization of the national accounts: More than 80 percent of the funding for large-scale statistical operations comes from external sources.

Finally, problems persist in coordinating the use of this funding with the scheduling of statistical work in the field. For instance, during collection of data for the second survey of the household consumption and informal sector of Chad (ECOSIT-2), disbursements were delayed. As a result, the survey was not fully completed; staff found it impossible to conduct all the field operations as planned initially.

0.2.2 Measures to ensure efficient use of resources are implemented

The cost of the three-year work program has been worked out in detail, but the greatest attention is, in fact, paid to donor-financed programs. To improve program monitoring, a grading system has been proposed to evaluate the degree of implementation of each of these activities, but this system is not yet enforced.

No procedure exists for evaluating staff work. The draft personnel bylaws of INSEED (to be submitted for approval by the Board of Directors) call for a performance assessment system as part of the staff promotion and categorization policy.

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

In the past, no real monitoring covered the relevance and usefulness of statistics in terms of user needs. Recently, however, INSEED has begun to systematically track information requests from its visitors, for identification and input into future programs.

0.4.1 Processes are in place to focus on quality

Article 13 of Law 13/PR/99 stipulates that the SSN must observe the basic principles of official statistics adopted by the Statistical Commission of the United Nations Economic and Social Council.

Basic training for statistics officers reflects this general concern for quality. INSEED has held seminars on national accounts concepts and measures, for the benefit of users, information providers, and persons engaged in the processing of primary data.

0.4.2 Processes are in place to monitor the quality of the statistical program

INSEED has prepared a three-year work plan for 2005–07, specifying for each program detailed objectives, activities, performance indicators, time frames, the responsible manager, the expected results, and the source of funding. This work plan has not been discussed either by the INSEED Board of Directors or by the CPS.

A number of international missions have visited Chad to help the country improve its capacity to produce national accounts. Three recent IMF missions have visited—one in 2000 (multisector statistics) and two in 2002 (preparation of General Data Dissemination System [GDDS] metadata and national accounts). An Afristat mission in 2000 examined the country’s readiness to receive a mission to implement the ERETES software module.³

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

INSEED’s work plan pays great attention to quality considerations. In particular, a special program maintains and strengthens the quality of existing personnel, promotes training, and improves the understanding of data sources. Another program is in place to change the base of the national accounts and to adopt the ERETES system.

The country’s system for compiling national accounts works both toward and against quality components. The computerized simplified system used for compiling national accounts for Chad includes convenient functionality for estimating automatically missing source data, while it also poses problems for introducing data from new data sources or data organized along formats unusual to the system. At the same time, the compilation of national accounts is part of the process for setting the macroeconomic framework, which calls for strict deadlines. These two considerations together tend to cause the awareness of meeting compilation deadlines to take too much precedence over other components of quality, in particular the accuracy and reliability of the estimation method.

1.1.1 Statistics are produced on an impartial basis

The INSEED Board of Directors consists of general directors from various ministries, the National Director of the BEAC, the Director of the University of N’Djamena, and two INSEED staff representatives It is chaired by the Director General of the Ministry of Economic Development. The board nominates a candidate as Director of INSEED, who is then appointed by decree of the Council of Ministers. No particular conditions are attached to this post, relating either to the term of the mandate or to the professional requirements for the job.

Professional independence seems to be respected in compiling the national accounts. INSEED staff who prepare the retrospective accounts say they have never been pressured by the government or by other domestic economic agents, with respect, for example, to the relationship between the results in the historical national accounts and the previous projected versions. This may reflect, however, not so much a culture of noninterference as a lack of interest. Any pressure is likely to come from international donors, who may have their own opinion on the economy’s performance and are ready to make it prevail.

As part of INSEED’s program to strengthen the dissemination of better-quality data, all technical publications must now be approved by a scientific review committee of senior officials from the department concerned. The formal evaluation committee for the national accounts has yet to be established. Thus, for the moment they are validated by the Director General alone: The data cannot be released without his consent.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

Given the shortage of human resources and the nature of the calculation system now in use, the data sources for the model have remained mostly the same. The difficulty of amending the national accounts procedures and software (as demonstrated by the technical problems in integrating oil industry information) makes it all the harder to upgrade data sources, to expand the number of respondents, or to develop new data sources.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

Chad has no public or media discussion of economic information and, in fact, no commentary at all outside the CCM. Thus, for the moment, no erroneous interpretation by the general public arises in a public forum. INSEED in fact has a program to encourage the press, radio, and television to give greater public prominence to statistics.

1.2.1 The terms and conditions under which statistics are collected, processed, and disseminated are available to the public

The terms and conditions under which statistics are collected, processed, and disseminated are not disclosed in the publications.

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

The access that government agents have to the data before their release is nowhere indicated. Local users, however, must generally be aware that these data are discussed in advance by the CCM, since it is through dissemination of that framework that the historical national accounts data are first released.

1.2.3 Products of statistical agencies/units are clearly identified as such

INSEED makes it a general policy to identify clearly its own products and those from other sources. However, its publications do not mention the user’s obligation to cite the sources when using the material.

1.2.4 Advance notice is given of major changes in methodology, source data, and statistical techniques

No advance notice is given of major changes affecting the data. In the September 2004 publication of the national accounts, a section briefly noted the sectors for which the methodology was updated in light of the April 2003 macroeconomic framework exercise. However, no details were provided about the impact or scope of the changes.

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

Apart from Law 13/PR/99, no INSEED document deals explicitly with ethical issues. The draft personnel bylaws merely refer, on this point, to Law 13/PR/99 and in particular to the chapter on statistical secrecy. To date, no regular reminders have been issued about the ethical rules that must govern staff activities.

It would be appropriate to issue systematic reminders of these rules, especially for temporary staff hired to collect and process source data.

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

The national accounts of Chad are compiled using the simplified national accounts system introduced in 1995 with United Nations Development Program support. This system works as a model that incorporates observed past data, current data, and equations for estimating missing variables. It also has equations for compiling historical data, as well as future estimates in the form of national accounts forecasts. Although this simplified system may be more or less germane to the 1993 SNA, the departures between both systems are not documented. The INSEED has no formal plan to migrate Chad’s national accounts to fully comply with the 1993 SNA.

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

Some of the 1993 SNA tables and accounts (listed below) are compiled regularly (at first quarterly and now semiannually):

• annual gross domestic product (GDP) and value added (calculated for 21 industries but published for 15 industries only) at current and constant 1995 prices;

• annual expenditures of GDP at current and constant 1995 prices;

• simplified consolidated accounts for the total economy (up to financial accounts); and

• annual rest-of-the-world accounts (up to net lending).

Annual value-added components at current prices are not shown—neither overall nor by activity. The annual accounts for the rest of the world, derived from the balance of payments (BOP), are not fully integrated with the national accounts (import-export differences). The consolidated accounts for the total economy are very simplified and show only the link of the basic national aggregates with the accounts for the rest of the world.

Although Chad does not prepare quarterly accounts, it recalculates its historical and projected annual accounts every six months. In the past, revisions to the historical accounts could affect the entire series back to 1997. Since publication of the 1995–2003 national accounts series in September 2004, all the accounts for 2002 and earlier are treated as final, and this approach will be consistently applied in the future: Thus, the figures for year n-2 and before are deemed final.

The question of the scope of economic activities included within the economic territory does not arise explicitly in compiling the national accounts. Instead, scope is determined by the treatment in the BOP—the data from which (apart from merchandise trade) are integrated, generally without modification, into the national accounts.

Some branches of foreign firms operating in Chad (except for those in the oil industry) are not reflected in the evaluations of domestic output (this is especially true of the local agencies of foreign airline companies).

In the absence of detailed source data, it is not clear whether the production boundary is consistent with the 1993 SNA. Own-account production of all goods for own final consumption is included in output, since the estimate is made by extrapolating from the results of the ECOSIT for 1995/1996 (ECOSIT-1) where this item is included. However, some problems appear with grain milling, which is consistently treated in all cases as a market activity (pertaining to manufacture), is valued at market prices, and is thus included in the market output even if performed for own account. On the contrary, grain milling should be considered as an industrial service on own account and thus be excluded from the boundary of production. Fetching water is also valued in all cases and treated as a good, which is consistent with the 1993 SNA.

The economic measure of the oil industry specifically includes exploration—the output of which is considered the production of an intangible asset—as well as the own-account capital formation related to these activities.

Agricultural work in progress is taken into account in the changes in inventories, since the crop year extends in most cases from October to April, thus covering portions of two calendar years. Most of the expenditure occurs at the beginning of the crop year. Production is therefore deemed to occur at that time, while growing crops are part of the stocks carried forward from one year to the next.

With respect to defense-related assets, the investment figures supplied by government are broken down between gross fixed capital formation (GFCF), compensation of employees, and intermediate consumption. This is on the basis of available data for the base period (fixed allocation keys). Therefore, it is difficult to tell what principle is implicitly adopted, even if it was correctly applied to the base period.

The primary sources used do not identify other intangible assets such as systems and standard-application computer software and databases (purchased or built in-house), entertainment, literary or artistic originals, patents, leases, and other transferable contracts (such as purchased goodwill). But given the country’s low level of development, these are unlikely to represent significant amounts.

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

The scope of the simplified system is limited to the production accounts and the consolidated accounts of the nation. The system does not include institutional sector accounts or even the allocation of primary income accounts. Thus, no institutional sector classification is found in Chad’s national accounts.

Nevertheless, a classification should emerge from the preparatory work for introducing a new national accounts base. At a workshop held in Libreville in May 2001, the BEAC launched work on updating the sectorization of public-sector entities applicable to the monetary statistics of the Central African Economic and Monetary Community (CEMAC, for Communauté économique et monétaire de l’Afrique centrale) countries. This initiative should serve as a starting point for a complete sectorization of the economy, prepared in cooperation among the INSEED, BEAC, and MEF, and this would apply to the national accounts.

The transactions classification used in the system generally follows the recommendations of the 1993 SNA, even if some of these transactions are treated in a simplified manner.

Nevertheless, the simplified system uses its own classification system of products and activities, with an automatic industry-product bridge (there being no secondary production). This is an extremely simplified and rigid system, designed to make use of the sources that were available at the time the base was established, and it cannot be readily stretched to take account of new products and new activities. This classification currently includes 21 products and industries, after inclusion of three oil activities and related products. Imported goods are not classified by product.

A project is near completion to validate the new classifications prepared on the basis of Afristat’s nomenclature of activities (Activity Classification of Afristat Member States [NAEMA]) and products (Product Classification of Afristat Member States [NOPEMA]). They will be used in the new national accounts base.

Expenditures by government are simply presented as an overall figure and are not examined by function; thus, there is no classification of the functions of government.

2.4.1 Market prices are used to value flows and stocks

Market output is valued at producer prices.

Output for own use is valued at equivalent market prices. This is the rule used to evaluate output and consumption in the 1995/1996 ECOSIT-1 survey, which serves as the basis of extrapolation for households’ own-use production.

Producer prices for agricultural production are estimated on the basis of consumer prices observed for the measurement of the producer price index and an assumption on the margin rate.

Intermediate consumption is valued with all taxes included (TTC, for toutes taxes comprises).

Transfer prices must exist, in particular for CotonTchad—the cotton company that embraces a whole series of activities starting with raw cotton (ginning, oil, and soap manufacture). However, corrections do not seem to be made, nor is it possible to do so, because no benchmark market exists for intermediate products.

Major export products are tracked specifically. However, imports are valued only globally, using BOP figures. This is a consequence of the problems encountered in gathering foreign trade data by product (there being no detailed customs data since 1997). A global adjustment is made to transform free on board (f.o.b.) import data from the BOP into cost, insurance, freight (c.i.f.) data from the goods and services balance table. The system would also seem to make other adjustments, because differences not only persist but they increased considerably in 2003.

2.4.2 Recording is done on an accrual basis

In theory, the rules used for recording flows accord with the 1993 SNA. Transactions and flows are recorded on an accrual basis.

While government expenditures are recorded on a commitment basis (relatively close to the accrual base of recording), revenues are recorded on a cash basis.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

In light of the economic structure of Chad, where nearly all enterprises consist of only one establishment, the issue of grossing or netting procedures is not relevant.

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

When the simplified system of national accounts was introduced in 1995, staff determined and still determine the content of the expected results on the basis of the types of information that could be collected at that time. The availability of information has not evolved much since then and is even more meager; this is the case, for example, with the statistics and tax declarations (DSFs, for Déclarations statistiques et fiscales). A few changes have taken place in the structure of the data collected: (1) Firms no longer in business have been deleted, (2) adjustments to individual industries have been applied, and (3) oil exploration and exploitation activities are now taken into account.

Overall, Chad suffers from a chronic scarcity of source data for national accounts. The source data received on a regular basis relate to agriculture, livestock, government finance, BOP, certain information on enterprises in the modern sector, imports of oil products, oil output, and crude oil export prices.

The rigidity of the simplified national accounts system makes it difficult, without a thorough overhaul of its internal calculation process, to take account of changes in products, activities, and information sources. For example, serious problems persist with the inclusion of mobile telephony and other fast-growing services as well as the products and activities relating to oil exploration and development. Data for most commercially traded services, tracked in part through a quarterly BEAC survey, are not integrated into the system but are estimated using an indirect procedure (global consumption per capita) on the basis of the ECOSIT survey.

Data collection is far from adequate for compiling the national accounts statistics, given the importance of services and the informal sector. This was in fact the rationale for introducing the simplified national accounts system. But little progress has been made since 1997 in mobilizing new sources, because of the lack of human and financial resources. This is particularly true in the case of customs data on foreign trade, which are not always available. Moreover, the solution offered by the existing program has tended to reduce pressure to find new data.

In agriculture, the statistics division of the Ministry of Agriculture had to abandon its survey, which covered about 1,600 farms, for lack of funds. Staff now prepare agriculture data on the basis of an oral field survey that tends to record the respondents’ opinion rather than objective information.

For livestock production, the staff collect data from veterinary stations. They estimate herd sizes on the basis of a 1976 census, applying hypothetical rates of growth and slaughter to update these variables and to evaluate production and consumption.

With a specially constructed questionnaire, the quarterly BEAC survey systematically tracks only about 140 enterprises in the modern sector. INSEED uses for its accounts some information relating to about 10 industrial enterprises that have monopoly status in their respective fields: in particular, sugar, cotton, beverages, tobacco, soaps, and oils. This information covers turnover, employment, the payroll, and physical output. (These industries generally produce only one product.)

There is no business register. A plan exists to create such a register, based on data collected by the Internal Revenue Department (DGI) and the private-sector National Social Provident Fund (CNPS). However, the plan has made little progress because of human and financial constraints. To improve national accounts estimates, a project has been suggested to conduct a census of producing units: This census could serve as a framework for the register. To date, a questionnaire has been prepared, but the project has not been launched, for lack of financing.

Income and expenditure surveys of households and the informal sector have been conducted but at irregular intervals. In 1970, a household and expenditure survey covered N’Djamena. The 1995 ECOSIT-1 covered the four main cities. ECOSIT-2, of nationwide coverage, was conducted between February 2003 and March 2004, and its results are now being processed. However, financial and technical problems hindered that survey, and information from ECOSIT-2 is still unavailable. That survey covered three issues: (1) households (social and demographic characteristics, housing, education, health, employment. etc.); (2) budget, with the emphasis on collecting household expenditure data; and (3) the informal sector.

The survey covered “all ordinary settled African households where no member works in an international agency or a diplomatic establishment.” The stratified sample originally contained 7,008 households, of which 1,056 were in N’Djamena, 854 were in other cities, and 5,088 were rural. The coverage of rural areas was new for Chad because previous surveys had been limited to N’Djamena (1970) or to the major cities (1995). The survey was initially constructed on the basis of the 1993 census, but field counts were conducted in selected survey areas to upgrade the sample and establish the probability of household selection. The survey was conducted with a fixed sample, that is, each household had to be interviewed in each round.

Data collection was organized in three rounds, and four months were supposed to elapse between rounds to take account of seasonal fluctuations. However, various circumstances have interfered: To begin with, travel is difficult during the rainy season, and the second round had to be postponed in some areas. Then various administrative problems occurred (late funding, delays in the printing survey forms), suspending data collection in some areas before all households were visited the second time.

The questionnaire had been tested in the field, but, at more than 70 pages, it was found to be excessively detailed.

INSEED maintains close working relationships with the Budget General Directorate (DGB), the MEF’s Studies and Forecasts Directorate (DEP), and the Treasury General Directorate (DGT)—the directors of which are members of the CCM. Annual government finance data are in this way transmitted to the national accounts team as they become available. The national accounts staff use the operating budget to identify the nature of the expenditure. They break down capital budget, using keys to bridge from the detailed structure of the base year. Local governments seem to be covered only in part (city of N’Djamena only), but in fact the financial autonomy of other local governments is apparently very limited.

As for other government agencies, the data only seem to take into account those agencies responsible for social security for public-sector employees—the National Pension Fund of Chad (CNRT, for Caisse nationale de retraite tchadienne)—and private-sector employees—the National Social Provident Fund (CNPS, for Caisse nationale de prévoyance sociale). The data do not cover other entities (such as funds or other incorporated or unincorporated units with some degree of autonomy) that might perhaps be included in this sector.

Data collection from administrative sources is inadequate. If it were better managed, it would certainly help make up the lack of information on formal sector activities. Wider data collection should be possible, but today, in any case, INSEED does not have the technical capacity to handle this information.

INSEED is supposed to receive a copy of the DSFs submitted to the DGI, but that requirement has never been enforced. As part of modernizing its national accounts, INSEED plans to make use of DSFs starting in April 2006, in collaboration with the DGI, for 2005 data. Discussions to this effect were opened with the DGI in May 2005.

The Government Flow of Funds Table (TOFE, for Tableau des opérations financières de l’État), in its current form, is inadequate for compiling the government sector accounts.

INSEED could no doubt use data from the social security funds in its new national accounts base.

Price statistics now in use are inadequate. With respect to final consumption goods, primarily the consumer prices are used, as observed in the process of compiling the consumer price index. These are usually transformed, using a fixed margin, into producer prices.

For industrial products of the modern sector, producer prices are available. For livestock products, prices are those observed on producer markets in the vicinity of production and are consumer prices on markets in the vicinity of consumption. For export products, which number very few (livestock, cotton, gum Arabic, and crude oil,) the various prices are known. Current resource limitations make it impossible to undertake supplementary surveys.

INSEED held meetings with the ministries’ statistics offices in 2003 to identify any developments that should be considered in the system for compiling the national accounts statistics. Such meetings however have not been repeated.

Occasional but irregular meetings are held to improve data sources. The most recent one, which included all the units involved in producing foreign trade statistics, took place in November 2004, but it does not seem to have produced any results.

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

With the current acute shortage of information, the staff responsible for preparing the national accounts tries to make maximum use of any source data that might support compilation or confirm the data already used. Questions about their reliability thus arise only later.

The coverage of total economic activities (in terms of value added) by source data obtained through recognized statistical procedures is very limited. The exception is oil extraction, where production is known but where the elements of value added, as well as related activities, are hardly measured. This reflects, in large part, the fact that most economic activities in Chad are conducted informally and are tracked only sporadically during specific statistical operations (the ECOSIT surveys of 1995 and 2003/2004).

The format of some of the data obtained indirectly from administrative documents is however not always consistent with national accounts needs. For example, data on the oil industry are very difficult to interpret for national accounts purposes.

In some cases, despite agreement on the basic information, discrepancies persist about how to treat the information: For example, INSEED considers the herd growth rates used by the Ministry of Livestock (ME, for Ministère de l’élevage) to be outdated, and it makes its own estimates.

3.1.3 Source data are timely

The estimation procedure implicit in the national accounts system masks the lack of source data. That is, it facilitates, for better or worse, accommodation to the lack of information, since the model makes it possible to estimate data that are not at hand. The work is thus done on an estimates basis, with the inherent risk that information that might exist but that might not be collected will not be introduced into the system.

Monitoring by INSEED, given its organizational shortcomings and its lack of materials (vehicles and fuel) and human resources, does not seem sufficiently rigorous to ensure the timely receipt of data, for example through reminders and follow-up visits to obtain missing information from respondents.

Actually, no real program promotes systematic data collection for compiling and semiannually updating the national accounts in accordance with a specific schedule. When establishing the macroeconomic framework and medium-term forecasts, the entire team (including the DEP) proceeds urgently to collect the information needed to update the database for preparing the historical and projected accounts.

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

There is no process for systematic prior revision of source data. The data are generally incorporated into the model’s spreadsheets as soon as they become available, because the historical accounts are not really compiled by stages and aggregation of intermediate results.

Staff do, however, systematically modify some source data during the process of preparing the accounts. This is the case, in particular, with the slaughter rates for livestock (since those used by the ME are considered too high), exports of livestock (where the data received are subject to systematic adjustments), and cotton exports at certain periods.

For the first version of the historical accounts, staff gradually replace when possible, by observed figures, some data from the system that were estimated in the forecasted version. They then run the program to obtain a new version of the accounts. They critically review that version, and if it seems inconsistent with the expected overall outcomes, they review the individual data incorporated in it.

The establishment of the new base for the national accounts will provide an opportunity to reconsider the calculation methods and the sources used.

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

The data are not systematically adjusted despite major swings in agricultural output. (The output may at times seem erratic because of the significant climatic fluctuations that Chad suffers from one year to another.) Adjustments to source data are thus made only when aberrations are detected. In this case, the techniques used are simple, because the estimation model almost always offers a substitute based on historical trends or on similarities of behavior observed in the past.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

Table 1:

Production Approach

Note: The base year for the rates is 2002, to exclude the oil boom, which has completely disrupted the structure of the national economy.

Table 1:

Production Approach

Industry	Weight in 2002 GDP	Main data sources
Agriculture, livestock, forestry, and fisheries	34 percent	Subsistence agriculture: physical quantities derived from land area and yields from oral farm surveys + consumption standards + producer or market prices.
		Non-monitored products: ECOSIT-1 (1995) estimates x population growth rate.
		Cash crops (cotton, sugar, gum Arabic): extension services, processing businesses, consumption standards, technical coefficients. Intermediate consumptions: by consumption ratios per hectare of crop.
		Livestock: various data from the livestock ministry, but INSEED’s own growth rates.
Oil extraction and oil processing industry	4 percent	Forestry, fisheries, hunting: consumption and export standards. Direct information on quantities extracted and exported
Manufacturing and crafts	9 percent	Agri-food: by branch, input quantities available.
		Modern industrial activities: some products are identified and processed using information provided by the businesses concerned. Remainder: in the absence of DSFs, a distribution key that has been unchanged since 1995.
Electricity, gas, and water	1 percent	DSFs from enterprises in each industry.
Construction and public works	1.5 percent	For GFCF:
		Household dwellings: ECOSIT-1 (1995).
		Public: Government capital accumulation account.
		Private: production industry related activities.
		Oil: direct sources.
		Capital repairs: percentage of productive activities in GDP.
Wholesale and retail trade	22 percent	Nonagricultural products: estimate of the portion of market supply to which an assumed trade margin is applied.
		Agricultural products: balance: resource = production + trade margin + stocks.
Transportation and communications	3 percent	Passenger road travel: ECOSIT-1 (1995).
		Merchandise road transport: transport imputation rate; tax rate.
		Services: ASECNA [African air safety agency].
		Air transport: activities reports.
Financial and insurance corporations	-------------------	Accounting documents from the establishments concerned
Actual and imputed rent	Overall, 9 percent	Average rents from ECOSIT projected and applied to a stock of housing
Other commercially marketed services		Personal services: ECOSIT estimates and projections by population.
	-------------------	Services to businesses: technical relationships with industries that consume them + exports (BOP)
General government	11 percent	Using information provided by government departments and agencies
Import taxes and duties that of subsidies	4 percent	Documents provided by Customs General Directorate

Note: The base year for the rates is 2002, to exclude the oil boom, which has completely disrupted the structure of the national economy.

View Table

Table 1:

Production Approach

Industry	Weight in 2002 GDP	Main data sources
Agriculture, livestock, forestry, and fisheries	34 percent	Subsistence agriculture: physical quantities derived from land area and yields from oral farm surveys + consumption standards + producer or market prices.
		Non-monitored products: ECOSIT-1 (1995) estimates x population growth rate.
		Cash crops (cotton, sugar, gum Arabic): extension services, processing businesses, consumption standards, technical coefficients. Intermediate consumptions: by consumption ratios per hectare of crop.
		Livestock: various data from the livestock ministry, but INSEED’s own growth rates.
Oil extraction and oil processing industry	4 percent	Forestry, fisheries, hunting: consumption and export standards. Direct information on quantities extracted and exported
Manufacturing and crafts	9 percent	Agri-food: by branch, input quantities available.
		Modern industrial activities: some products are identified and processed using information provided by the businesses concerned. Remainder: in the absence of DSFs, a distribution key that has been unchanged since 1995.
Electricity, gas, and water	1 percent	DSFs from enterprises in each industry.
Construction and public works	1.5 percent	For GFCF:
		Household dwellings: ECOSIT-1 (1995).
		Public: Government capital accumulation account.
		Private: production industry related activities.
		Oil: direct sources.
		Capital repairs: percentage of productive activities in GDP.
Wholesale and retail trade	22 percent	Nonagricultural products: estimate of the portion of market supply to which an assumed trade margin is applied.
		Agricultural products: balance: resource = production + trade margin + stocks.
Transportation and communications	3 percent	Passenger road travel: ECOSIT-1 (1995).
		Merchandise road transport: transport imputation rate; tax rate.
		Services: ASECNA [African air safety agency].
		Air transport: activities reports.
Financial and insurance corporations	-------------------	Accounting documents from the establishments concerned
Actual and imputed rent	Overall, 9 percent	Average rents from ECOSIT projected and applied to a stock of housing
Other commercially marketed services		Personal services: ECOSIT estimates and projections by population.
	-------------------	Services to businesses: technical relationships with industries that consume them + exports (BOP)
General government	11 percent	Using information provided by government departments and agencies
Import taxes and duties that of subsidies	4 percent	Documents provided by Customs General Directorate

Note: The base year for the rates is 2002, to exclude the oil boom, which has completely disrupted the structure of the national economy.

Apart from production and exports, information on the oil sector is inadequate and does not meet national accounting needs. In 2005, it represented more than 38 percent of GDP, after only 4 percent in 2002.

The accounts rely heavily on indirect estimates. The modern sector, apart from the oil industry, is very poorly represented. While significant amendments have been made since 1995, the accounts need an appropriate observation system.

Intermediate consumption estimates are based in all cases on fixed ratios. Fixed ratios, indeed, are one of the most common techniques for estimating the national accounts of Chad, but it is very difficult to determine their extent. Even in agriculture, where direct information is available on the most important products, ratio-based estimates account for a significant portion of calculations.

The following techniques are used to evaluate specific issues and categories:

• Output of owner-occupied dwellings is valued as the estimated rentals that tenants would pay for similar accommodation. This information is taken from the ECOSIT-1.

• Work in progress is deemed to include growing crops and livestock reared for consumption purposes, as well as fishery stocks. Inventory valuation does not reflect price fluctuations between the periods of entering and exiting the stocks. Quantity stocks are valued first, and the average price for the period is then applied to these quantities. Thus, no adjustment is made for holding gains/losses accruing on inventories; these appear to a certain extent in the trading margin because of the calculation method used.

• No calculation is done for fixed capital consumption; all valuations are gross values. The calculations of nontradable output of government and of own-account capital formation are thus undervalued.

• Cash-based data for general government are not converted to an accrual basis.

• The single indicator method is most widely used to calculate GDP volume indicators. The same product volume index is applied to the industry’s production (equivalent to output of the product), to its intermediate consumption, and to its value added.

• Volume measures of taxes and subsidies on products are estimated by extrapolating the tax or the subsidy from the base year using a volume indicator for transactions subject to a particular tax or subsidy. Yet, given the level of aggregation of the accounts, this principle is applied at an overly aggregated level.

• Output volumes of trade margin are estimated by extrapolating the base-year trade margins. The average volume of sales are used but at an aggregated level, because there are no truly systematic product balances.

• GDP volume changes are measured by applying base-year price indices (1995) to quantities for the current period (all products and industries are treated as elementary products). There is no chaining of indices.

Table 2:

Expenditure Approach

Table 2:

Expenditure Approach

Source of demand	Weight in 2002 GDP	Main data sources
Household final consumption	74 percent	Residual estimate
Government final consumption	15 percent	Direct estimation from information supplied by the MEF
Gross fixed capital formation	57 percent	Direct estimation by product and industry
Changes in inventories	Nonsignificant	Relates exclusively to livestock and growing crops
Imports and exports of goods and services	Imports: 63 percent	Merchandise: adjusted data from the BOP
		- direct data by product for exports
		- direct data for imports in the oil industry
	Exports:	- adjustment of cross-border trade: data from the BOP
	16 percent
		Services: data from the BOP.

View Table

Table 2:

Expenditure Approach

Source of demand	Weight in 2002 GDP	Main data sources
Household final consumption	74 percent	Residual estimate
Government final consumption	15 percent	Direct estimation from information supplied by the MEF
Gross fixed capital formation	57 percent	Direct estimation by product and industry
Changes in inventories	Nonsignificant	Relates exclusively to livestock and growing crops
Imports and exports of goods and services	Imports: 63 percent	Merchandise: adjusted data from the BOP
		- direct data by product for exports
		- direct data for imports in the oil industry
	Exports:	- adjustment of cross-border trade: data from the BOP
	16 percent
		Services: data from the BOP.

There are no changes in inventories apart from those relating to livestock raised for food purposes and to standing crops. There is no buffering of the impact of the production shock on consumption.

Household final consumption expenditure is determined globally as a residual. Detailed elements of this consumption are also calculated, however, to estimate output of products of national origin (agri-food products for which there is no direct information, craft products), when this is estimated from demand.

Government final consumption expenditure is compiled globally from information on current expenditure, recognizing that this is estimated from direct production costs. Consumption of fixed capital is not calculated.

Gross fixed capital formation is compiled using several criteria: herd size by type, housing, government GFCF, and other values for the oil industry, capital repairs, and others.

For domestic consumption, ratios more than five years old are frequently used—certainly in more than 20 percent of cases.

3.4.1 Intermediate results are validated against other information where applicable

Little complementary information is available to validate the data. GDP is estimated from the production approach, and global final consumption is calculated as a residual. Because of the lack of foreign trade data, no proper supply and use tables exist to combine the imported and domestically produced products. The system treats these as totally complementary.

The lack of customs data (the latest are for 1997) means that imported products are not disaggregated.

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

Staff conduct checks to detect any discrepancies, in particular, in administrative data sources. They analyze ratios of intermediate consumption to output, structure of GDP, etc. and eliminate discrepancies.

Nevertheless, neither household actual final consumption, nor any of its components, are systematically analyzed. The total value of actual final consumption at constant prices does not seem to have been viewed as an interesting aggregate; it does not even appear separately in the publications. INSEED does not provide explication about the fact that sharp fluctuations exist in annual volume changes for an aggregate in which major components are estimated by trend.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

The process of compiling the accounts involves consultation and comparisons between the MEF, IMF, and BEAC, to prepare the submissions to the CCM.

Analysts have assessed the import and export discrepancies with the BOP statistics for 2001 and prior years. They found that the discrepancies resulted essentially from divergent figures on exports of livestock and differences in the method of recording imports (f.o.b. for the BOP; c.i.f. for the national accounts). Since 2002, these differences have reached considerable proportions (more than 25 percent of GDP). The differences would seem to be linked to the way the oil industry is treated, but no comment on them has appeared in INSEED’s publication Comptes et agrégats de la Nation (National Accounts and Aggregates), in which the two series are presented.

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

Given their number, it would seem that the successive estimations of the historical accounts, when they are finalized, are not subject to any regular analysis by INSEED. This might be explained by the lack of continuity in the team in charge of the project over recent years, by the complexities of the model used, and by the many different versions available.

4.1.1 Periodicity follows dissemination standards

The staff compiles GDP estimates annually. It revises them semiannually during the macroeconomic framework-setting exercise, generally in April and October of each year. The first provisional estimates for a given year are thus generally available in May of the following year.

4.1.2 Timeliness follows dissemination standards

INSEED distributes the estimated data for the national accounts to members of the CCM and to international agencies at the committee’s meetings in April and October. The data are also available to the general public as of May and November (but only on request), following their approval by the director of INSEED.⁴ The data relate to the provisional accounts for the last calendar year and those for previous years, which are generally reviewed on this occasion. This distribution also includes a forecast for the current and following years. Although the level of detail communicated is not known, it is probably consistent with the detail in which the macroeconomic program itself is reported, because these figures seem to be the point of departure for that program.

In September 2004, INSEED released 200 copies of Comptes et agrégats de la Nation, covering the period 1995–2004 (data for 2004 are forecasts). Plans are to publish this document systematically at the end of each year, meaning the provisional data for any year would be released in October–November of the following year. The publication would still cover the historical series since 1995. However, publication following the April meeting would be preferable, insofar as it could then include an estimate for the immediately preceding year, in its provisional version.

4.2.1 Statistics are consistent within the dataset

The calculation used in the simplified accounting system guarantees internal consistency of the national accounts statistics, both at current prices and at constant 1995 prices. No statistical discrepancy exists between the two GDP estimates, because the total final private consumption is calculated as a residual.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

The calculation method relies on time series dating back to 1995. Also, consistent time series cover the period 1983–95, compiled by extrapolation at the time the base year 1995 was established.

When new observation data come in, the system normally recalculates the entire series back to 1995. New data may relate to any year in the series. Some technical experts think certain adjustments should affect the entire time series. (Given the system’s complexity, this view is not shared by everyone.) The national accounts statistics would then always have to be considered semi-definitive, because they would always be subject to changes, even for periods long ago. Thus, the INSEED has decided to freeze all results prior to 2002. That is, the agency will make no further corrections to data before that time and will apply this procedure systematically in the future. Consequently, in the first historical estimation of the accounts (n) in April (n+1), the accounts for (n-2) and previous years will be considered as definitive and may no longer be revised. It will be important to verify that this indeed is true, henceforth, because INSEED’s credibility demands that it guarantee users that its time series will be stable, which currently is not the case.

The following table shows the fluctuations in the published national accounts series:

Table 3:

Comparison of GDP Volume Growth Rates by Sources

Note: Only the data from National Accounts and Aggregates 1995–2004 and those from the INSEED website are consistent over the entire period. Figures from the Macroeconomic Outlook for the second half of 2004, published later than the National Accounts and Aggregates 1995–2004, are not consistent with the announced freeze of estimates prior to 2002. Figures released by BEAC do not correspond to any of the other series.

Table 3:

Comparison of GDP Volume Growth Rates by Sources

In percent	1998	1999	2000	2001	2002	2003
Macroeconomic framework February	5.9	-0.5	-0.7	9.6	10.5	9.2
2004, INSEED
Macroeconomic outlook 2/2 2004,	7.0	-0.8	-0.6	10.4	8.4	12.5
INSEED
National accounts and aggregates	7.0	-0.8	-0.6	9.6	10.1	10.5
1995/2004, September 2004, INSEED
INSEED website	7.0	-0.8	-0.6	9.6	10.1	10.5
BEAC website	6.0	-0.6		9.4	9.7	11.9

Note: Only the data from National Accounts and Aggregates 1995–2004 and those from the INSEED website are consistent over the entire period. Figures from the Macroeconomic Outlook for the second half of 2004, published later than the National Accounts and Aggregates 1995–2004, are not consistent with the announced freeze of estimates prior to 2002. Figures released by BEAC do not correspond to any of the other series.

View Table

Table 3:

Comparison of GDP Volume Growth Rates by Sources

In percent	1998	1999	2000	2001	2002	2003
Macroeconomic framework February	5.9	-0.5	-0.7	9.6	10.5	9.2
2004, INSEED
Macroeconomic outlook 2/2 2004,	7.0	-0.8	-0.6	10.4	8.4	12.5
INSEED
National accounts and aggregates	7.0	-0.8	-0.6	9.6	10.1	10.5
1995/2004, September 2004, INSEED
INSEED website	7.0	-0.8	-0.6	9.6	10.1	10.5
BEAC website	6.0	-0.6		9.4	9.7	11.9

Note: Only the data from National Accounts and Aggregates 1995–2004 and those from the INSEED website are consistent over the entire period. Figures from the Macroeconomic Outlook for the second half of 2004, published later than the National Accounts and Aggregates 1995–2004, are not consistent with the announced freeze of estimates prior to 2002. Figures released by BEAC do not correspond to any of the other series.

A very brief note in the September 2004 publication mentions some important changes added since previous versions. For each sector, the periods involved are indicated, but neither the impact nor the scope of the changes is specified.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

Compilers of the national accounts use BOP statistics from the BEAC and government finance figures from the MEF. However, the statistical indicators produced by these sources are not always consistent with those in the national accounts. Thus, total imports and exports of goods are not always the same in the national accounts and in the BOP, particularly because of the impact of imports related to the oil industry. These statistics, however, are reproduced without change in the National Accounts and Aggregates 1995–2004, with mention of their source but without any bridge tables or any explanation for the differences. The situation is especially serious for the BOP, because the consolidated national accounts show in this case an adjustment (without any commentary) equal to about 28 percent of GDP for 2003.

4.3.1 Revisions follow a regular and transparent schedule

Solidly established, the revision cycle follows the rhythm of meetings of the CCM. The cycle was quarterly until 2002 and has been semiannual since then. The ministries, the monetary and financial authorities, and the major donors are generally aware of this even if they are not specifically informed. The same cannot necessarily be said for the general public.

4.3.2 Preliminary and/or revised data are clearly identified

With the first two tables in National Accounts and Aggregates 1995–2004 showing the main economic aggregates and key indicators, the heading indicates whether the data are revised (1995–2000), provisional (2001), estimates (2002, 2003), or forecasts (2004). However, no defined meaning is presented for these terms in the case of the historical accounts. Also, this heading does not appear in the rest of the publication.

4.3.3 Studies and analyses of revisions are made public (see also 3.5.1)

Although INSEED has been publishing revised series systematically since 1997, it does not indicate any analysis of the scope of revisions. It mentions only the nature of the revisions and the economic sectors concerned.

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

For lack of resources, dissemination of the national accounts in a specific publication was suspended in 1997 but was resumed in September 2004. Although it is called a semiannual publication, it is in fact intended to be released only annually.

This publication provides, first, the main macroeconomic aggregates and key indicators, followed by commentaries on the analysis of economic growth for the period 1995–2001, growth in 2002 and 2003, and the outlook for 2004. This is followed by the statistical tables and annexes.

INSEED provides very little in the way of national accounts data on its website; GDP is shown only as a global aggregate at current and constant prices.

5.1.2 Dissemination media and format are adequate

The September 2004 publication was printed in 200 copies: 120 copies were distributed within government and to donors. The other 80 copies were made available to the public, and stocks were exhausted in May 2005. Data are very aggregated, general comments are provided about the economic performance of the economy, and little reference is made to the data presented in the tables. Although this represents an important improvement over the previous situation, neither the extent of this dissemination, nor the content of the publication itself, is really up to the standards.

Before and since that date, some of the most important data have been released every six months through the publication dealing with the macroeconomic outcome (Macroeconomic Framework, and now the Macroeconomic Outlook). Nonetheless, the data were released only after their approval by the INSEED Director and only as a supplement to the macroeconomic framework itself.

INSEED plans to post more data from the national accounts on its website (currently, only total GDP at current and constant prices is shown). It plans also to disseminate those data electronically to users on request. A small budget exists for maintaining the website, but the staff who are supposed to maintain the website have not received the necessary training.

Users can obtain more detailed information on request.

5.1.3 Statistics are released on a preannounced schedule

The national accounts data are disseminated on a limited scale with the macroeconomic framework, i.e., twice a year, in May and November. The data constitute the base of the framework. Thus, the publication calendar is theoretically that of the CCM, and users are aware of this fact. But the deadline may not be respected if the CCM meets late or if the INSEED Director reserves his decision to publish them (as is currently the case). No official schedule has been made public, however.

5.1.4 Statistics are made available to all users at the same time

Given the current process of calculating and disseminating the national accounts data together with the economic forecasts, CCM members receive those data before other users, who are not explicitly informed of this fact.

5.1.5 Statistics not routinely disseminated are made available upon request

It seems that users interested in doing so can gain access to more detailed data by applying to the INSEED publications office. This possibility is not publicly announced.

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

The only metadata available today to the public are those prepared in English for the GDDS and posted at the IMF’s DSBB. Moreover, they have not been updated since they were posted in July 2002. Their availability is not indicated in any of the dissemination channels used in Chad. Further, these metadata are not sufficiently detailed and do not provide any information on the departures between the simplified system of national accounts applied to Chad and the 1993 SNA.

5.2.2 Levels of detail are adapted to the needs of the intended audience

As explained in paragraph 5.2.1, only one version in English of the metadata is made available to the public; thus, its content is not adapted to the specific needs of various audiences. Even the French version of the metadata, made available by the IMF to the authorities, is not posted at any website or published in any document.

5.3.1 Contact points for each subject field are publicized

All statistical publications provide coordinates (e-mail, telephone, fax, and mailing address) for contacting INSEED and its Director General for further information. Specific contact points for each subject field are not publicized.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

There is no formal catalog of publications; there is merely a listing of the publications available and their price.

Table 4:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for National Accounts

(Compiling Agency: National Institute of Statistics and Economic and Demographic Studies – INSEED)

Table 4:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for National Accounts

(Compiling Agency: National Institute of Statistics and Economic and Demographic Studies – INSEED)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element		NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
			O	LO	LNO	NO
0. Prerequisites of quality
	0.1 Legal and institutional environment			X			The current legislative framework is not fully operational. Interinstitutional cooperation is not adequate. When individual data are disseminated-directly or indirectly-because of the small number of operators in some economic branches, interested respondents are not notified of such situation.	Short-term (ST). Have the CSS and the CPS meet. Have the Board of Directors meet and decide on the personnel bylaws and the work plan. Notify officially respondents when their individual data will inevitably be –directly or indirectly– disseminated.
	0.2 Resources				X		INSEED’s financial independence is not guaranteed. It does not offer attractive pay rates. INSEED is sometimes slow in making use of its funds.	Medium-term (MT). Guarantee permanent financing for INSEED’s long-term programs.
	0.3 Relevance			X			The relevance and serviceability of the national accounts data suffer from gaps in the source data.	MT. Improve the monitoring of users’ needs and build these into the new system.
	0.4 Other quality management			X			The process of compiling the accounts focuses on timeliness, to the detriment of other aspects of quality.	MT. Take advantage of the change of base year and system to incorporate all existing data sources.
1. Assurances of integrity
	1.1 Professionalism			X			Technical criteria are not sufficiently weighted for selection of the Director General.	MT. Reinforce the technical nature of INSEED.
	1.2 Transparency				X		Publications do not mention the conditions governing data collection, processing, and dissemination. Access to statistics by CCM members prior to their release to the public is not publicized. No advance notice is given of major changes in methodology, source data, and statistical techniques.	ST. Mention in INSEED publications the conditions governing data collection, processing, and dissemination. Publicized early access of CCM members to statistics. Give advance notice of any major change in methodology, source data, or statistical techniques.
	1.3 Ethical standards			X			INSEED has no specific ethical standards applicable to its staff. Standards applicable to civil servants in general are not systematically indicated to staff, particularly to short-term contractuals.	ST. Adopt ethical standards specific to INSEED and make sure all staff, in particular temporary staff, are aware of these standards.
2. Methodological soundness
	2.1 Concepts and definitions				X		Simplified system basically restricted to production and goods and services accounts. The departures between the simplified system applied to Chad’s national accounts and 1993 SNA are not documented. There is no plan to migrate to the 1993 SNA.	ST. Prepare documentation on the departures between the simplified system and 1993 SNA. Design and adopt a migration plan to the 1993 SNA.
	2.2 Scope			X			The sequence of accounts recommended in the 1993 SNA is not completely compiled. Some foreign-based businesses operating in Chad are not included.	MT. Move ahead with introduction of the ERETES system, which produces more complete tables. ST. Include the missing activities of foreign-based businesses operating in Chad.
	2.3 Classification/sectorization				X		No institutional sectorization exists. Classifications of products and activities are specific to Chad.	MT. Adopt an institutional sectorization in common with all other participants in the statistics system. Adopt the comprehensive Afristat classification of products and activities.
	2.4 Basis for recording			X			Market prices are used, with a few exceptions. Source data on government finance are not on an accrual basis, but on a cash basis.	MT. Implement ERETES
3. Accuracy and reliability
	3.1 Source data				X		The compilation of ECOSIT-2 survey is not completed. No detailed data are available on the informal sector or on marketable services, except for the base year. Detailed data on foreign trade are unavailable since 1997. The Agriculture survey is unavailable since 2001.	ST. Complete the compilation of ECOSIT-2 survey. MT. Improve the use made of administrative data. Produce and use detailed statistics on foreign trade. Repeat the agricultural survey.
	3.2 Assessment of source data				X		There is no systematic assessment of source data.	ST. Assess critically and systematically the data used and perform cross-checks.
	3.3 Statistical techniques				X		There is heavy reliance on fixed ratios and on indirect estimates. Unique indicator for volume of output and of value added. All products and activities are considered as elemental.	ST. Compare ECOSIT 1 and ECOSIT 2 to assess the validity of assumptions about fixed ratios, and reevaluate them if necessary.
	3.4 Assessment and validation of intermediate data and statistical outputs				X		Opportunities for assessing intermediate data are overlooked.	ST. Conduct a systematic review of household consumption by product.
	3.5 Revision studies				X		Data are reviewed twice a year, but no systematic studies of changes are conducted between successive versions.	ST. Monitor successive versions of the accounts and explain any differences to users.
4. Serviceability
	4.1 Periodicity and timeliness			X			The data are available on an annual basis, but no precise schedule promotes their dissemination after the meeting of the Macroeconomic Framework Committee.	ST. Establish a strict release calendar and adhere to it.
	4.2 Consistency			X			Foreign trade data since 2002 are sharply inconsistent with those of the BOP.	ST. Review foreign trade data and analyze the reasons for discrepancies with the BOP. Develop consensus on estimates for the BOP and the national accounts.
	4.3 Revision policy and practice			X			Revisions are mentioned in publications, but their amount is not quantified.	ST. Establish a classification that will clearly designate the successive degrees of revision and adhere to it strictly.
5. Accessibility
	5.1 Data accessibility				X		There was no hardcopy publication between 1997 and 2004. No preannounced schedule is available. The data offered at the website are overly aggregated.	ST. Revise and adhere to the calendar for hardcopy publication. Revise the publication of the accounts, in particular the contents of the tables. Expand dissemination at the INSEED website. Confine to a minimum the revisions to the historical series.
	5.2 Metadata accessibility				X		The metadata in English available at the IMF’s DSBB are not periodically updated and detailed enough. No metadata in French are made available to the public.	ST. Update GDDS metadata posted in English at the DSBB and provide more details herein. Make metadata in French available to the public. Make a practice of adding methodological notes to publications.
	5.3 Assistance to users			X			Only the Director General is indicated as the contact person for all statistics compiled by the INSEED. There is no catalogue of publications.	ST. Indicate specific contact person(s) for national accounts statistics. Establish and disseminate a catalog of publications.

View Table

Table 4:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for National Accounts

(Compiling Agency: National Institute of Statistics and Economic and Demographic Studies – INSEED)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element		NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
			O	LO	LNO	NO
0. Prerequisites of quality
	0.1 Legal and institutional environment			X			The current legislative framework is not fully operational. Interinstitutional cooperation is not adequate. When individual data are disseminated-directly or indirectly-because of the small number of operators in some economic branches, interested respondents are not notified of such situation.	Short-term (ST). Have the CSS and the CPS meet. Have the Board of Directors meet and decide on the personnel bylaws and the work plan. Notify officially respondents when their individual data will inevitably be –directly or indirectly– disseminated.
	0.2 Resources				X		INSEED’s financial independence is not guaranteed. It does not offer attractive pay rates. INSEED is sometimes slow in making use of its funds.	Medium-term (MT). Guarantee permanent financing for INSEED’s long-term programs.
	0.3 Relevance			X			The relevance and serviceability of the national accounts data suffer from gaps in the source data.	MT. Improve the monitoring of users’ needs and build these into the new system.
	0.4 Other quality management			X			The process of compiling the accounts focuses on timeliness, to the detriment of other aspects of quality.	MT. Take advantage of the change of base year and system to incorporate all existing data sources.
1. Assurances of integrity
	1.1 Professionalism			X			Technical criteria are not sufficiently weighted for selection of the Director General.	MT. Reinforce the technical nature of INSEED.
	1.2 Transparency				X		Publications do not mention the conditions governing data collection, processing, and dissemination. Access to statistics by CCM members prior to their release to the public is not publicized. No advance notice is given of major changes in methodology, source data, and statistical techniques.	ST. Mention in INSEED publications the conditions governing data collection, processing, and dissemination. Publicized early access of CCM members to statistics. Give advance notice of any major change in methodology, source data, or statistical techniques.
	1.3 Ethical standards			X			INSEED has no specific ethical standards applicable to its staff. Standards applicable to civil servants in general are not systematically indicated to staff, particularly to short-term contractuals.	ST. Adopt ethical standards specific to INSEED and make sure all staff, in particular temporary staff, are aware of these standards.
2. Methodological soundness
	2.1 Concepts and definitions				X		Simplified system basically restricted to production and goods and services accounts. The departures between the simplified system applied to Chad’s national accounts and 1993 SNA are not documented. There is no plan to migrate to the 1993 SNA.	ST. Prepare documentation on the departures between the simplified system and 1993 SNA. Design and adopt a migration plan to the 1993 SNA.
	2.2 Scope			X			The sequence of accounts recommended in the 1993 SNA is not completely compiled. Some foreign-based businesses operating in Chad are not included.	MT. Move ahead with introduction of the ERETES system, which produces more complete tables. ST. Include the missing activities of foreign-based businesses operating in Chad.
	2.3 Classification/sectorization				X		No institutional sectorization exists. Classifications of products and activities are specific to Chad.	MT. Adopt an institutional sectorization in common with all other participants in the statistics system. Adopt the comprehensive Afristat classification of products and activities.
	2.4 Basis for recording			X			Market prices are used, with a few exceptions. Source data on government finance are not on an accrual basis, but on a cash basis.	MT. Implement ERETES
3. Accuracy and reliability
	3.1 Source data				X		The compilation of ECOSIT-2 survey is not completed. No detailed data are available on the informal sector or on marketable services, except for the base year. Detailed data on foreign trade are unavailable since 1997. The Agriculture survey is unavailable since 2001.	ST. Complete the compilation of ECOSIT-2 survey. MT. Improve the use made of administrative data. Produce and use detailed statistics on foreign trade. Repeat the agricultural survey.
	3.2 Assessment of source data				X		There is no systematic assessment of source data.	ST. Assess critically and systematically the data used and perform cross-checks.
	3.3 Statistical techniques				X		There is heavy reliance on fixed ratios and on indirect estimates. Unique indicator for volume of output and of value added. All products and activities are considered as elemental.	ST. Compare ECOSIT 1 and ECOSIT 2 to assess the validity of assumptions about fixed ratios, and reevaluate them if necessary.
	3.4 Assessment and validation of intermediate data and statistical outputs				X		Opportunities for assessing intermediate data are overlooked.	ST. Conduct a systematic review of household consumption by product.
	3.5 Revision studies				X		Data are reviewed twice a year, but no systematic studies of changes are conducted between successive versions.	ST. Monitor successive versions of the accounts and explain any differences to users.
4. Serviceability
	4.1 Periodicity and timeliness			X			The data are available on an annual basis, but no precise schedule promotes their dissemination after the meeting of the Macroeconomic Framework Committee.	ST. Establish a strict release calendar and adhere to it.
	4.2 Consistency			X			Foreign trade data since 2002 are sharply inconsistent with those of the BOP.	ST. Review foreign trade data and analyze the reasons for discrepancies with the BOP. Develop consensus on estimates for the BOP and the national accounts.
	4.3 Revision policy and practice			X			Revisions are mentioned in publications, but their amount is not quantified.	ST. Establish a classification that will clearly designate the successive degrees of revision and adhere to it strictly.
5. Accessibility
	5.1 Data accessibility				X		There was no hardcopy publication between 1997 and 2004. No preannounced schedule is available. The data offered at the website are overly aggregated.	ST. Revise and adhere to the calendar for hardcopy publication. Revise the publication of the accounts, in particular the contents of the tables. Expand dissemination at the INSEED website. Confine to a minimum the revisions to the historical series.
	5.2 Metadata accessibility				X		The metadata in English available at the IMF’s DSBB are not periodically updated and detailed enough. No metadata in French are made available to the public.	ST. Update GDDS metadata posted in English at the DSBB and provide more details herein. Make metadata in French available to the public. Make a practice of adding methodological notes to publications.
	5.3 Assistance to users			X			Only the Director General is indicated as the contact person for all statistics compiled by the INSEED. There is no catalogue of publications.	ST. Indicate specific contact person(s) for national accounts statistics. Establish and disseminate a catalog of publications.

0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

The Government Flow of Funds Table (TOFE, for Tableau des opérations financières de l’État):

Decree 141/PR/MEF/2003 on the organization of the Ministry of Economy and Finance (MEF) stipulates (Article 33) that the Directorate of the Treasury and Public Accounts (DTCP), which falls under the Treasury General Directorate (DGT), is responsible for preparing the weekly and monthly statements of government financial operations.

That same decree (Article 13) makes the Research and Forecasting Directorate (DEP), which falls under the Budget General Directorate (DGB), responsible for “conducting the macroeconomic framework setting exercise, in collaboration with other offices and agencies: to this end, it compiles the Government Flow of Funds Table (TOFE), on the basis of information collected from the appropriate agencies.” The TOFE comprises revenue, expenditure, lending minus repayments, a balance, and financing. The TOFE, together with debt statistics (see below), serves as the government finance statistics (GFS) for Chad.

The DTCP is the focal point for the conduct of government budget transactions, which means it has regular financial information available and produces monthly cash-based statements of government financial operations and detailed accounting statements. It provides these statements to the DEP for preparing the quarterly and annual TOFE. The information thus available may allow the compilation of a monthly cash-based TOFE.

A TOFE Committee was established by Decree 058/MF/SE/DG/00 of the Minister of Economy and Finance to prepare the TOFE. The DEP serves as its secretariat, and the Treasury provides its legal counsel. Its membership embraces all the source data providers in the MEF and the BEAC. Its mandate is to facilitate collaboration between these agencies and to validate the TOFE. The fact that the quarterly and annual TOFE⁵ draw essentially upon accounting and administrative records using a common analytical framework ensures consistency in the methods used by the DTCP and the DEP. The TOFE Committee constitutes a working arrangement to promote consistency of methods and statistical outputs and to explain any discrepancies among the sources.

Public debt:

Article 18 of the above-mentioned decree makes the Debt Directorate (DD) of the DGT responsible for preparing the public debt statistics.

The DD, as ordonnateur (the payment authorizing authority), tracks domestic and external debt transactions. For the external debt, the DD has a copy of all loan contracts, and donors provide it with notices of disbursement and requests for repayment upon maturity. As to the domestic debt, it arises not from the issuance of treasury bills or borrowings but essentially from the financial difficulties of the government, which has drawn upon statutory advances from the central bank and has incurred commitments that it has not honored. The DD is responsible for determining the amount of this debt and seeking ways to settle it, and therefore has at hand the information needed to calculate the statistics.

Nevertheless, the responsibility for disseminating the GFS (TOFE and debt statistics) is not specified in any law and is not clear in practice. The MEF regards its responsibility in this matter as limited to communicating the GFS to other data-producing agencies, to regional and international organizations, and to users on demand. The MEF notes that INSEED has the mandate to disseminate official statistics. The terms and conditions under which INSEED performs this task are not defined in any formal understanding. Actually, INSEED, and the BEAC as well, disseminate annual TOFE only.

Because coverage is limited to the central government entities included in the General State Budget (BGE, for Budget général de l’État), production of the TOFE is not subject to any legal constraints in terms of access to information, since it relies primarily on accounting and budgetary sources.

0.1.2 Data sharing and coordination among data-producing agencies are adequate

Apart from the existence of the TOFE Committee, no other formal provision exists for the exchange of source data or outputs among statistics-producing agencies. Nevertheless, the MEF communicates the TOFE to the BEAC and to INSEED regularly.

Also, the CCM—in charge not of statistics but of forecasts and preparation of economic budgets (see paragraph 0.1.2 in National Accounts)—could provide a useful forum for exchanging views. The forum would be not only on the TOFE but also on the information needs of the BEAC and INSEED relating to source data on government finance.

The MEF is satisfied with the exchange of information with INSEED. The BEAC, which is a member of the TOFE Committee, reports monetary statistics to the Treasury, which uses these to provide bank financing data for the TOFE.

The DD represents the MEF on the CBP instituted by the BEAC. The CBP constitutes a framework that promotes consistency of sources for foreign debt statistics.

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

To the extent that the TOFE source data are aggregate accounting data on a large number of taxpayers, the rules governing source data confidentiality are observed for tax returns. In addition, because the sectoral coverage is limited to the central government agencies covered by the Budgetary Central Government (BGE, for Budget général de l’État), the source data involve only the governmental agencies and units. Moreover, employees are legally bound by civil service and government accounting regulations to ensure the confidentiality of individual information in their possession.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

The MEF is legally entitled to access all financial information for the budgetary central government. Moreover, as the government entity responsible for financial control of the public sector, the MEF may demand access to financial information of all central or local government entities, as well as the government’s social security and retirement funds.

More specifically, by virtue of Organic Finance Law 11–62 of May 11, 1962 (Articles 174 to 177), the government accounting officers (comptables publics) are required to compile a monthly statement comparing revenues by budget line and a summary statement of payments by budget chapter. The central treasurer is also supposed to prepare a monthly statement of budgetary and treasury transactions, covering the Treasury’s financial situation as well as the balances of budgetary revenues and expenditures for each budget and special account. This law also requires that the annual compte de gestion be submitted within three months after the close of the fiscal year.

Beyond these legal obligations, the TOFE Committee is a major working arrangement that ensures access of the DEP to source data.

The MEF produces the other principal source data for preparing the TOFE. Information on the successive stages of expenditure is produced by the Commitments Directorate (DE, for Direction des engagements), which comes under the same General Directorate as the DEP. As well, the DD supplies external and domestic debt data. Domestic bank financing is covered directly in the monetary survey produced by the BEAC.

With respect to the public debt statistics, the decree organizing the MEF specifies that the DD must centralize the payment-due notices issued by creditors so that payment can be ordered (by mandatement) and so that payments and arrears can be recorded. It must also collect data on loans and participate in the issuance of borrowings, together with the DTCP.

The data providers generally comply with the legal and administrative arrangements for producing and communicating the source data. Although the law does not provide any penalties for failure to respect these provisions, the MEF has sufficient powers of administrative discipline to ensure that this information is communicated.

Because the TOFE relies essentially on available accounting and administrative data, compilation does not impose an additional work burden beyond the normal budgetary and accounting activities and financial control of government operations.

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

Under the DTCP, the Division of Statistics and Data Analysis (DSAD, for Division des statistiques et analyse des données) was created recently; its duties are still currently being performed by other divisions of the DTCP. The DSAD already has three economists on staff—a number seemingly sufficient to meet statistical needs. The division will be responsible for aggregating the source financial data monthly for the TOFE. The work of collecting and compiling the data relies on the resources of the treasury accounting network and on three divisions that are devoted respectively to the compilation of revenue transactions, expenditure transactions, and accounting.

The Forecasting Unit (SP) of the DEP, responsible for preparing the quarterly and annual TOFE, has a staff of four economists. Even with the other responsibilities that have been devolved onto this unit, its staffing level is deemed adequate.

The Payment Authorization and Monitoring Unit (SMS, for Service du mandatement et du suivi), part of the DD, compiles the debt statistics. For this purpose it has a statistics section, with a single middle-ranking officer. It is thus insufficiently staffed. Nevertheless, the SMS has support from two other units within the DD: the Domestic Debt Unit and the Foreign Debt Unit, which collect and aggregate data as part of their duties.

DSAD, SP, and SMS staff attend seminars organized generally by regional organizations such as Afristat, the BEAC, and the BAD, as well as by international organizations including the IMF. However these events do not form part of a formalized training plan. The DEP recently adopted a training plan designed by an external consultant. The plan includes aspects relating to preparation of the provisional TOFE, as well as training in computer tools.

MEF units are at constant risk of losing their best officers to other jobs or other organizations. Pay levels for MEF personnel, including bonuses, while generally higher than those for other ministries, are less attractive than those offered, for example, by the university. Individual salary is determined by the level of formal education and the person’s ranking within the administrative hierarchy.

The SP has two computers, while the SMS and the DSAD have one computer each. Internet access is generally available only to directors, which limits the ability to exploit the resources of the Web in GFS work. Staff enter and process source data for the TOFE in Excel, which is sufficient to meet the current needs of the statistics units. However, the DTCP is beginning to feel the need for an integrated computer application dedicated to compiling data from the network of treasury accounts offices.

The DD uses the Debt Management and Financial Analysis System (DMFAS)⁶—a software supplied by the United Nations Conference on Trade and Development. This program sets an international standard for monitoring and analyzing the debt and ensuring efficient management of debt stocks and flows within a single application. Only limited use is made of the program’s functionalities, however.

The need for computer security measures has not yet been taken seriously. There are no formal practices for safeguarding data. However, the files containing the statistical series are generally saved on several computers, and backup copies are kept.

Offices and equipment are not very functional. Few meeting rooms are available, and this makes teamwork difficult.

The DTCP relies for delivery of source data on the postal service, which does not function well in remote locations. The DTCP is examining the possibilities of partnering with the BEAC to have government accounting materials delivered, together with the documents remitted by the local BEAC offices.

Funding for preparation of the GFS comes exclusively from the annual budget of the Treasury and Budget directorates and cannot be identified in isolation.

0.2.2 Measures to ensure efficient use of resources are implemented

The procedures for evaluating the performance of staff involved in compiling the TOFE are the same as those applicable to all public servants. A rating system, based on criteria of diligence, competence, output and conduct, is used to determine career progress and pay.

There is no measurement of the costs of producing the TOFE. However, because the source data come primarily from accounting and administrative documents produced in the course of ongoing statutory and regulatory work, the direct costs of production are modest.

The MEF generally benefits from external expertise for improving its performance as regards budget monitoring, provided under technical assistance programs financed by donors. However, Chad has received very little technical assistance relating specifically to the GFS in recent years.

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

The statistics are thoroughly integrated into the budget preparation process, the financial framework exercise, and multiyear planning. The statistics also meet the needs expressed by decision makers.

Moreover, because the statistics are presented in a manner very similar to that of A Manual on Government Finance Statistics, 1986 (GFSM 1986), they also meet the main information needs of regional and international organizations. The economic classification used makes it possible to derive information on the principal government finance aggregates.

The statistics do not however respond completely to the needs of the national accounts, primarily because their scope is limited to central government as covered by the BGE. The TOFE does not cover miscellaneous central government agencies (ODACs, for organismes divers d’administration centrale), social security, or local governments. However, INSEED has not formally expressed a need in this regard.

There is no formal process for surveying users’ needs. The BEAC, as a member of the TOFE Committee and CCM, and the INSEED, as a member of the CCM, are able to express their information needs at meetings of these committees. The needs of other users are not surveyed.

0.4.1 Processes are in place to focus on quality

The MEF accords particular importance to the accuracy and reliability of the data it uses as inputs to its statistical work. The preferred sources of data are accounting and administrative documents, where standards of accuracy and reliability are high.

The data collected are compiled and validated by the TOFE Committee, which brings together the agencies that supply the source data and those that produce statistics. There are thus systematic controls of data accuracy, and adjustments are made when discrepancies are observed. Nevertheless, the utility and accessibility of the data are not sufficiently taken into account.

The compilers of the TOFE rarely make use of statistical techniques for improving the quality and consistency of their information output as a whole. Thus, any accounting corrections that are made after the statistics are produced are not incorporated, and discrepancies arising from the basis of recording transactions are not properly adjusted. However, such corrections are generally of minor scope.

0.4.2 Processes are in place to monitor the quality of the statistical program

The processes in place for monitoring accuracy and reliability of the TOFE are in general satisfactory.

The accounting data are subjected to strict monitoring and control for accuracy and reliability. An integrated expenditure management system (CID, for Circuit intégré des dépenses) monitors the reliability of the expenditure data recorded on a commitments basis, other than those relating to salaries. Data on bank financing come from the monetary statistics produced by the BEAC and are not reconciled with the correspondent data resulting from the treasury accounts, as they should be. Staff systematically reconcile with donors’ data the data on the external debt derived from credit and debit notices sent by the BEAC.

Once the TOFE has been compiled and released, no regular quality monitoring is done. Nor are there any periodic reviews to identify steps necessary to maintain quality requirements. The DEP is working with an external expert to improve its budget monitoring performance.

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

Data accuracy and reliability take preference over timely compilation of GFS and their accessibility. Although budget execution data are available promptly through the daily and weekly budget monitoring procedures, the TOFE is prepared only after data that are deemed more reliable are in hand.

Nevertheless, to shorten production times and increase reliability, the DTCP has plans to improve the government chart of accounts and to computerize production of the General Balance of Accounts (BGC, for Balance générale des comptes). The directorate is considering hiring an external consultant.

1.1.1 Statistics are produced on an impartial basis

The source data are taken primarily from accounting data prepared in accordance with legislation that guarantees the professional independence of the government accounting officers (comptables publics).

Article 54 of the Convention governing the Central African Economic Union provides that budgetary legislation, national accounts, and macroeconomic data, must be harmonized within five years from the date this convention entered into force (June 1999). The Council of Ministers is to adopt regulations and directives to achieve this harmonization and a schedule for implementation. However, the TOFE has not yet been harmonized.

A culture focused on professionalism is also encouraged by the formalized process for producing the TOFE, involving the TOFE Committee, and by the fact the authorities use the TOFE as a reference document in government financial and budgetary planning.

Staff professionalism has improved, as evidenced by the fact that the DEP has recently contracted economists with increased qualifications in terms of university degrees and/or experience than was the case previously. The DEP has adopted a training plan that includes modules on preparing the TOFE. MEF officers occasionally take part in seminars and training events sponsored by regional or international organizations.

Although the government is known to have privileged access to the GFS before their dissemination, there is no indication of political interference in their preparation or dissemination.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

Data sources are selected primarily on the basis of their reliability. Thus, the preferred data sources are administrative and accounting records, the reliability of which is assured. Production of the TOFE consists primarily in the reorganization of component data: statistical techniques are used only exceptionally.

The MEF distributes the TOFE to a restricted list of correspondents. The BEAC and INSEED decide on dissemination, without any prior consent from the MEF. These two agencies also decide the formats and the level of detail of the published TOFE.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

The MEF’s target public consists of professionals who are familiar with macroeconomic statistics and who cooperate in producing the TOFE. The producing agency provides explanations of the data to its correspondents, particularly through the work of the formal committees. The annual TOFE made public by the BEAC and INSEED does not contain any explanatory documentation on the information produced.

Generally, the MEF does not make a practice of intervening when the data are misinterpreted or misused by the public at large. One case involves some quarterly statistics named “Treasury Revenues and Expenditures” (under the heading “Government Finance”), compiled by INSEED on its own initiative and disseminated in its Bulletin trimestriel de conjoncture-a quarterly economic bulletin. These data have long contained obvious internal inconsistencies, such as negative revenues and expenditures, that have never evoked comment from the MEF.

1.2.1 The terms and conditions under which statistics are collected, processed, and disseminated are available to the public

The MEF does not publish any documentation on the terms and conditions governing production of the TOFE, other than through the GDDS metadata for Chad. Metadata have been available since July 2002 in English at the IMF’s Data Dissemination Standards Bulletin Board (DSBB) but have not been updated since that time.

Several decrees, laws, and regulations are published in the official Gazette and are publicly available in bookstores and libraries. They comprise the 2003 decree organizing the MEF and specifying the responsibilities of its directorates for the production of statistics, the Organic Finance Law establishing the basic provisions for accounting and budgetary monitoring, and various regulations regarding government accounting and budgetary classification and accounts. Compilations of legislation and regulations governing the public finances are also available in major bookstores and offer further means of public access to this information.

The MEF does not determine the terms and conditions for disseminating the GFS: INSEED and the BEAC disseminate the statistics on their own initiative. The publications of INSEED and the BEAC make no mention of any sources where further information can be obtained.

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

The GDDS metadata indicate that the government has privileged access to the TOFE before it is released. The BEAC and INSEED also have privileged access to the TOFE, and this is a matter of public record at least, because these two agencies disseminate it.

1.2.3 Products of statistical agencies/units are clearly identified as such

The producing agency is not always clearly identified in the annual TOFE, disseminated by the BEAC on its website and in the bulletin Études et statistiques and by INSEED in the brochure Comptes économiques et financiers (Economic and Financial Accounts).

More generally, the MEF is not involved in the dissemination of the TOFE and does not require the BEAC and INSEED to cite the MEF as a source for the TOFE published.

1.2.4 Advance notice is given of major changes in methodology, source data, and statistical techniques

The published TOFE is not accompanied by any commentary. The INSEED and BEAC limit themselves to including the GFS among a series of tables summarizing the sectoral and macroeconomic statistics.⁷ The fact that the agency producing the GFS is not involved in the dissemination makes it impossible to guarantee that changes of methodology or of source data will be announced in advance or at the time of dissemination. The BEAC and INSEED alone are informed of modifications decided, during meetings of the TOFE Committee and CCM.

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

No specific ethical rules concern the production of GFS. However, MEF staff operate in accordance with administrative rules, such as the General Civil Service Statutes, as well as the public accounting regulations, pertaining to professional secrecy and conflict of interest policy. New staff are made aware of the ethical standards when they join the organization, but they are not formally and periodically reminded of such standards afterwards.

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

The principles for compiling the statistics generally follow the analytical framework of the GFSM 1986. The structure of the TOFE, however, departs slightly from that recommended by this manual in order to allow for special monitoring of oil revenues and other expenditure commitments earmarked from those revenues.

The definitions and concepts of revenues, expenditures, and financing are generally consistent with the recommendations of the GFSM 1986, with three major exceptions. First, domestic lending minus repayments is included in financing, as recommended in the Government Finance Statistics Manual 2001 (GFSM 2001). Second, repayments from onlending are recorded as nontax revenues, whereas they should be treated as financing. Last, financing includes an item for “exceptional financing”, i.e., resources from foreign donors. The portion of this financing that corresponds to budgetary assistance (grants) should be considered as revenue in GFS, even though it is recorded as financing for the monitoring by the IMF of the Poverty Reduction and Growth Facility-supported program.

Oil revenues paid by the production and marketing consortium are handled under a special regime established by Decree 095/PR/MEF/04 of March 18, 2004. According to those rules, oil revenues include indirect revenues, corresponding to taxes, levies on companies, and customs duties. Oil revenues also include direct revenues corresponding to royalties and dividends, from which a provision is set aside for repaying the government’s debt to the European Investment Bank and to the World Bank for financing the oil investments.

Direct and indirect revenues are paid into an account held in the name of Chad at the Citibank in London. Ten percent of these revenues is deposited in the Fund for Future Generations account, while the remainder goes through two government accounts held in two national banks before being deposited with the BEAC. Of the amount deposited, 80 percent goes into a stabilization account, 15 percent to the treasury account, and 5 percent to an account for the oil-producing region. All direct and indirect oil revenues, as well as the provision deducted at source for repaying the external debt, are treated as revenues in the TOFE, and this is sound practice. As well, bank transactions and the above-mentioned provision are correctly recorded on the financing side.

There is an overall balance of expenditures and revenues on a commitments basis, as well as an overall balance on a cash basis, after changes in arrears. Financing is broken down into external and domestic financing.

No program has been adopted for the transition to the analytical framework of the GFSM 2001.

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

The coverage is limited to the budgetary central government.

The general government sector for Chad consists of central government, including the social security and pension funds, and local governments. Chad has two social security funds: the National Social Provident Fund (CNPS, for Caisse nationale de prévoyance sociale) and the National Pension Fund of Chad (CNRT, for Caisse nationale de retraite tchadienne). The CNPS covers private-sector employees and certain public employees working under contract, while the CNRT covers civil servants. Because these pension funds are mandatory and based on defined benefits, they are treated as part of the general government sector in accordance with the 1993 SNA methodology.

The central government sector is broken down into central government covered by the BGE, special treasury accounts, and ODACs.

Local governments have financial, administrative, and political autonomy. Their own revenues are limited and consist essentially of market-stall license fees and land rentals. Most of their financial resources consist of transfers from the BGE.

The scope of the TOFE is that of budgetary central government, which includes the Office of the President of the Republic, the Prime Minister’s Office (Primature), the Parliament, the General Secretary of Government, the National Mediator, and 19 ministries. Certain budgetary entities, however, such as the university and hospitals, have autonomous budgets, recording revenues and expenditures that are not integrated into the TOFE. Nevertheless, the MEF considers that are such transactions are insignificant in their amounts.

Thus, the TOFE does not cover the entire central government. Also, the TOFE does not cover the special accounts of the Treasury and of the ODACs—public entities of an administrative nature that engage in quasi-budgetary activities.

Coverage is limited to the budgetary central government, with respect to preliminary, sub-annual, or annual data. Data on local government operations are not collected.

The statistics show outstanding debt, both domestic and external. The scope of the statistics includes direct debts of the Treasury, passed-through debts (onlending), and guaranteed debts outstanding. Nonetheless, the statistics produced are global and do not show items for each of these three types of debt; as a result, the scope of coverage could not be verified and confirmed.

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

The classifications/sectorizations used do not always meet international standards.

The TOFE is disseminated in a single summary table showing the main economic classifications of revenues, expenditures, and financing at a level of detail that covers the principal operations. This classification is generally close to that of the GFSM 1986.

However, some discrepancies reflect primarily a concern about tracking oil revenues and related expenditures and monitoring certain specific operations. The MEF notes that this classification was adopted to meet the needs of certain donors—an issue that has to be managed through the production of specific follow-up statistics and not through the reconfiguration of the TOFE. This practice leads to a classification that does not always meet international standards. Actually, the classification adopted in the BGE adheres more closely to the GFSM 1986 than the TOFE and uses a more coherent economic classification.

Revenues are classified in the TOFE into two broad categories: (1) oil revenues from royalties, dividends, and taxes on the oil industry and (2) non-oil revenues. The latter shows breakdowns classified in accordance with the GFSM 1986 recommendations.

Expenditures are broken down into current and capital (investment) expenditures. Current expenditures are classified to show the economic nature of spending (wages and salaries, goods and services, etc.). A memorandum item shows expenditures of earmarked oil revenues. Capital spending is broken down into two categories: domestically financed investment and foreign-financed investment. The first category includes direct investments, investments of oil revenues, and an item for counterpart funds, showing the government contribution to social projects cofinanced by donors. The second category covers investments financed under the Heavily Indebted Poor Countries Initiative (HIPC Initiative) and those financed through loans and grants. These breakdowns do not conform to the GFSM 1986 classification.

Financing is classified as external financing, domestic financing, and exceptional financing. This last category relates in effect to external financing from bilateral and multilateral donors. The portion of this financing that corresponds to budgetary assistance (grants) should be classified as revenues, and the other portion, as external financing.

The details provided in each category employ headings that do not always allow the operation to be clearly identified. While some of the details provided are undoubtedly useful, the categories do not systematically group items into a readily understandable economic classification: This practice makes it difficult to read and interpret the statistics. The MEF notes that this presentation, which does not always meet international standards, was adopted to meet the needs of certain donors.

An item, termed the financing gap, demonstrates a gap between the cash-based deficit and its financing. This gap arises primarily from the often limited coverage of transactions, adjustments to expenditures recorded on a commitments basis to bring them to a cash basis, and the use of monetary statistics to determine the item for domestic bank financing. There is no reconciliation with data from the BGE, which shows different figures.

The stock of debt is broken down into domestic and external debt. Actually, the Chad domestic debt does not include any classical debt instrument, such as government bonds, but covers arrears that arose from past financial difficulties. The DD has compiled an inventory of this debt with a view to settling it. The classification of this debt is broken down into dettes conventionnées, that is, commitments for which restructuring agreements have been signed and other legal commitments. Prior to the settlement of arrears in 2002, the domestic debt also included such arrears, and this was good practice. When the treasury accounts with the banks show a negative balance, this is not classified, because it has to be under domestic debt.

The external debt is classified by creditor. There is no classification by initial or residual maturity.

2.4.1 Market prices are used to value flows and stocks

The TOFE records only flows, which are valued globally on the basis of current prices. They correspond, in the case of revenues, to amounts actually received; in the case of expenditures (excluding interest on debt), to amounts committed; and in the case of interest on the debt, to amounts as shown in the payment schedule.

For interest on the external debt, the amounts shown in the scheduled payments reflect the exchange rate at due date. In the case of late payment, any resulting differential is recorded as a change in arrears.

The stock of debt is recorded at its face value, without interest. The stock of external debt is converted into local currency at the international exchange rate (from the IMF website) for the last day of the period under review.

2.4.2 Recording is done on a cash basis

Revenues are recorded on a cash basis at the time payments are made. Expenditures are recorded on a commitments basis, at the stage preceding the issuance of the payment order (ordonnancement). The change in arrears item allows a return to the cash basis. Nevertheless, this post overlooks certain expenses that have been committed but for which no payment order has been issued, and this produces a slight distortion in the cash-based deficit.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

The TOFE records transactions on a gross basis, while financing is recorded on a net basis, consistent with the principles of the GFSM 1986. Accounting adjustments made at the end of the year are applied to the last month of the year. Tax refunds are given in the form of a tax credit and are deducted from tax revenues receivable, which are thus recorded on a net basis.

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

Data collection programs are based on complete sources that result from the centralization and production of accounting and budgetary documents on execution of the BGE. Theses documents mainly comprise the BGC, budget execution reports, and data on the bank financing of the government supplied from the BEAC.

The network of government accounting officers (comptables publics) records revenue, expenditure, and treasury transactions on a daily basis. Monthly accounting statements are compiled at the main treasury office and are recapitulated to produce the BGC and the statements of revenues and expenditures.

Expenditures are recorded in the accounts on the basis of payment order (ordonnancement). Nevertheless, in preparing the TOFE, the DEP uses a survey produced by the Commitments Directorate reporting expenditures at the different stages: commitment, payment order, recording, and payment by the payment officer. The TOFE records expenditures on a commitments basis.

The budgetary classification of revenues and expenditures allows for easy movement to the TOFE presentation. Moreover, the classification system makes it possible to identify transfers between levels of government, although the current institutional coverage does not require that these statistics be consolidated.

Financing is obtained from data provided by the monetary survey, in the case of domestic financing, and from the DD, in the case of external financing. The data provided by the DD are supplemented by data reported from the Investment Directorate (DI) to provide a level of detail that covers project loans, program loans, and exceptional external financing.

Data on local government are not collected, and there is no official register to identify public-sector units. Geographic coverage of the national territory is complete for the categories of transactions reported in the TOFE.

Because the data collection program is based on the accounting and budgetary work of the MEF, changes in the data sources reflect the evolution of regulation in this area.

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

The source data come from accounting and administrative records that summarize information using the same concepts, definitions, and analytical framework as the TOFE, and the same sectorization.

Revenue data are recorded on a cash basis in the accounting statements that accompany the BGC, consistent with the TOFE recording basis.

Expenditure data, excluding interest on the debt, come from administrative records produced by the DE. They summarize expenditures at the different levels of commitment, payment order, recording, and payment, using the budgetary classification. The commitment (engagement) stage is the one used for the TOFE. An adjustment item is used to calculate the cash-based deficit. However, this item overlooks certain expenditures that have been committed but for which no payment order has been issued (DENOs, for dépenses engagées non ordonnancées).

The monetary statistics are used to determine domestic bank financing in the TOFE. This practice calls for two observations. First, the scope of the monetary statistics may be broader, because the TOFE covers only the BGE. Second, the cash-based deficit in the TOFE results from the recording of transactions at the time when checks are collected (receipts) and issued (expenditures), while the monetary statistics are recorded at value date.

External financing and interest on the data are extracted from the source data reported by the DD, which cover the same scope. These data identify repayments of principal and interest corresponding to normal maturities and to arrears, as well as debt cancellations under the HIPC Initiative.

The levels of detail in the source data are sufficient to allow the economic classification of transactions in accordance with the GFSM 1986, as well as the monitoring of transactions financed from earmarked oil revenues.

This level of detail allows the identification of transfers between levels of government. The portion of central government revenues transferred to the ODACs and the municipalities appears in an item called recettes affectées (earmarked revenues), broken down by the economic nature of the revenue item, and transfers are classified by beneficiary.

Supplementary data sources are generally not used when compiling the TOFE.

3.1.3 Source data are timely

Source data are generally provided during the month following the period of reference. The accounting data from some of the more remote accounting offices arrive late, by as much as two months, essentially because of problems with postal service. The DTCP is examining ways to improve the timeliness of transmission, in cooperation with the BEAC, which has the means for transporting documentation through its network.

The time limits applicable to data from accounting sources are established in the Organic Finance Law. For data from other administrative sources, reporting takes place in the context of the monitoring and control tasks that have been devolved to departments and agencies. The units producing these data are also members of the TOFE committee.

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

The compilation of the data in Excel files makes it possible to test their internal consistency, as well as to perform certain cross-checks between data, especially for expenditures. Moreover, the data-reporting agencies are members of the TOFE Committee, which is responsible for preparing the GFS, and data can therefore be validated and any discrepancies readily eliminated.

Tests for consistency are conducted, in particular by examining outlier amounts.

The accounting data on revenues and expenditures are accompanied by detailed statements reviewed and approved by the ordonnateurs (the payment authorizing authority) before the data are sent to the MEF. In addition, because the revenue departments (taxes, customs) are also members of the TOFE Committee, they can verify the consistency of the accounting and administrative records.

As for expenditure data drawn from the budget monitoring reports produced by the DE, they can be compared against the cash-based expenditure accounts, to check their consistency.

The data on the external debt are cross-checked against two information sources—the DD and the DI.

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

The compilation process consists mostly of reorganizing accounting and administrative data; therefore, no statistical techniques are needed.

Moreover, as accuracy of source data takes precedence over timeliness, missing data are not estimated, even if this results in a delay in compiling the TOFE.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

There is little adjustment to the source data. However, a change in an arrears-adjustment item is used to convert expenditures recorded on a commitment basis to a cash basis. This item covers expenditures for which payment orders had been issued but were not paid. As mentioned before, it overlooks certain expenses that have been committed but for which no payment order has been issued.

3.4.1 Intermediate results are validated against other information where applicable

The TOFE does not identify intermediate results that could be validated from other macroeconomic statistics, and in fact very few comparable statistics are available.

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

For the external debt, flow data are systematically compared with stocks data. The integration of flows and stocks results from the use of a single application to handle these operations. It should be noted that this integration also takes account of exchange-rate fluctuations, which makes it impossible to obtain perfect equivalence.

The TOFE data on the flow of domestic debt cannot be compared to the stock of that debt, which consists largely of government arrears that the DD has inventoried for settlement.

Few related indicators exist to verify the behavior of intermediate series for the TOFE. For example, the behavior of tax revenues cannot be verified by that of GDP, because of the lags introduced by collection performance and by the use of different recording bases: cash in the case of the GFS and accrual in the case of the national accounts.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

Data on domestic bank financing supplied by the BEAC are directly inputted in the TOFE. Discrepancies with this item, as obtained from the BGC, are not analyzed. External debt statistics are compared with those from creditors, as well as with the BOP statistics.

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

Archiving practices are not of a quality to allow for revision studies and analyses.

4.1.1 Periodicity follows dissemination standards

The periodicity is annual for dissemination of statistics on central government operations to the general public by the INSEED and BEAC, thus not meeting the requirements of the GDDS for a quarterly dissemination. The external debt statistics are disseminated on a quarterly basis, while domestic debt statistics are not disseminated at all.

4.1.2 Timeliness follows dissemination standards

The TOFE is generally disseminated nine months after the end of the reference year, at the BEAC and IZF websites. This timeliness is not consistent with the GDDS recommendation for one quarter after the reference date. The quarterly statistics on the external debt are disseminated by INSEED three months after the end of the period.

4.2.1 Statistics are consistent within the dataset

Because the quarterly statistics are based on cumulative source data from the beginning of the year, the annual statistics are actually those for the last quarter of the year. Thus, checking consistency between successive quarterly TOFE pertaining to the same year is of little significance.

For revisions made to data within the accounting annual deadline to the end-of-year TOFE, the revisions are entirely attributed to the fourth quarter TOFE. Also, any errors and omissions are attributed to the fourth quarter, although they may relate to previous quarters. If corrections are made after closing date, they will be attributed to the following period. This practice distorts any comparison among the quarterly data over the long period. That distortion, however, is not deemed significant enough to affect the relevance of the TOFE.

The TOFE provides flow data only, and changes in stocks are not calculated. For the debt statistics, flows and stocks are handled within the same application, using the DMFAS (in French, SYGADE), which provides for proper integration of flows and stocks.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

Five-year time series are available for the TOFE and do not show inconsistencies over time. When a change occurs in methodology or presentation, the historical data are not reconstructed, and no methodological note is attached to the tables disseminated.

The DEP produces, but does not publish, a semiannual economic analysis in which it compares the GFS against budget forecasts. No commentary is provided on any changes in trends from one year to the next. However, the trends of revenues and expenditures seem coherent with economic and fiscal developments.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

The TOFE is not reconciled with the national accounts statistics or specifically with the general government accounts.

The same concept of residency is used for the GFS and the BOP statistics. However, differences in transaction coverage appear in the tracking of oil revenues and expenditures financed from them (see paragraph 2.1.1). Thus, the TOFE reflects, but the BOP statistics do not reflect, the withholdings from oil revenues performed outside the country for repaying the external debt and revenues earmarked for the Fund for Future Generations account kept outside the country. Recording these withholdings is an internationally accepted good practice.

With respect to the debt statistics, the DD is a member of the BOP Committee. External debt data are compared annually between the DD and the BEAC, during the work of this committee.

The TOFE data on domestic bank financing are taken from the broad money survey. The automatic consistency of these two sources does not flow unless there is a reconciliation of the data in these two statistical frameworks. As part of the work of monitoring government financial transactions, the DTCP conducts a monthly reconciliation of data from the BGC and those from the broad money survey. However, this accounting reconciliation is not taken into account in the TOFE.

4.3.1 Revisions follow a regular and transparent schedule

No formal procedure exists for data revisions. Staff replace the preliminary statistics by final statistics as soon as these become available. According to the MEF, current users of the statistics (BEAC, INSEED, and MEF sections) expect that certain necessary revisions will be made. Because the TOFE is published so late, the data they contain are generally final; thus there is no revision cycle.

4.3.2 Preliminary and/or revised data are clearly identified

The preliminary statistics are not always identified as such. However, the TOFE, compiled with large delays, is based essentially on reliable accounting and administrative sources, which limits the scope of corrections. Moreover, because accuracy takes precedence over timeliness in compiling the TOFE, the MEF rarely resorts to preparing the statistics on the basis of provisional data that may be available from budget execution records.

4.3.3 Studies and analyses of revisions are made public (see also 3.5.1)

The MEF does not conduct any revision studies or analysis and does not generally archive the preliminary data.

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

Data dissemination is generally consistent with the analytical framework of the GFSM 1986. The disseminated TOFE consists of a single summary table showing the main economic classifications for revenues, expenditures and financing, as well as the various budgetary balances. Although no supplementary tables are provided, the level of detail is satisfactory. No additional series are disseminated to satisfy users’ needs for greater detail.

5.1.2 Dissemination media and format are adequate

The MEF is not responsible for disseminating GFS. As a result, this data-producing agency has no control over the dissemination media and format, which may actually vary from one period to the next. The BEAC disseminates the TOFE on its website and in print, and the INSEED disseminates it in print only.

5.1.3 Statistics are released on a preannounced schedule

There is no preannounced schedule for release of the GFS, nor is there a well-established pattern of assured compliance with implicitly recognized timetables.

5.1.4 Statistics are made available to all users at the same time

The public is not informed of the release of the statistics or of the procedures for accessing them. However, the channels through which statistics are disseminated are such that they are available to all users at the same time. The government, the BEAC, and INSEED have privileged access to the TOFE.

5.1.5 Statistics not routinely disseminated are made available upon request

In addition to the TOFE published by INSEED and the BEAC, other more detailed statistics are available on budget execution. These statistics are prepared for internal use only and are not made available to external users, even upon request.

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

The MEF does not publish any documentation on sources and methods used to establish GFS for Chad.

The GDDS metadata posted in English at the IMF website in July 2002 offer a description of sources and methods, but those metadata have not been updated since that time

5.2.2 Levels of detail are adapted to the needs of the intended audience

Only one version of the metadata, as posted at the DSBB, is available; thus, GFS metadata for Chad are not adapted to the specific needs of various audiences. The French version of the metadata, made available by the IMF to the authorities, is not posted at any website, nor is it published in any document.

5.3.1 Contact points for each subject field are publicized

In Chad’s metadata for GFS, posted at the DSBB, contact persons and units are indicated for the TOFE and the debt statistics.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

There is no catalog of publications.

Table 5:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Government Finance Statistics

(Compiling Agency: Ministry of Economy and Finance – MEF)

Table 5:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Government Finance Statistics

(Compiling Agency: Ministry of Economy and Finance – MEF)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element		NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
			O	LO	LNO	NO
0. Prerequisites of quality
	0.1 Legal and institutional environment			X			Responsibility for disseminating the GFS is not specified in any text and does not correspond to an established practice.	Short-term (ST). Give the body that compiles the GFS the legal responsibility to disseminate them on a regular basis, in accordance with GDDS requirements, and in adequate formats (GFS bulletin, Internet).
	0.2 Resources				X		Facilities are not very functional, and there are few meeting rooms. The GFS methodology and concepts are not widely disseminated among staff. Internet connections are available only to senior managers. There is no formal practice for guaranteeing efficient use of resources.	ST. Adopt a plan for training in GFS methodology and concepts. Upgrade facilities to make them more functional. Extend Internet access to the units involved in the compilation of GFS.
	0.3 Relevance			X			Users’ needs are not surveyed. The limited institutional coverage reduces the relevance of the TOFE.	Medium-term (MT). Introduce a formal mechanism for surveying users’ needs (annual survey).
	0.4 Other quality management				X		Practice focuses on data accuracy and reliability to the detriment of serviceability (periodicity and timeliness) and accessibility (commentaries, dissemination schedule). No formal mechanism exists for monitoring statistical quality.	ST. Strike a better balance between quality dimensions (accuracy and reliability, serviceability and accessibility): Improve dissemination timeliness (publish preliminary data promptly, for subsequent replacement by final data in accordance with a fixed calendar) and disseminate comments on the tables (graphs, explanation of trends, etc.)
1. Assurances of integrity
	1.1 Professionalism			X			The MEF undertakes no action to disseminate GFS; BEAC and INSEED are de facto disseminating some GFS on their own absent an agreement with MEF. The MEF does not make a practice of intervening when GFS are misinterpreted or misused.	ST. Specify the conditions for disseminating the GFS (formats, timeliness etc.). The MEF should formulate observations in case of erroneous interpretation or misuse.
	1.2 Transparency			X			The terms and conditions under which statistics are compiled as well as the government’s privileged access to statistics are specified in the GDDS metadata only. The data sources not always mentioned. Methodological changes are not always announced when introduced.	ST. Publicize in national sources the terms and conditions under which statistics are compiled and the government’s privileged access to statistics. Prepare and disseminate a technical note on GFS sources and methods. Ensure that methodological changes in GFS compilation are systematically indicated.
	1.3 Ethical standards		X
2. Methodological soundness
	2.1 Concepts and definitions			X			No process has been adopted for moving to the analytical framework of the GFSM 2001.	ST. Adopt a medium-term plan for moving to the GFSM 2001 analytical framework.
	2.2 Scope				X		The TOFE cover only the Central Government Budget (BGE). They do not cover certain transactions of BGE entities. ODAC, social security, and local governments are not included in the scope of the TOFE.	MT. Include the transactions of all BGE entities in the TOFE (universities, hospital budgets, etc.). Expand the TOFE sectoral coverage to include off-budget accounts (ODAC) and social security. Subsequently, extend coverage to local governments.
	2.3 Classification/sectorization			X			Some details provided in the TOFE are not systematically grouped by economic nature. There is no functional classification. Debt statistics are not classified by maturity. Repayments from onlending are treated as revenues.	ST. Systematically group details from TOFE in accordance with their economic classification. Use the GFSM 2001 classification system. Produce a classification of public debt statistics by initial and residual maturity. Reclassify repayments from onlending under financing.
								MT. Prepare a bridge table between the administrative budget classification (by chapter among other criteria) and the functional classification.
	2.4 Basis for recording			X			Expenditures are recorded on a commitment basis. Complete GFS on a cash basis are not prepared.	ST. Prepare the complete GFS on a cash basis. Record budgetary expenditures in the TOFE on a payment-order (ordonnancement) basis so that they can be reconciled with a cash basis, and payment arrears can be evaluated.
								MT. Compile debt statistics on both an accrual and cash basis.
3. Accuracy and reliability
	3.1 Source data		X
	3.2 Assessment of source data			X			Source data for revenues, expenditures, and external financing are verified and cross-checked with other information sources, but this is not done for domestic financing.	ST. Calculate domestic financing using BGC data and verify the deficit/financing identity.
	3.3 Statistical techniques			X			The TOFE require few statistical techniques. Nevertheless, the bridge table for estimating the cash-based deficit overlooks certain DENO items.	ST. Monitor DENO items (expenditure commitments for which no payment order has been issued) to include them in the TOFE.
	3.4 Assessment and validation of inter-mediate data and statistical outputs			X			Domestic bank financing from the monetary survey is not reconciled with the BGC.	ST. Systematically reconcile financing data from the monetary survey and from the BGC.
	3.5 Revision studies					X	Archiving practices are not adequate to allow revision studies.	ST. Ensure systematic archiving of preliminary and definitive data, and conduct revision studies.
4. Serviceability
	4.1 Periodicity and timeliness				X		The periodicity of TOFE dissemination is annual. The domestic debt statistics are not disseminated. The MEF does not control dissemination, which is frequently late.	ST. Disseminate complete GFS (TOFE and external and domestic debt statistics) with periodicity and timeliness consistent with GDDS recommendations.
	4.2 Consistency			X			The consistency of statistical series is generally verified, apart from the deficit/financing identity. The monetary statistics are used to determine domestic financing in the TOFE but are not reconciled with BGC data.	ST. Identify discrepancies between deficit and financing for balancing the TOFE. Prepare and disseminate monthly GFS with time series for the TOFE and the public debt (annual, quarterly, monthly).
	4.3 Revision policy and practice				X		There are no revision policies or practices, because the preliminary statistics use reliable sources, and revisions are usually small.	MT. Establish and disseminate a calendar for GFS revisions. Disseminate revision studies.
5. Accessibility
	5.1 Data accessibility				X		The TOFE is presented in a single summary table. There is no supplementary table with more detail, and the TOFE is not accompanied by any commentary or graphical presentations. The GFS are not released in accordance with a preannounced schedule.	ST. Add comments and graphs to the published statistics, explaining and illustrating changes and trends. Publish more detailed TOFE. MT. Prepare and publish a functional classification of expenditures, at least annually. Adopt and publish a TOFE dissemination calendar.
	5.2 Metadata accessibility				X		The metadata in English available at the IMF’s DSBB are not periodically updated. No metadata in French are made available to the public.	ST. Update and disseminate the GDDS metadata in English and disseminate locally the GDDS metadata in French. Prepare and publish a technical note on TOFE sources and methods.
	5.3 Assistance to users				X		The GDDS metadata indicate the persons responsible for the statistics, but the metadata are not available in French. No catalog of publications is available.	ST. Include in the published TOFE the name of a contact person for obtaining clarifications or supplementary information. MT. Prepare and publish a catalog of publications (indicating those available and how to obtain them).

View Table

Table 5:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Government Finance Statistics

(Compiling Agency: Ministry of Economy and Finance – MEF)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element		NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
			O	LO	LNO	NO
0. Prerequisites of quality
	0.1 Legal and institutional environment			X			Responsibility for disseminating the GFS is not specified in any text and does not correspond to an established practice.	Short-term (ST). Give the body that compiles the GFS the legal responsibility to disseminate them on a regular basis, in accordance with GDDS requirements, and in adequate formats (GFS bulletin, Internet).
	0.2 Resources				X		Facilities are not very functional, and there are few meeting rooms. The GFS methodology and concepts are not widely disseminated among staff. Internet connections are available only to senior managers. There is no formal practice for guaranteeing efficient use of resources.	ST. Adopt a plan for training in GFS methodology and concepts. Upgrade facilities to make them more functional. Extend Internet access to the units involved in the compilation of GFS.
	0.3 Relevance			X			Users’ needs are not surveyed. The limited institutional coverage reduces the relevance of the TOFE.	Medium-term (MT). Introduce a formal mechanism for surveying users’ needs (annual survey).
	0.4 Other quality management				X		Practice focuses on data accuracy and reliability to the detriment of serviceability (periodicity and timeliness) and accessibility (commentaries, dissemination schedule). No formal mechanism exists for monitoring statistical quality.	ST. Strike a better balance between quality dimensions (accuracy and reliability, serviceability and accessibility): Improve dissemination timeliness (publish preliminary data promptly, for subsequent replacement by final data in accordance with a fixed calendar) and disseminate comments on the tables (graphs, explanation of trends, etc.)
1. Assurances of integrity
	1.1 Professionalism			X			The MEF undertakes no action to disseminate GFS; BEAC and INSEED are de facto disseminating some GFS on their own absent an agreement with MEF. The MEF does not make a practice of intervening when GFS are misinterpreted or misused.	ST. Specify the conditions for disseminating the GFS (formats, timeliness etc.). The MEF should formulate observations in case of erroneous interpretation or misuse.
	1.2 Transparency			X			The terms and conditions under which statistics are compiled as well as the government’s privileged access to statistics are specified in the GDDS metadata only. The data sources not always mentioned. Methodological changes are not always announced when introduced.	ST. Publicize in national sources the terms and conditions under which statistics are compiled and the government’s privileged access to statistics. Prepare and disseminate a technical note on GFS sources and methods. Ensure that methodological changes in GFS compilation are systematically indicated.
	1.3 Ethical standards		X
2. Methodological soundness
	2.1 Concepts and definitions			X			No process has been adopted for moving to the analytical framework of the GFSM 2001.	ST. Adopt a medium-term plan for moving to the GFSM 2001 analytical framework.
	2.2 Scope				X		The TOFE cover only the Central Government Budget (BGE). They do not cover certain transactions of BGE entities. ODAC, social security, and local governments are not included in the scope of the TOFE.	MT. Include the transactions of all BGE entities in the TOFE (universities, hospital budgets, etc.). Expand the TOFE sectoral coverage to include off-budget accounts (ODAC) and social security. Subsequently, extend coverage to local governments.
	2.3 Classification/sectorization			X			Some details provided in the TOFE are not systematically grouped by economic nature. There is no functional classification. Debt statistics are not classified by maturity. Repayments from onlending are treated as revenues.	ST. Systematically group details from TOFE in accordance with their economic classification. Use the GFSM 2001 classification system. Produce a classification of public debt statistics by initial and residual maturity. Reclassify repayments from onlending under financing.
								MT. Prepare a bridge table between the administrative budget classification (by chapter among other criteria) and the functional classification.
	2.4 Basis for recording			X			Expenditures are recorded on a commitment basis. Complete GFS on a cash basis are not prepared.	ST. Prepare the complete GFS on a cash basis. Record budgetary expenditures in the TOFE on a payment-order (ordonnancement) basis so that they can be reconciled with a cash basis, and payment arrears can be evaluated.
								MT. Compile debt statistics on both an accrual and cash basis.
3. Accuracy and reliability
	3.1 Source data		X
	3.2 Assessment of source data			X			Source data for revenues, expenditures, and external financing are verified and cross-checked with other information sources, but this is not done for domestic financing.	ST. Calculate domestic financing using BGC data and verify the deficit/financing identity.
	3.3 Statistical techniques			X			The TOFE require few statistical techniques. Nevertheless, the bridge table for estimating the cash-based deficit overlooks certain DENO items.	ST. Monitor DENO items (expenditure commitments for which no payment order has been issued) to include them in the TOFE.
	3.4 Assessment and validation of inter-mediate data and statistical outputs			X			Domestic bank financing from the monetary survey is not reconciled with the BGC.	ST. Systematically reconcile financing data from the monetary survey and from the BGC.
	3.5 Revision studies					X	Archiving practices are not adequate to allow revision studies.	ST. Ensure systematic archiving of preliminary and definitive data, and conduct revision studies.
4. Serviceability
	4.1 Periodicity and timeliness				X		The periodicity of TOFE dissemination is annual. The domestic debt statistics are not disseminated. The MEF does not control dissemination, which is frequently late.	ST. Disseminate complete GFS (TOFE and external and domestic debt statistics) with periodicity and timeliness consistent with GDDS recommendations.
	4.2 Consistency			X			The consistency of statistical series is generally verified, apart from the deficit/financing identity. The monetary statistics are used to determine domestic financing in the TOFE but are not reconciled with BGC data.	ST. Identify discrepancies between deficit and financing for balancing the TOFE. Prepare and disseminate monthly GFS with time series for the TOFE and the public debt (annual, quarterly, monthly).
	4.3 Revision policy and practice				X		There are no revision policies or practices, because the preliminary statistics use reliable sources, and revisions are usually small.	MT. Establish and disseminate a calendar for GFS revisions. Disseminate revision studies.
5. Accessibility
	5.1 Data accessibility				X		The TOFE is presented in a single summary table. There is no supplementary table with more detail, and the TOFE is not accompanied by any commentary or graphical presentations. The GFS are not released in accordance with a preannounced schedule.	ST. Add comments and graphs to the published statistics, explaining and illustrating changes and trends. Publish more detailed TOFE. MT. Prepare and publish a functional classification of expenditures, at least annually. Adopt and publish a TOFE dissemination calendar.
	5.2 Metadata accessibility				X		The metadata in English available at the IMF’s DSBB are not periodically updated. No metadata in French are made available to the public.	ST. Update and disseminate the GDDS metadata in English and disseminate locally the GDDS metadata in French. Prepare and publish a technical note on TOFE sources and methods.
	5.3 Assistance to users				X		The GDDS metadata indicate the persons responsible for the statistics, but the metadata are not available in French. No catalog of publications is available.	ST. Include in the published TOFE the name of a contact person for obtaining clarifications or supplementary information. MT. Prepare and publish a catalog of publications (indicating those available and how to obtain them).

0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

Chad is one of six member countries of the Monetary Union of Central Africa (UMAC, for Union monétaire de l’Afrique centrale). The Bank of Central African States (BEAC, for Banque des États de l’Afrique centrale) serves as the Central Bank of UMAC, and its currency is the CFA franc. The Research Department (DET, for Direction des études) at BEAC headquarters in Yaoundé and the National Directorate for Chad (DNT, for Direction nationale pour le Tchad) jointly collect the source data, process them, and disseminate the monetary statistics for Chad. No other agency is involved in compiling the monetary statistics for Chad.

The BEAC’s responsibility for compiling monetary statistics is clearly defined in Article 22 of its charter (Title II), which stipulates that the central bank is to “centralize bank risks in member countries, participate in preparing their balance of payments, and compile the monetary statistics.” To this end, “it is authorized to collect, from the national authorities, banks, finance companies, and economic agents of member countries that are required to provide them, all data useful for the conduct of monetary policy, and for ensuring the security of bank and financial transactions” (Article 25). Articles 56 and 58 of the BEAC charter provide that “the Governor shall publish monthly balances for the Bank” and that “the Bank shall prepare and publish a statement of its accounts at the end of each month. It shall also prepare a balance sheet at the end of each fiscal year.”

Decree 20/PRE.T of January 19, 1965 regulates the bank activities in Chad; it constitutes the regulatory framework on the basis of which the BEAC collects, produces, and disseminates statistics.

The monthly BEAC survey for Chad is prepared jointly by the DNT and headquarters. Compilation of this report is done in three stages:

1) the DET sends to the DNT (by telex) the provisional accounting data for transactions performed by BEAC headquarters for the account of Chad, including transactions with the IMF.

2) upon receipt of these data, the Statistics and Monetary Analysis Unit (SSAM, for Service des statistiques et analyses monétaires) of the DNT prepares the BEAC balance sheet for Chad, recording all transactions conducted by headquarters and by the DNT.

3) the BEAC balance sheet for Chad is sent to the DET for verification and validation.

In compiling the monetary statistics, the BEAC also uses the monthly balance sheets and schedules prepared by banks and finance companies (altogether known as établissements de crédit) for supervisory purposes of the Banking Commission of Central Africa (COBAC, for Commission bancaire de l’Afrique centrale). The BEAC sends these monthly returns to the DNT in four copies, one of which is for compiling the monetary statistics. The return forms are standardized and identical for all banks and finance companies of the zone, whatever kind of license they may have.

0.1.2 Data sharing and coordination among data-producing agencies are adequate

The BEAC is the only agency authorized to compile monetary statistics and disseminate them to national and international users. For collecting source data from depository corporations other than the central bank, the BEAC is assisted by the COBAC, which establishes the reporting forms and the nature of the data to be reported. For those respondents that still submit their monthly returns on printed forms, the BEAC collects them at its various offices and transmits a copy to the COBAC. For the increasing numbers of respondents who are now using the COBAC-developed and managed centralized electronic reporting system,⁸ the COBAC transmits to the BEAC the data from each respondent that it needs in the exercise of its functions, including compilation of the monetary statistics. There is effective cooperation between the BEAC and the COBAC in both cases.

Depository corporations with reporting obligations to the BEAC and the COBAC are required by law to declare their monthly balance sheets and schedules in a manner consistent with the regional chart of accounts for banks and finance companies (PCEC, for Plan comptable des établissements de crédit). During the development of the Cerber system, the PCEC was updated and the reportable data were consequently revised. Banks and finance companies in the region were involved in these reforms, although only when the process was at a fairly advanced stage.

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

The SSAM is responsible for compiling and analyzing the monetary statistics. The SSAM processes the data for compiling statistics with software that allows its agents to obtain an overview of the balances of all commercial banks, as well as summarized individual balances for control and analysis purposes.

The SSAM staff responsible for compiling the monetary statistics are advised of the importance of maintaining the confidentiality of the data and information they handle and of the disciplinary penalties to which they are liable. These officers, like all BEAC personnel, are subject to the provisions of the personnel bylaws for senior officers. They are subject as well to the general provisions and conditions of the annex to the BEAC work contract, which provides that “employees are bound by professional secrecy” and “any breach of professional secrecy may be punished by penalties ranging up to dismissal without compensation” (Articles 31.7 and 31.8).

The only staff who have access to the monetary statistics database are those responsible for overseeing the source data and compiling the statistics. That database is protected by a password. Only aggregate statistics covering at least three reporting agencies may be released to the public. In publications, the only identifiable transactions are those of the central bank, the Treasury, and the Postal Checking Center (CCP, for Centre des chèques postaux), in recognition of their specific delegated responsibilities. For this reason, disclosure of their transactions does not breach the rules of statistical secrecy.

Individual respondents’ documents are filed and protected. There is strict control over access by outsiders to DNT premises: Presentation of a personalized badge is required. The DNT premises that house the SSAM are surrounded by a security fence.

Reporting forms do not spell out the rights and obligations of respondents or the purpose for which the information is collected. Nevertheless, the banks are aware that data they transmit are used for purposes of banking supervision and preparation of monetary statistics. Providers of source data do not receive in return any statistical information that would allow them to compare their own activity against that of the banking sector as a whole.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

Article 8 of Decree 20/PRE.T of January 19, 1965 and Article 36 of the annex to the convention on harmonization of bank regulations in UMAC stipulate that banks and finance companies must provide the central bank with information, clarifications, and justifications needed for the conduct of its delegated responsibilities. The long-standing 30-day deadline within which the commercial banks were to submit their monthly returns has been reduced to 15 days. Penalties may be imposed for failure to respect this deadline: a fine of 100,000 CFA francs per day for the first 15 days of delay and 300,000 CFA francs per day thereafter (Article 48 of the above-mentioned annex). The SSAM reports that it consistently sends out reminders to banks that are late with their reports and that it has recently imposed penalties on some banks.

Although bank regulations do not apply to the Post Office (Article 2 of the above-mentioned annex), the CCP is required to provide the BEAC with any information deemed useful to its functions. A bilateral agreement exists between the BEAC and the CCP governing the transmittal of such data.

The information collected by the DNT seems generally to be satisfactory and sufficient to allow the SSAM to compile the monetary statistics.

The operations of microfinance institutions in Chad are not currently included in the monetary statistics. This type of activity is expanding rapidly, with the resulting creation and expansion of networks of savings and loan cooperatives. The authorities will need to respond to this trend by imposing a periodic reporting requirement. Thus, the transactions of these organizations can be included in the monetary statistics, to the extent that they meet the definition of depository corporations pursuant to the Monetary and Financial Statistics Manual (MFSM).

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

The SSAM is staffed by a unit director, who has a university degree and has held the position one year, and by a middle-ranking officer, who has been on the job 10 years. If a member of the team is absent, an officer from the Credit and Monetary Markets Unit of the DNT can be seconded to perform the work of compiling the monetary statistics.

SSAM staff work with a computer connected to the DNT network. Processing of the source data and preparation of the statistics are fully automated. The software developed by BEAC is used by the national directorates in the six member countries of UMAC, as well as at headquarters.

There is no specific budget for financing BEAC’s statistical work.

0.2.2 Measures to ensure efficient use of resources are implemented

The human, financial, and material resources available to the monetary statistics team at DNT are decided by headquarters, which maintains consistency between assigned responsibilities and resources in the six national directorates of the BEAC and at headquarters.

As is the case with other personnel, the performance of monetary statistics staff is appraised annually by the immediate superior. Senior management of DNT and headquarters then review this appraisal.

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

The BEAC does not consult users about the nature and quality of the monetary statistics it prepares. No opinion survey of monetary statistics users has ever been conducted. Nor does the BEAC take any initiative to identify new needs that users might express.

0.4.1 Processes are in place to focus on quality

Monetary statistics staff are made aware of the importance of quality in their work. Banks and finance companies complete their monthly returns in accordance with standard formats designed by the DET on the basis of accounting standards. Those formats are contained in an internal technical document entitled Codification of monetary statistics, which displays model balance sheets and schedules relevant for monetary statistics compilation. This Codification also describes the procedures to be applied by SSAM for collecting, compiling, controlling, and disseminating the monetary statistics, as well as the relevant responsibilities of headquarters and the national directorates. This methodological reference document, developed by headquarters for all the national directorates, guarantees the quality and homogeneity of procedures across time and space.

All new employees assigned to SSAM have immediate access to a complete procedures manual, and the quality of their work therefore will not depend on the (perhaps incomplete) instructions left behind by their predecessors. However, the Codification will have to be thoroughly revised once the new PCEC is applied to the source data for the monetary statistics. At the same time, the BEAC has undertaken a revision of its own chart of accounts, and this will require a further amendment to the Codification for purposes of producing a new monthly survey.

0.4.2 Processes are in place to monitor the quality of the statistical program

The BEAC has working groups made up of headquarters and DNT staff to review the methodology used in compiling the monetary statistics.

The COBAC instituted a new PCEC on February 15, 1998, and commercial banks and finance companies are supposed to apply it for recording their transactions in the Cerber system (Cerber is a French acronym for “collection, exploitation, and feedback to the banks and financial corporations relating to regulatory reporting”). This new system, developed by COBAC, reorganizes information channels between the banks and the COBAC Secretariat.

The Cerber system works as follows:

1. Commercial banks and finance companies no longer prepare their periodic returns in hardcopy.

2. They send an electronic file to the COBAC Secretariat with all the necessary basic information, drawn from their accounts in accordance with a standardized structure and codification.

3. The COBAC automatically checks the data transmitted and rejects any transmission that contains anomalies (banks and finance companies have the checklist).

4. Once the data are validated, the COBAC prepares the regulatory returns in hardcopy and sends them to each respondent.

5. Respondents send the COBAC an acknowledgment of receipt of these returns, signed by a manager in certification that the information communicated is correct.

6. The COBAC provides BEAC users (the national offices, the DET, the Credit and Money Markets Division at headquarters) with these periodic returns in the form of an Excel file, by e-mail.

Beginning soon, the monetary statistics are supposed to be compiled from data gathered through the Cerber system. The definitive switch will take place over a transitional period during which the SSAM will have to compile two versions of the monetary statistics: (1) one version based on the balance sheet and two schedules currently used (Statement A, detailed monthly balance sheet; Statement B, loans granted to clients; and Statement D, client deposits) and (2) the other version based on the data produced by the Cerber system. This dual set of monetary statistics will show the impact on those statistics of adopting the new PCEC and the Cerber system. As of the end of June 2005, the Cerber system was not yet in effect in Chad.

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

The procedures for compiling the monetary statistics are fully described in the Codification of monetary statistics issued by the DET, and BEAC staff adhere to them scrupulously—to the point, indeed, where the feeling of confidence this generates may at times affect the critical mindset required to validate the source data or the resulting product. On occasion, the BEAC has published national monetary statistics with major gaps in the series—gaps subsequently found to result from clear errors in the source data or from improper assignment of data in the course of processing.

In May 2002 the BEAC hosted a seminar in Libreville for monetary statisticians from the six national directorates and headquarters, intended to establish a work program for implementing the MFSM methodology. As of May 2005, however, no movement at all has been made in this direction.

Finally, the BEAC does not actively seek out the reactions or preferences of official users or of the general public in relation to monetary statistics. Thus in recent years, it was only at the insistence of agencies external to BEAC that significant gaps were resolved between the net bank financing position of central government (NPG) as calculated by the MEF and as calculated by the DNT.

1.1.1 Statistics are produced on an impartial basis

The BEAC compiles monetary statistics independently of national governments. The Governor, the Vice Governor, the Secretary General, and all employees of the Bank are bound, while in office, to respect the principles of independence and neutrality inherent in their duties (Article 46 of the BEAC Charter).

The National Director appoints SSAM officers; he selects appointees among staff available in the DNT. The BEAC recruits staff with academic backgrounds at the regional level, either by competition or by written application at the decision of BEAC, after examining their candidacy qualifications. New recruits receive one year of wide-ranging instruction at the headquarters training center. They are then given working assignments for two or three months in a national office, and those whose appraisal ratings are better than 12 out of 20 are assigned to positions in the national directorates or at headquarters.

BEAC staff are expected to pursue their professional development. They are encouraged to produce papers and studies, some of which will be published in the BEAC’s bulletin, Études et Statistiques. They have the opportunity to take training assignments and to attend seminars, conferences, and meetings with other experts.

The Governor of the BEAC, who by convention is a Gabonese, is appointed by the conference of CEMAC heads of state, on the recommendation of the BEAC Board of Directors, for a renewable term of five years. He may, however, be relieved of his duties if he does not fulfill the conditions necessary to the exercise of those duties, or for grave misconduct.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

The BEAC, completely independent from national governments, is the only agency authorized to compile and disseminate the monetary statistics, and no other agency is involved in the choice of sources and of techniques for compiling those statistics. The DET at BEAC headquarters makes the main decisions on the production process of monetary statistics, based on technical considerations.

The BEAC generally disseminates the monetary statistics as soon as they become available. The authorities declare that observed delays are wholly attributable to lack of diligence on the part of BEAC and should not be interpreted as reflecting any intent to withhold data or to alter the information they convey.

The source data provided by the targeted population of respondents are sufficient for producing the monetary statistics but are not in accordance with most current international standards (see paragraph 2.1.1 below). Financial corporations communicate these data in a standard reporting format as described in the Codification of monetary statistics. BEAC obtains the data through an exhaustive collection exercise conducted at the date accounts are closed, for all the financial corporations covered by the current scope of the monetary statistics. The reporting format had not been revised since it was established in 1985 until when the new PCEC and the Cerber system were developed in the late nineties.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

The BEAC reports that it has never detected any erroneous interpretation or misuse of its monetary statistics. However, the BEAC does not make public the kind of background information that might forestall such risks. For example, BEAC publications make no explicit mention of the fact that liquidated banks are excluded from the scope of the monetary statistics. Thus the liquidation of the national savings bank (CNE, for Caisse nationale d’épargne) and its deletion from monetary statistics coverage have never been documented. Also, the CNE remains on the list of financial corporations in Chad, as published in the BEAC’s Études et Statistiques, with an indication that its accounting data are unavailable. The BEAC does not have a formally established public relations unit, such as a press office or external communications section.

1.2.1 The terms and conditions under which tistics are collected, processed, and disseminated are available to the public

The BEAC charter, the bank law, and a methodological note on monetary statistics are posted at the BEAC website.

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

Government officials, including those of the MEF, have access to the statistics before they are publicly released. This fact is not reported to the public.

1.2.3 Products of statistical agencies/units are clearly identified as such

All publications relating to the monetary statistics are issued under the seal of the BEAC, without explicit reference to the DNT. In the GDDS metadata published at the IMF website the names of the persons responsible for the monetary statistics are indicated. This information was not updated, however, when the head of the SSAM was replaced.

When the BEAC publishes statistics it did not produce, it indicates the data source.

1.2.4 Advance notice is given of major changes in methodology, source data, and statistical techniques

The SSAM points out that the move to the MFSM methodology will constitute a significant innovation in current classification practice, especially in the broadening of institutional coverage and the compilation of data in terms of stocks and flows. These methodological changes should, when implemented, be the subject of advance notice to the public, to be handled by the DET. At the end of May 2005, however, the likely impact of the move to the new PCEC and the Cerber system had not been announced to monetary statistics users.

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

In accordance with the BEAC’s personnel bylaws, staff are subject to professional secrecy under penalty of sanctions. At the time of their recruitment, employees are made aware of the required ethical standards, to which they must commit themselves in writing. There is no interference from outside the BEAC in the work of the DET and the national offices in compiling the monetary statistics.

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

The analytical framework used by the DNT in compiling the monetary statistics is consistent with the concepts and principles of the IMF’s Guide to Money and Banking Statistics in International Financial Statistics (Guide). It is not, however, consistent with the MFSM methodology.

The BEAC has not prepared any document indicating the main discrepancies between the current methodology and that of the MFSM, and it has not really taken any steps toward the MFSM methodology.

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

The scope will be more consistent with internationally accepted standards when it covers microfinance data.

The financial sector in Chad comprises a central bank, represented by the DNT, seven commercial banks, the CCP, savings and loan cooperatives that provide microfinance, two insurance companies, and an insurance broker. The monetary statistics for Chad cover the central bank, the commercial banks, and the CCP, all of which are depository corporations as regards the MFSM methodology.

On the other hand, the savings and loan cooperatives, although they are depository corporations, are not covered in the monetary survey, and their operations are not included in the monetary statistics of Chad. This gap reflects that fact that periodic balances are not collected from these cooperative networks. The CEMAC has decided to regulate the microfinance sector and to require regular reporting, at a frequency that has yet to be determined. Because the DNT has no estimate of the volume of transactions by microfinance institutions, it is difficult to assess how severely this deficiency affects the monetary survey.

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

Several classification and sectorization systems are not consistent with internationally accepted standards, guidelines, or good practices.

Transactions with nonresident units are classified as foreign assets and liabilities. The residency rule, used to distinguish between the reciprocal claims and liabilities of residents and nonresidents, is consistent with the notion of residence in the Balance of Payments Manual, fifth edition (BPM5).

Nonfinancial sectors are divided into (1) general government, (2) public nonfinancial corporations, and (3) the private sector. The sectorization of public units is not consistent with international classifications. The set of monthly returns includes, in particular, two schedules: “breakdown of transactions with government and public agencies” and “breakdown of transactions with public enterprises.” As a guide to respondents, these forms are distributed with the left-hand column already completed, listing the public entities of the respondent’s country. In principle, the exhaustive listing of governments, public agencies, and enterprises can only enhance the reliability of the commercial banks’ declarations. Yet this will only be so if the lists are regularly updated (at least once a year) to reflect any change in the status of units—something that is not done. The current lists, perhaps dating as far back as the Codification of monetary statistics drafted in 1985, are severely out-of-date and thus of poor relevance.

Assets and liabilities are overly aggregated. In the central bank survey and in the table “Avoirs extérieurs nets du système monétaire au sens large” (“Net foreign assets of the monetary system-broad money”), the assets are not broken down by type of financial instrument.

The NPG covers all transactions of general government, whereas it should include only the central government.

Consistent with the Guide and the MFSM, security repurchase agreements are treated as collateralized loans rather than as outright sales of securities.

2.4.1 Market prices are used to value flows and stocks

Financial assets are valued at market prices.

Assets and liabilities denominated in foreign currencies are valued at the average market exchange rate for the last working day of the month, consistent with the Guide and the MFSM.

The stock of monetary gold is valued at market prices. The countervalue of gold holdings in CFA francs is adjusted every six months (June and December), in light of the London market price for the last working day of the six-month period, consistent with the MFSM.

The CFA franc equivalent of SDR holdings is adjusted once a year, in May, on the basis of the rate set by the IMF on April 30, whereas international standards recommend a monthly revaluation.

The BEAC does not compile statistics based on flows, whereas the MFSM recommends that data should be presented not only as stocks but also as flows, divided into three components (transactions, revaluations, and other changes in the volume of assets).

Nonperforming claims are reported net of specific provisions for losses, whereas the MFSM recommends reporting at book value.

2.4.2 Recording is done on an accrual basis

The new PCEC recommends that all transactions be recorded on the day the payment order (ordonnancement) is issued. Banks and finance companies are supposed to prohibit the recording of entries ex post facto after the close of business on a given business day. Only on an exceptional basis should they allow a supplementary period for recording transactions that could not be recorded on the proper date. This practice is consistent with the MFSM and the Guide.

Accrued interest (i.e., interest accumulated but not yet due) on loans and deposits is recorded in specific accounts known respectively as “associated claims” and “associated debts.” This practice is consistent with the Guide, which calls for accrued interest on loans and deposits to be recorded as individual items under “other assets” or “other liabilities.” By contrast, the MFSM recommends that accrued interest be incorporated into the financial asset that generated it.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

Data on assets and liabilities are collected on a gross basis, and there is not supposed to be any netting of asset and liability items. This practice is consistent with the Guide and the MFSM.

In BEAC publications, some statistical tables, such as those on the net position of government and net foreign assets, show net positions, accompanied by the underlying gross data. These tables are presented as follows:

• (i) Net position of government =

  • Net position of government vis-à-vis the BEAC

    • Claims

    • - Liabilities

  • + Net position of government vis-à-vis the IMF (IMF loans) + Net position of government vis-à-vis the banks

    • Claims on government

    • - Deposits of government

• (ii) Foreign assets of the monetary system =

  • Gross foreign assets of the BEAC (gold, SDRs, reserve position with the IMF, operating account with French treasury [credit balance calculated by convention for Chad], others)

  • + Foreign assets of banks

  • - Gross foreign liabilities of the BEAC (drawings from the IMF, operating account with French treasury [debit balance calculated by convention for Chad], others)

  • - External liabilities of the banks (postal debt, others)

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

Banks and finance companies, established in member countries of the UMAC, use a chart of accounts that was overhauled in 1998 with the new PCEC. Pursuant to the implementation of the Cerber system, in which most UMAC member countries, but not Chad, actively participate, the regulatory time limit for submitting monthly balances to the COBAC should be reduced by half, from 30 days to 15 days.

Work on the PCEC began in 1993, involving many meetings among all the government departments and professional organizations concerned. The BEAC, however, did not fully participate in this process and has had little involvement in implementing the project. As a result, it would seem that the balance sheet and schedules will not initially include all the necessary or desirable data for the compilation of monetary statistics in accordance with the MFSM. For example, in the Cerber system, the sectorization of public entities has not been revised and still reflects the out-of-date lists.

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

The source data are not completely consistent with the MFSM classification of financial assets and valuation principles. Some of the source data from banks and other financial corporations do not come with a sufficient degree of breakdown by sector and by financial instrument for classifying assets and liabilities in accordance with the MFSM and even the Guide.

An adjustment procedure succeeds only partially in correcting the amount of banknotes in circulation, as it shows up on the liability side of the DNT. Because banknotes issued by each of the national directorates in UMAC member countries circulate freely,⁹ the amount of actual currency circulating in Chad cannot be determined simply as the difference between the total of banknotes issued in Chad and the total of banknotes returned to BEAC offices in Chad. When banknotes are presented by the commercial banks to BEAC offices they are sorted by country, using their distinctive lettering, and are then repatriated to the national directorate of the issuing country. This repatriation produces a decrease in the amount of outstanding banknotes on the balance sheet of the national directorate in the country receiving those repatriated banknotes. At the same time, the amount of banknotes outstanding on the balance sheet of the national directorate of the country sending the repatriated banknotes increases by the same amount.

Apart from the repatriation procedure, no other adjustment is made to the domestic circulation of currency. Banknotes from all CEMAC member countries are in the vaults of the BEAC, awaiting sorting by country; other banknotes are held by the commercial banks; and still others are in circulation. However, no adjustment is made to domestic circulation figures on any of these three scores such as would improve the reliability of national monetary statistics.

The lack of precision in the estimate is all the greater when it is recognized that, within CEMAC as a whole, currency outstanding represents about 40 percent of the means of payment and more than 25 percent of the money supply, and cross-border flows of banknotes are very high. The national monetary aggregates calculated by the BEAC are therefore underestimated for those countries that are net importers of banknotes through cross-border flows, and they are underestimated for those countries that are net exporters through those flows.

It is reasonable to assume that a country’s status as a net importer or exporter of banknotes persists over time, since cross-border flows of currency reflect structural trade flows between countries. The results of the banknote sorting by the BEAC between April 2004 and March 2005 suggest that (1) 70 percent of the banknotes issued by the DNT are circulating within Chad, and 30 percent of them in Cameroon, and (2) 70 percent of the banknotes circulating in Chad were issued by the DNT, while 30 percent of them come from other CEMAC member countries, in particular Cameroon (20 percent) and the Central African Republic (6 percent). For these reasons, the money supply as calculated for Chad carries a degree of uncertainty that is not negligible.

The commercial banks send their monthly balance sheets and attached schedules to the SSAM on printed forms. Before SSAM staff enter the data, they perform consistency checks on the general balance of assets and liabilities, on the consistency of the attached schedules with the balance sheet, and on the consistency of the amount posted for refinancing transactions with other banks on the interbank market.

3.1.3 Source data are timely

The banks generally respect the regulatory deadline (30 days) for transmitting their monthly returns. If they are late, they receive a reminder from the SSAM and are assessed a penalty.

The balance sheet for the DNT is prepared at headquarters. As of May 2005, the last available balance sheet was for March 2005.

Data for the CCP are often transmitted late because its nationwide network is so broad. The latest available data are often carryovers. They are not systematically revised when new figures become available.

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

The DNT does not use data from censuses or sample surveys in compiling the monetary statistics. When the SSAM detects inconsistencies or movements that deviate from trends, it contacts the reporting corporation for explanations and corrections if necessary.

SSAM also verifies and cross-checks the data by using information available in other DNT units, in particular the Money Market Unit.

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

The DNT enters data manually into the computerized monetary statistics database. Thereafter, data processing is entirely automated in Excel, which reduces the risk of errors. Successive aggregations and consolidations undertaken in the process of compiling the statistics respect general principles of accounting (balanced assets and liabilities). No estimations or adjustments to the source data are required in compiling the monetary statistics.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

The BEAC does not publish seasonally adjusted data series and does not perform any estimations or carryovers in the case of missing source data or nonresponse, except for the CCP. In the case of CCP, data are carried forward from one month to the next. No specific procedure exists for adjusting data or for estimating missing or late data.

3.4.1 Intermediate results are validated against other information where applicable

Staff may perform cross-checks on the basis of information (such as data regarding interbank market transactions) available in other DNT units. However, no comprehensive secondary data are available that staff could use to verify the monetary statistics.

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

No reconciliation is done for data on stocks and flows. Staff compile the monetary statistics exclusively on the basis of data expressed as stocks only.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

Generally, in the case of a significant or unexplained fluctuation in the monetary statistics, the DNT contacts, directly or by letter, the declaring institution that supplied the data that caused the fluctuation. However, over the last years, significant unexplained fluctuations have been detected only well after dissemination of monetary statistics to the public or their submission to the IMF, even though they should have been identified and cleared during the validation phase before dissemination.

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

The process of compiling the monetary statistics does not entail any systematic revision. The monetary statistics, deemed definitive when released, will be revised only by the occasional correction. The process of compiling the monetary statistics is stable, and error corrections are infrequent.

4.1.1 Periodicity follows dissemination standards

The central bank survey and the survey for other depository corporations are compiled monthly; thus, their periodicity is consistent with the GDDS.

4.1.2 Timeliness follows dissemination standards

The GDDS recommends that the central bank survey be disseminated within two months after the reference month, and the broad money survey within three months after the reference month. At the BEAC website, all these data are generally available between two and three months after the reference month. This means the GDDS timeliness standard is often met for the broad money survey but almost never for the central bank survey. The deadlines are almost never respected when it comes to publication in Études et Statistiques, which for a long time now has been published more than a year late.

4.2.1 Statistics are consistent within the dataset

Broad internal consistency is present between the various aggregated surveys and the detailed tables for certain aggregates. This consistency relates only to stocks, since flow data are not available.

Nevertheless, the consistency of reciprocal claims and liabilities between depository corporations and other financial corporations’ claims cannot be verified, since other financial corporations are not currently covered by the monetary statistics of Chad. It is not possible to reconcile the monetary aggregates with the TOFE to verify the net position of government calculated for each dataset, because the NPG calculated by the BEAC is now treated as a source input in preparing the TOFE: Through this device, the considerable discrepancies observed in the past have disappeared.

Also, among some of the monetary statistics tables published in Études et Statistiques, some discrepancies appear between (1) the amount of the banks’ foreign liabilities (liabilities side of the banks’ balance sheet) and that shown in the external position; (2) the total of quasi-money deposits from the banks survey and that from the broad money survey; and (3) the amount of quasi-money (from the broad money survey) and the sum of its components. To date, the BEAC has not taken any steps to reconcile these discrepancies.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

The statistics are available as monthly data, in accordance with a methodology that has remained unchanged since 1994 in the BEAC’s monetary statistics software used by the DNT. The historical series are not systematically adjusted following changes in the methodology, source data, or statistical techniques. However, such changes are very rare.

BEAC publications do not address the possible causes of breaks in the series. Thus, the changes to the institutional coverage of the monetary statistics resulting from liquidation of the CNE and transformation of the Banque de développement du Tchad (an other financial intermediary) into the Commercial Bank Tchad (a depository corporation) have not been indicated in the publications.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

There is no guarantee of consistency between the government finance statistics and the monetary statistics. The existing discrepancies reflect primarily the differing institutional coverage of the concepts of government employed respectively by the BEAC and the Treasury.

4.3.1 Revisions follow a regular and transparent schedule

The process of compiling the monetary statistics does not entail a preliminary version or any revision procedure. The monetary statistics are always deemed definitive when released to the public.

4.3.2 Preliminary and/or revised data are clearly identified

Not applicable.

4.3.3 Studies and analyses of revisions are made public (see also 3.5.1)

Not applicable.

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

The monetary statistics are presented only in the form of tables. These tables are not accompanied by any supplementary information to facilitate their analysis. The monetary statistics are not accompanied by any graphs or by any seasonally adjusted series that would make it easier to detect monetary trends.

A few succinct comments are published in Études et Statistiques, dealing primarily with monetary trends within the CEMAC as a whole, rather than with any trends observed in each member country. For the more recent monetary statistics, a monthly note on the broad money survey of each member country is available at the BEAC website.

5.1.2 Dissemination media and format are adequate

The BEAC disseminates the monetary statistics in three publications:

• Études et Statistiques is in principle a monthly bulletin, but in fact its frequency varies and it sometimes appears more than a year late. The last version was dated April 2004, containing monetary statistics to that date.

• The Monthly Bulletin of Monetary Statistics is meant to be published much more promptly than Études et Statistiques. However, this bulletin, designed to deal only with the monetary statistics, has been printed only once in early 2002 with monetary statistics as of end-December 2001.

• The BEAC Annual Report dates last from 2003.

These three publications are in French, and they have no English translation.

The BEAC website is much more up-to-date than its hardcopy publications. At the end of May 2005, the most recent monetary statistics posted at the website were for March 2005.

5.1.3 Statistics are released on a preannounced schedule

There is no preannounced schedule for the release of monetary statistics.

5.1.4 Statistics are made available to all users at the same time

The monetary statistics are made available to all users at the same time, except that some official agencies, such as the MEF, receive the monetary statistics before their publication.

5.1.5 Statistics not routinely disseminated are made available upon request

Statistics not routinely disseminated can be made available upon request, but the DNT has not received specific requests from external users recently.

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

A summary methodological note concerning the analytical accounts of the banking sector and of the central bank—concepts, classifications, recording basis, characteristics, and definitions—is available in the GDDS metadata at the IMF’s DSBB, in English. The website does not provide the French version of these files.

The BEAC publishes a methodological note, in French, on the compilation of the monetary statistics, in its bulletin Études et Statistiques and at its website.

5.2.2 Levels of detail are adapted to the needs of the intended audience

The published information apparently meets the requirements of the general public but is not sufficiently detailed for more informed users.

The French version of the metadata, made available by the IMF to the authorities, is not posted at any website, nor is it published in any document.

5.3.1 Contact points for each subject field are publicized

Monetary statistics users may contact the responsible compiler, whose name is indicated in the GDDS metadata published at the IMF website. However, this contact information was not updated when the new head of the SSAM was appointed.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

The DNT offers a documentation service and a library, where the public may consult available documentation on-site and obtain information on the monetary statistics. Users may subscribe to certain publications. The publication list is available at the BEAC website, but the modest prices per publication and subscription fees are not indicated. The Études et Statistiques bulletin does not mention the unit price, the subscription fees, or the procedure for requesting the service.

Table 6:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Monetary Statistics

(Compiling Agency: Bank of Central African States – BEAC, National Directorate for Chad – DNT)

Table 6:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Monetary Statistics

(Compiling Agency: Bank of Central African States – BEAC, National Directorate for Chad – DNT)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element	NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
		O	LO	LNO	NO
0. Prerequisites of quality
0.1 Legal and institutional environment		X
0.2 Resources		X
0.3 Relevance				X		There is no periodic or structured consultation to obtain users’ opinions on the monetary statistics	Medium-term (MT). Survey a representative sample of data users for their opinions on the quality of monetary statistics and their emerging needs.
0.4 Other quality management			X			The Cerber system is not yet operational in Chad.	Short-term (ST). Activate the Cerber system using the tests and verifications provided by COBAC.
1. Assurances of integrity
1.1 Professionalism		X
1.2 Transparency			X			No policy is in place to inform the public in advance of major changes in methodology. The public is not informed of the access of government officials to monetary statistics prior to their release.	ST. Release advance notice of methodological changes. Inform the public that government officials have access to monetary statistics prior to their release by implementing an adequate warning in BEAC’s publications and on its website.
1.3 Ethical standards		X
2. Methodological soundness
2.1 Concepts and definitions			X			There is no plan for transition to the MFSM.	ST. Prepare a program for transition to the MFSM.
2.2 Scope				X		The monetary statistics do not cover microfinance data.	ST. Compile promptly an inventory of micro finance institutions and collect their balances for inclusion in the monetary survey as of 2006.
2.3 Classification/sectorization				X		The sectorization of government entities is not consistent with international classifications.	ST. Update the list of government entities and sectorize them appropriately.
2.4 Basis for recording			X			Accrued interest on deposits and loans is not recorded as part of the financial assets to which they refer.	MT. Include accrued interest with the financial assets to which they relate, consistent with the MFSM methodology.
3. Accuracy and reliability
3.1 Source data			X			The DNT survey included in the monetary statistics is available much later than the aggregate survey for other depository corporations.	MT. Develop an application for preparing the DNT survey and send it electronically to the SSAM.
3.2 Assessment of source data		X
3.3 Statistical techniques		X
3.4 Assessment and validation of inter-mediate data and statistical outputs			X			It occurs sometimes that a significant or unexplained fluctuation in the monetary statistics is detected only after their release.	ST. Improve the validation process of monetary statistics before their release.
3.5 Revision studies	X
4. Serviceability
4.1 Periodicity and timeliness			X			The timeliness of dissemination of the central bank aggregates is not consistent with the GDDS.	ST. Shorten the time taken to compile the DNT survey.
4.2 Consistency				X		There are apparent inconsistencies between the monetary statistics tables. There is no reconciliation with the GFS as regards the net position of the government, or with the BOP as regards net foreign assets of the banking system.	ST. Establish an automatic system of verification and control between tables. Ensure permanent consistency between the monetary statistics and the GFS (net position of the government) and the BOP (net foreign assets).
4.3 Revision policy and practice	X
5. Accessibility
5.1 Data accessibility			X			There is no commentary on breaks in the series. The Études et Statistiques bulletin last appeared in April 2004. There is no advance release calendar. Government officials have access to monetary statistics prior to their general dissemination.	MT. Supplement publications of the monetary statistics tables with notes, commentaries, and graphs. Establish and publish a schedule for disseminating the monetary statistics. Release monetary statistics to all users at the same time.
5.2 Metadata accessibility			X			The GDDS metadata presented at the IMF website are not up-to-date.	ST. Transmit metadata updates to the IMF at least once a year.
5.3 Assistance to users			X			There is no catalog of publications. Access procedures for user assistance remain confidential.	ST. Publish a catalog of all existing publications. Advise the public of the user assistance available.

View Table

Table 6:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Monetary Statistics

(Compiling Agency: Bank of Central African States – BEAC, National Directorate for Chad – DNT)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element	NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
		O	LO	LNO	NO
0. Prerequisites of quality
0.1 Legal and institutional environment		X
0.2 Resources		X
0.3 Relevance				X		There is no periodic or structured consultation to obtain users’ opinions on the monetary statistics	Medium-term (MT). Survey a representative sample of data users for their opinions on the quality of monetary statistics and their emerging needs.
0.4 Other quality management			X			The Cerber system is not yet operational in Chad.	Short-term (ST). Activate the Cerber system using the tests and verifications provided by COBAC.
1. Assurances of integrity
1.1 Professionalism		X
1.2 Transparency			X			No policy is in place to inform the public in advance of major changes in methodology. The public is not informed of the access of government officials to monetary statistics prior to their release.	ST. Release advance notice of methodological changes. Inform the public that government officials have access to monetary statistics prior to their release by implementing an adequate warning in BEAC’s publications and on its website.
1.3 Ethical standards		X
2. Methodological soundness
2.1 Concepts and definitions			X			There is no plan for transition to the MFSM.	ST. Prepare a program for transition to the MFSM.
2.2 Scope				X		The monetary statistics do not cover microfinance data.	ST. Compile promptly an inventory of micro finance institutions and collect their balances for inclusion in the monetary survey as of 2006.
2.3 Classification/sectorization				X		The sectorization of government entities is not consistent with international classifications.	ST. Update the list of government entities and sectorize them appropriately.
2.4 Basis for recording			X			Accrued interest on deposits and loans is not recorded as part of the financial assets to which they refer.	MT. Include accrued interest with the financial assets to which they relate, consistent with the MFSM methodology.
3. Accuracy and reliability
3.1 Source data			X			The DNT survey included in the monetary statistics is available much later than the aggregate survey for other depository corporations.	MT. Develop an application for preparing the DNT survey and send it electronically to the SSAM.
3.2 Assessment of source data		X
3.3 Statistical techniques		X
3.4 Assessment and validation of inter-mediate data and statistical outputs			X			It occurs sometimes that a significant or unexplained fluctuation in the monetary statistics is detected only after their release.	ST. Improve the validation process of monetary statistics before their release.
3.5 Revision studies	X
4. Serviceability
4.1 Periodicity and timeliness			X			The timeliness of dissemination of the central bank aggregates is not consistent with the GDDS.	ST. Shorten the time taken to compile the DNT survey.
4.2 Consistency				X		There are apparent inconsistencies between the monetary statistics tables. There is no reconciliation with the GFS as regards the net position of the government, or with the BOP as regards net foreign assets of the banking system.	ST. Establish an automatic system of verification and control between tables. Ensure permanent consistency between the monetary statistics and the GFS (net position of the government) and the BOP (net foreign assets).
4.3 Revision policy and practice	X
5. Accessibility
5.1 Data accessibility			X			There is no commentary on breaks in the series. The Études et Statistiques bulletin last appeared in April 2004. There is no advance release calendar. Government officials have access to monetary statistics prior to their general dissemination.	MT. Supplement publications of the monetary statistics tables with notes, commentaries, and graphs. Establish and publish a schedule for disseminating the monetary statistics. Release monetary statistics to all users at the same time.
5.2 Metadata accessibility			X			The GDDS metadata presented at the IMF website are not up-to-date.	ST. Transmit metadata updates to the IMF at least once a year.
5.3 Assistance to users			X			There is no catalog of publications. Access procedures for user assistance remain confidential.	ST. Publish a catalog of all existing publications. Advise the public of the user assistance available.

0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

The Bank of Central African States (BEAC, for Banque des États de l’Afrique centrale) is responsible for preparing the external balance of payments (BOP) for CEMAC member countries.¹⁰

Decree 164-F of July 5, 1967 (Articles 4 and 5) gives the BEAC’s National Office for Chad (DNT) the authority to collect the data needed to establish the national BOP. That decree empowers the DNT to collect source data from businesses, governments, and international organizations, either directly or through banks and the post office.

Consequently, the DNT serves as the Secretariat for the BOP Committee (CBP)—the national body responsible for reviewing data and validating BOP statistics. The CBP was created by Article 2 of Decree 164-F. It is chaired by the Secretary-General of the MEF.

This prerogative was confirmed by Law 13/PR/99 of June 15, 1999, regulating statistics in Chad. Article 2 of that law makes the offices of the central bank that prepare the economic, monetary, financial, and BOP statistics members of the SSN.

This responsibility is also reflected in Article 22 of the BEAC Charter adopted on July 11, 2003, which provides that the central bank must centralize bank risks in member countries, participate in preparation of the BOP, and prepare the monetary statistics.

Within the DNT, the BOP Unit (SBDP, for Service de la balance des paiements) is responsible for establishing the BOP, while the Directorate of External Financial Relations (DRFE, for Direction des relations financières extérieures) at BEAC headquarters in Yaoundé oversees and coordinates the work of the national offices. It also maintains liaison with international organizations for supplying particular data and with the IMF on methodological issues.

0.1.2 Data sharing and coordination among data-producing agencies are adequate

Overall, existing mechanisms for data sharing and coordination among statistics-producing agencies are adequate. However, weaknesses appear in some of the source data, such as customs data, which make them unusable for the DNT.

The CBP reflects the public agencies that produce source data and statistics for the BOP. The agencies provide the SBDP with the most recent information available, for inclusion in the BOP.

Beginning in 2005, the BEAC has started holding, prior to its meeting with the CBP, an annual technical meeting with the principal directorates of the MEF and the INSEED. The meeting constitutes a framework for coordination among data-producing agencies. Participants exchange views on the data produced by the SBDP as well as on any reprocessing that is done before presentation of the draft BOP to the CBP.

Foreign trade statistics produced by the Directorate of Customs and Indirect Taxation (DDII) should normally be included in the BOP, but the DDII has been producing them too late for inclusion in the BOP. Thus the SBDP must itself estimate import and export data on the basis of surveys.

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

Article 13 of Decree 164-F of July 5, 1967 prohibits information collected by the BEAC from being used for purposes other than statistical. The BEAC is not authorized to disclose individual transactions or data in its possession. It guarantees the confidentiality of the information collected for BOP purposes and is banned from disclosing that information to any person or agency not involved in preparing the BOP statistics. The letters that accompany the mailed-out questionnaires mention the confidential nature of the data and the fact they are used exclusively for compiling BOP statistics.

During visits to businesses, SBDP agents remind respondents that the information provided will be used only in aggregate form and only for BOP purposes. Data collected by the SBDP through its compulsory surveys are covered by statistical secrecy, unless the individual or business explicitly authorizes disclosure. Consequently, the data may in no case be used for fiscal or economic control purposes.

This obligation of confidentiality is also stipulated in the staff rules and regulations of the BEAC, whose agents are subject to professional secrecy. They may neither report nor disclose any facts or information of which they become aware in the exercise of their duties, without prior authorization from the Governor. BEAC staff members who breach professional secrecy are subject to punishment, ranging up to dismissal without compensation.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

The obligation to report source data is confirmed by Law 13/PR/99 of June 15, 1999 regulating statistics in Chad (chapter II). That law establishes the BEAC’s authority clearly and without ambiguity, and no actual or potential conflict of responsibilities arises in preparing the external accounts.

Data collection is done in accordance with rules established by the BEAC. Respondents have 30 days to return the completed questionnaire. In practice, however, many questionnaires reach the BEAC late, having the effect of postponing CBP meetings and delaying publication of the BOP.

Law 13/PR/99 requires natural and legal persons to provide accurate responses to survey questionnaires and forms, within the stipulated deadlines. The Senior Statistics Council (CSS, for Conseil supérieur de la statistique) has the power to issue warnings and impose sanctions on persons violating these rules. The BEAC has not resorted to this device, however, preferring to work cooperatively with respondents and sensitize them to their responsibilities.

Recognizing the workload required of statistical correspondents, the BEAC helps them complete the questionnaires and invites them to contact SBDP agents to clarify any doubts they may have. BEAC agents also use telephone contacts and field visits to settle reporting problems.

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

With only three officers, including the section chief, the SBDP is inadequately staffed to compile and disseminate the BOP under ideal conditions. Moreover, two of the SBDP’s officers are subject to temporary transfer, as needed, to other sections of the DNT. The SBDP urgently needs the additional services of at least one middle manager and a secretary-accountant to process promptly the data collected and to produce the quarterly statistical statements needed for the conduct of economic policy. The turnover of staff in the SBDP is too high, and, as a result, staff assigned in this unit do not have sufficient professional training.

Working conditions and equipment in the office are satisfactory. The computing resources allocated to BOP statistics work are adequate. Each officer has a computer that processes data automatically using the BEAC’s BOP software, and the office is also equipped with three printers.

On the other hand, financial resources are not adequate. The SBDP does not have the means of transportation needed to collect data from respondents, and the travel allowances accorded to officers are minuscule in comparison with the amount of travel.

The BEAC is now developing a means of extending the software application to bank forms. This will allow bank statements to be processed automatically, thereby reducing the risk of error and delay. The banks will test data transmission using the new format in the first quarter of 2006.

0.2.2 Measures to ensure efficient use of resources are implemented

As with other DNT personnel, officers assigned to the SBDP undergo an annual performance review. However, managers do not specifically evaluate BOP work. Nor is there any costing of BOP statistics production. However, the new computer application mentioned above will allow each officer to tally the number of forms processed during the working day. This will also make it possible to measure personnel needs in light of the volume of information received for processing. The BEAC does not use outside experts to evaluate BOP compilation methodologies and procedures.

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

To date, no specific measures are in place to ensure that current statistics meet the needs of data users. Nor is there any structured and periodic process of consultation to review the usefulness of existing statistics and to identify emerging data requirements.

0.4.1 Processes are in place to focus on quality

The BEAC adheres to principles of reliability in BOP statistics. Thus, every officer assigned to the SBDP undergoes a practical training program. Provision is made for training assignments in another BEAC national directorate to improve officers’ familiarity with BOP operations. As well, SBDP officers may take part in outside training sessions, locally or abroad, relating to the external sector.

In line with the recommendations of the IMF’s Balance of Payments Manual, fifth edition (BPM5), the BEAC undertook a review in 1995 of its system for collecting source data and compiling statistics and has adapted its methodology. The BEAC has also introduced a common computer application in all national offices. The first version of the questionnaire was revised in light of observations from businesses, and the computer application was amended accordingly. The latest version installed in the DNT dates from May 2005.

A Guide for Users of the BOP Software and a technical document—describing report forms, questionnaires, and other source data submissions and possible content for each items herein—entitled Codification of the Balance of Payments are available to compilers.

0.4.2 Processes are in place to monitor the quality of the statistical program

The DNT is able to measure the response rate to its questionnaires using a module in the BOP software designed for this purpose. The SBDP revises the list of respondents annually to reflect the trend in national economic activity. On the other hand, despite its efforts to date, the DNT has been unable to organize regular meetings with the CBP, which should in principle take place nine months after the end of the reference period.

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

The CBP has the authority to consider the quality of the source data collected, their statistical processing, and the dissemination of results. It may issue an opinion on the methodology used and make recommendations on data sources and strategies for improving statistical production.

The BEAC communicates its preferences to members of the CBP, and their viewpoints are taken into account. The inclusion of the Directorate of Veterinary Services (DSV, for Direction des services vétérinaires) and Oil Directorate in the CBP, as announced by the authorities, should improve the reliability of source data on transactions monitored by these units.

1.1.1 Statistics are produced on an impartial basis

The provisions governing the production of statistics are consistent with the principles of professional independence. This concern is confirmed in Article 46 (paragraph 3) of the BEAC charter, which specifies that the Governor, the Vice-Governor, the Secretary-General (during their term in office), as well as all officers of the central bank must respect the principles of independence and neutrality inherent in their functions. These provisions ensure that the BEAC observes impartiality in the exercise of its responsibilities. The National Director is appointed and removed by the Board of Directors on the recommendation of the Governor, with the concurrence of the government of Chad. The National Director is appointed for an indefinite term.

BEAC officers are recruited through competition. Successful candidates must undergo a period of professional training of up to one year before they are confirmed. Depending on the position to be filled, a university degree is required. The BEAC encourages its staff to pursue ongoing training, either under its own sponsorship or at the officer’s initiative.

The BEAC actively promotes professionalism among its SBDP officers by providing or authorizing professional training in light of their capacities, the positions they occupy, and their own operational needs. Thus, officers newly assigned to the SBDP undergo a training period in one of the BEAC centers to familiarize them with the methodology and procedures for compiling the BOP.

With prior approval from headquarters, managers may apply to participate in training sessions, seminars, conferences, and meetings organized by international or regional organizations. However, middle-level managers currently assigned to the SBDP have not yet had the opportunity to take such training. The BEAC encourages its officers to publish studies in its monthly bulletin, Études et Statistiques. Officers may also take part in national meetings or workshops of interest to the BEAC.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

Most decisions concerning BOP statistics production are taken by the DRFE at BEAC headquarters, which coordinates all the work pertaining to this dataset within CEMAC.

All data collection and aggregation techniques reflect statistical requirements, specifically those of the BPM5. Internal methodological notes are prepared for the use of BEAC staff and consequently are not published.

A press release notifies the media about meetings of the CBP. BOP data are disseminated to users on the basis of statistical considerations.

The results of SBDP surveys are subjected to internal review before they are submitted for approval by the CBP.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

The BEAC reserves the right to comment publicly on statistics and to correct any errors in the interpretation of the statistics it produces. However, according to the DNT, BOP data have never been the subject of erroneous commentary in the media.

The BEAC uses its press office to ensure media coverage of its data.

1.2.1 The terms and conditions under which statistics are collected, processed, and disseminated are available to the public

BEAC publications do not disclose the regulatory provisions of the BEAC governing the collection, processing, and dissemination of data on the external sector. In particular, the BEAC Annual Report and the Balance of Payments of the Republic of Chad, Year N (BOP Annual Report)¹¹ contain only annotated statistics.

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

The MEF and INSEED, as suppliers of raw statistical data, have access to the provisional aggregates before their release, but this is not made public. The existence and membership of the CBP do not make, by themselves, the public at large aware of early governmental access to BOP statistics.

1.2.3 Products of statistical agencies/units are clearly identified as such

The BOP statistics produced by the BEAC are clearly identified as its product and are published under its seal, in accordance with its mandate. The DNT reports that, in the case of joint publication, the name of each producing entity is specified.

1.2.4 Advance notice is given of major changes in methodology, source data, and statistical techniques

Users of BOP statistics are not given advance notice of major changes in methodology, source data, and statistical techniques. However, when survey results are released, users are advised of any new statistical technique used. The innovations introduced by BPM5 were covered in a summary note to the BOP Annual Report, which explains the principal methodological changes. The same occurred at the time of the move to the Balance of Payments Manual, fourth edition (BPM4).

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

Ethical rules exist and are understood by staff. The behavior expected of central bank personnel is clearly specified in the staff rules and regulations of the BEAC, a copy of which is given to each officer upon appointment.

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

Since 1995 the BEAC has been compiling BOP statistics in accordance with the principles of BPM5. This holds with respect to the principle of double-entry accounting, the classification of goods, services, and income, and the embedding of the net residual in the errors and omissions item. The BEAC’s Codification of the BOP uses the concept of residency and economic interest for sectorizing natural or legal persons and their transactions with the outside world.

Transactions are recorded on the date on which an economic value is created, transformed, exchanged, extinguished, or is subject to a transfer of ownership. Transaction value is determined on the basis of the amount actually paid, except where there is special agreement among the parties, for example in the case of grants. The basic rule for foreign direct investment transactions is the 10 percent of ownership threshold. Direct investment transactions are recorded by the direction of the transaction: inward direct investment by nonresidents and outward direct investment by residents.

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

The scope of the BOP statistics is broadly consistent with BPM5 principles. The current codification covers all transactions between resident and nonresident institutional units as defined in the BPM5. International and regional organizations and embassies accredited to the country are not considered as residents in accordance with the BPM5 principles. Travelers, students, persons receiving medical treatment, athletes, cross-border and seasonal workers, as well as locally engaged staff of Chadian embassies abroad are deemed residents of their home countries.

The Codification of the BOP explicitly provides that any transaction between residents and nonresidents must be accounted for in the BOP statistics.

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

The classification and sectorization system used for the BOP statistics is broadly consistent with the BPM5 principles. Thus, institutional units are correctly sectorized, and transactions are classified in accordance with the standard components table of the BPM5. External public debt transactions are attributed to the public sector, while transactions by public enterprises are classified under “other sectors.”

2.4.1 Market prices are used to value flows and stocks

The valuation rules used for recording transactions are broadly consistent with the market valuation principles of the BPM5. Thus, goods transactions are valued on an f.o.b. basis, both for exports and for imports. However, for imports declared c.i.f., the amount is divided between f.o.b. goods (67 percent), freight (30 percent), and insurance (3 percent). The SBDP indicated that this key is obsolete and requires updating.

Transactions in foreign currency are converted to domestic currency, the CFA franc, using the transaction exchange rate, following the BPM5 principles. With respect to transaction estimates derived from stocks data, they are converted to CFA francs at the average exchange rate for the applicable period.

2.4.2 Recording is done on an accrual basis

Transactions are recorded in the BOP on an accrual basis. Thus, services are recorded when they are provided or received, loan drawings are entered at the time of actual disbursement, and loan repayments are entered at their due date. However, those transactions recorded through bank settlement reports are on cash basis.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

BOP recording conventions require that current and capital account transactions be reported and recorded on a gross basis, while financial account transactions be recorded separately by category (direct investments, portfolio investments, other investments), net of financial commissions and charges. These recording conventions are generally applied in producing the BOP for Chad.

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

The source data are largely obtained from comprehensive programs.

The data collection program used in CEMAC is appropriate for compiling the BOP statistics. It consists essentially of two sources: the standard or special questionnaires, and the bank forms. The standard or special questionnaires are designed to record transactions by resident businesses, government entities, embassies, nongovernmental organizations (NGOs), and international agencies operating in Chad. The bank forms cover external financial transactions by households and businesses.

BEAC headquarters reports to the DNT the following: (1) the external position of the monetary authorities, in particular the account for transactions and position with the IMF; (2) the external position of the commercial banks; (3) the statement of billets déplacés (displaced banknotes, i.e., resulting from intra-currency union movements—for further explanations, see Monetary Statistics, paragraph 3.1.2);¹² (4) documentation on transactions by the French Development Agency; and (5) commissions charged to the Chadian government. According to the SBDP, the last two items of information are no longer transmitted regularly to the DNT; requests have to be sent out if this information cannot be obtained from other sources.

The Investment Directorate (DI) and Debt Directorate (DD) report public debt transactions to the SBDP. The Directorate of Cooperation and Monitoring of Development Projects and Programs reports information on project grants, while the DSV of the ME provides volume data on exports of livestock, hides, and meat.

The SBDP surveys a large sample of businesses. It updates this list to reflect mergers, closures, and creations of businesses. The membership directory published by of the Chamber of Commerce, Industry, Agriculture, Mining, and Crafts is consulted during updates. The survey procedure is satisfactory in terms of concepts, but in practice the spontaneous response rate is low. The SBDP must then send out reminders and perhaps visit companies to help those complete questionnaires. The questionnaires for businesses as well as for other entities are sent out to respondents every three months. The responses are supposed to be returned to the SBDP within one month. For 2005, 126 economic agents will be surveyed, distributed as follows: 6 government departments and agencies; 53 industrial and commercial businesses; 7 airline companies; 4 hotels; 12 nongovernmental organizations; 27 embassies and international organizations; 11 banks and insurance companies; 3 forwarding agents; and 3 construction companies.

The reporting forms submitted by the commercial banks cover external financial transactions by households and businesses, including inward and outward transfers and manual exchange transactions (currency purchases and sales). Under the data collection system used within CEMAC, the commercial banks must report to the central bank all transactions conducted abroad—either for their clients or for their own account. These transfers may be handled through correspondent accounts or through the BEAC. Declarations also list manual exchange transactions.

The SBDP collects data on government financial transactions from administrative sources. These data are satisfactory as far as external public debt transactions are concerned: such transactions are detailed and classified by donor and by type of operation. This is not the case, however, with central government expenses, because data on diplomatic and other foreign expenses are late in arriving, and the SBDP must use budget provisions as an approximation.

As regards trade of goods, information obtained by the SBDP from the DSV is also inadequate, because of problems in identifying the exact number of livestock exported. Further, the SBDP does not supplement its own data on trade with that obtained through the computer software package Automated System for Customs Data (ASYCUDA) used by the DDII. Indeed, aggregate data produced through ASYCUDA are not more reliable than those collected through questionnaires returned to the BEAC by main operators. However, the former could be taken into account to inform the compilation process of trade data principally based on the latter.

The SBDP does not conduct any household surveys. Source data on households’ external transactions are collected via the bank reporting forms.

Overall, the SBDP considers that data collection programs are sufficiently open and flexible to take account of new developments affecting data sources. In practice, there are no meetings with the business community. However, the financial press is monitored so that information on international transactions and on economic agents can be taken into account in compiling the BOP.

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

Source data from the surveys are on the whole consistent with the required definitions, classifications, valuation, and time of recording. On the other hand, respondents may be late in reporting source data for BOP compilation by the DNT. For example, during preparation of the 2004 statistics the SBDP found that, four months after the end of the reporting year, only 37 of the anticipated 126 questionnaires had been returned (29 percent).

In compiling the BOP, the DNT refines the figures by recording inflows and outflows of banknotes issued by it and circulating within the CEMAC. In doing so, it (1) uses the distribution key developed by headquarters and (2) accounts for fraudulent livestock exports and their counterpart in merchandise imports. Also, when goods transactions are declared c.i.f., it applies the 33 percent coefficient to exclude the freight and insurance components from merchandise data, which are then reclassified under the corresponding headings.

3.1.3 Source data are timely

Respondents are advised when the questionnaires are sent out that the deadline for returning the data is 30 days after the end of the reporting quarter. Reminder letters are sent when there is no response. In fact, the data are usually received after the deadline. Thus, SBDP staff must call on statistics correspondents, generally after the second quarter of year N+1, to complete the document for an annual period. For businesses established in the Western Logone region, the BEAC agency at Moundou serves as a channel for collecting statistical information.

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

Although the SBDP updates the sample survey population to add or delete economic agents, it does not specifically monitor sampling errors. It verifies survey results by comparing them with previous series, using information available within the DNT or the government. It uses information taken from the bank forms as a substitute for missing data in the surveys. Computerization offers the possibility of automatically replacing a company’s missing questionnaire by data on that company found in the forms prepared by the banks. Atypical differences are discussed with respondents.

The DNT verifies annual information taken from administrative data after the provisional results are prepared, and it submits them for assessment by correspondents (DI, DD, DTCP, and INSEED), before the CBP meets. The authorities mentioned that the DSV and the Oil Directorate will join this group as of 2006. The DNT also submits data on the major exporting and importing businesses in the form of a BOP for each economic agent to verify the accuracy of the data reported.

However, data on imports seem underestimated, and complementary data sources should be considered to inform their compilation process.

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

Overall, data compilation employs sound statistical techniques. To reduce the risk of processing error and to facilitate analysis of the responses, the questionnaires to respondents are precoded in accordance with BPM5 principles. Also, a possible sign (debit or credit) of the accounting entries is indicated. For the bank forms, four versions are printed in the form of multiple-choice questionnaires (interbank debit and credit settlements, manual debit, and credit exchange transactions). Data entry errors are corrected, as is any confusion between units of account or of classification. In the absence of response to the business questionnaires, the data are retrieved from the bank forms.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

Other statistical procedures employ adequate methods. An example of the useful adjustments made is the standard distribution of displaced banknotes within the CEMAC, for evaluating informal trade. A specified proportion of the value of exported and imported banknotes to and from other CEMAC countries is allocated to trade in goods, the rest being allocated to such transactions as personal travel, workers’ remittances, and bank deposits. Exports are adjusted to take account of unreported livestock transactions, a portion of which is attributed to goods imports.

3.4.1 Intermediate results are validated against other information where applicable

Intermediate data taken from the surveys are compared to other data from independent sources. Thus, the results from questionnaires completed by NGOs are compared with data from the directorate responsible for monitoring NGOs. Also, food aid data are compared with information provided by the Ministry of Agriculture. In practice, data collected from the acacia gum (Arabic gum) industry are reconciled with those from INSEED. As well, data provided by the MEF on project grants are compared with those from the Ministry of Planning, Development, and Cooperation (MPDC).

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

The DNT routinely assesses the behavior of certain series by examining the time trend, the level of economic activity, and other related indicators. Thus, it compares data on freight and insurance with the values of the imported goods.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

Errors and omissions are investigated using a determined ratio, such as the relationship of errors and omissions to cumulative merchandise debit and credit entries. If that ratio exceeds five percent, the SBDP will conduct verifications, adjustments, and reconciliation between current transactions and capital account and financial operations, as well as any supplementary surveys needed to reduce this discrepancy.

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

Revisions to the provisional BOP statistics are periodically assessed as part of the monetary programming work, particularly during the fiscal year review. Analyses of revisions are performed after updated provisional data are compiled. They are discussed in an analytical note explaining the discrepancies found between the initial and revised preliminary data. The statistical series and the notes are intended primarily for use by the monetary authorities. Once they have been validated, they may be provided to other users, such as the national or international financial authorities. When approved by the CBP, the data are considered final and are published; thus they are not subjected to further revisions.

4.1.1 Periodicity follows dissemination standards

The periodicity of annual BOP statistics for Chad follows the GDDS standards.

4.1.2 Timeliness follows dissemination standards

The timeliness of BOP statistics for Chad does not follow the six- to nine-month GDDS standard.

The indicative schedule (where N is the reference year) for the BOP statistics work in all CEMAC member countries is as follows:

• first-quarter, year N+1: end of data collection from questionnaires and bank forms;

• second-quarter, year N+1: data coding and entry;

• month 7, year N+1: analysis of results and correction of discrepancies;

• month 8, year N+1: review and approval of data by DRFE;

• month 9, year N+1: CBP meeting; and

• month 10, year N+1 and following: proofreading, printing and distribution of the BOP Annual Report.

In practice, this calendar is not adhered to because of late reporting by respondents. For example, the 2003 BOP was approved by the CBP only on April 26, 2005 instead of in September 2004, i.e., already seven months later than recommended in the GDDS.

The annual BOP statistics are supposed to be published, in the BOP Annual Report, in the months immediately following approval of the figures. On this point, the DNT is out of step both with the BEAC’s provisions and with the GDDS recommendations. The SBDP will attempt to overcome this weakness by releasing the data for 2002 and 2003 simultaneously.¹³

To date, BOP statistics posted for Chad cover 1995–2002.

4.2.1 Statistics are consistent within the dataset

The BOP statistics are internally consistent. The concepts, definitions and classifications for producing quarterly and annual BOP statistics are the same. Financial transactions data are reconciled with changes in the international investment position (IIP). These flows represent transactions and exchange rate changes. Over the long run, the errors and omissions item has been stable.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

Annual time series following the BPM5 format are available since 1995. Unusual changes are explained in the commentary in the BOP Annual Report. The DRFE has reconstructed data prior to 1995 (originally compiled using the BPM4 format) in the wake of the reform, to conform to the new standards. However, historical BOP statistics show inconsistencies that hamper their use for surveillance and monitoring purposes.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

The BOP statistics show weak consistency with other macroeconomic datasets.

Thanks to the bridge tables and the annexes, government budgetary and financial data can be identified and reconciled with those of the BOP. The coordination that goes on prior to the CBP meeting, which brings together respondents from the DRFE, DTCP, INSEED, and SBDP, makes it possible to establish consistency with the data supplied by the MEF and the MPDC. In addition, the CBP meeting provides a setting to other institutional members of the committee for ensuring that the BOP statistics are generally consistent with the source data they have supplied to the SBDP.

Total imports and exports of goods are not always the same in the national accounts and in the BOP statistics, and the situation is worse when oil industry data are figured in.

The BOP statistics are consistent with the monetary statistics just because the SBDP takes the external position of the central bank and commercial banks as calculated by the BEAC.

4.3.1 Revisions follow a regular and transparent schedule

As part of the monetary programming work, successive preliminary data are revised as needed during the year. The revised data are transmitted to the monetary programming authorities before their approval as definitive. They become definitive once the President of the CBP approves the report as amended with the observations of the CBP’s institutional members. To date, the definitive BOP data have never been subjected to a revision.

4.3.2 Preliminary and/or revised data are clearly identified

The provisional data, not published, contained in the monetary programming framework are preceded by the mention “Estimates.” The definitive data are designated “Réalisations” (“actual outcomes”) and are identical to the data approved.

4.3.3 Studies and analyses of revisions are made public (see also 3.5.1)

Nonapplicable.

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

The BOP statistics are presented in a way that largely facilitates proper interpretation and comparison.

The BOP statistics are published each year in the BOP Annual Report. The last available publication is as of 2003. BOP statistics for 2004 have not yet been approved, and relevant tables are sent out on request.

The BOP statistics tables are presented according to the standard components of BPM5. The statistics for year N are available during the course of year N + 1 as a time series covering two years only. The statistics are supported by commentaries comparing developments during the period covered with those of the previous year. Also available are a note on the BOP accounts, the BOP tables, the bridge tables, the annex tables, and the IIP tables. No graphs are available.

The BOP statistics also appear in the BEAC Annual Report, published by headquarters, and summaries are included in the Chad’s Annual Report of the National Credit Council (Conseil national du crédit), prepared by the DNT. Excerpts from the BEAC Annual Report are posted on its website at www.beac.int.

Finally, BOP statistics are also presented in the Comptes économiques et Prévisions published annually by the INSEED.

5.1.2 Dissemination media and format are adequate

Statistics are disseminated in ways that meet users’ needs. The CBP meeting is followed by a press release describing the highlights of the BOP adopted. The BOP Annual Report is then published in hard copy and made available to users.

5.1.3 Statistics are released on a preannounced schedule

Although the SBDP’s work schedule indicates a dissemination calendar, this is not announced in advance. As a rule, data for year N are supposed to be published before the end of year N + 1: there is no further precision as to the date.

5.1.4 Statistics are made available to all users at the same time

CBP members have access to the data before the data are publicly released. The statistics are made available to all users immediately after their adoption, and they are delivered as a report to certain recipients, such as government departments and agencies, the university library, the country’s documentation centers, the resident representative office of the IMF, other national directorates of the BEAC, and the most important respondents that have contributed to preparing the BOP statistics. The report is available for on-site public consultation at the DNT library.

5.1.5 Statistics not routinely disseminated are made available upon request

Unpublished and nonconfidential data may be obtained in detail on demand.

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

The only metadata available today to the public are those prepared in English for the GDDS and posted at the IMF’s DSBB. Moreover, they have not been updated since they were posted in July 2002. Their availability is not indicated in any of the dissemination channels used in Chad.

5.2.2 Levels of detail are adapted to the needs of the intended audience

As explained in paragraph 5.2.1, only one version in English of the metadata is made available to the public, and, thus, the content is not adapted to the specific needs of various audiences. Even the French version of the metadata, made available by the IMF to the authorities, is not posted at any website, nor is it published in any document.

5.3.1 Contact points for each subject field are publicized

BEAC publications do not indicate the name of any contact other than the central bank. Some BEAC publications, such as the Money Market Bulletin of the BEAC Zone (Bulletin du marché monétaire de la zone BEAC) and Études et Statistiques, contain information for contacting BEAC headquarters. The DNT library is open to the public for on-site consultation. If users encounter difficulties in interpreting or understanding the contents or methodology used in preparing the statistics, the SBDP offers to assist them.

Contact names and numbers (telephone and fax) for the SBDP and for the DNT’s research unit are indicated in the GDDS metadata at the IMF website.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

The Études et Statistiques bulletin and the Money Market Bulletin are available for a fee. Other BEAC publications are free of charge.

Table 7:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Balance of Payments Statistics

(Compiling Agency: Bank of Central African States – BEAC, National Directorate for Chad – DNT)

Table 7:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Balance of Payments Statistics

(Compiling Agency: Bank of Central African States – BEAC, National Directorate for Chad – DNT)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element		NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
			O	LO	LNO	NO
0. Prerequisites of quality
	0.1 Legal and institutional environment		X
	0.2 Resources				X		Staffing levels are inadequate to ensure prompt processing of forms returned by commercial banks.	Short-term (ST). Add immediate staff to process declarations more promptly; staffing can be cut later when the planned computerization of forms returned by commercial banks is in place.
	0.3 Relevance				X		There is no device in place for measuring relevance.	Medium-term (MT). Introduce a mechanism to monitor and improve relevance of BOP statistics
	0.4 Other quality management			X			No attempt is made to ascertain the opinions of users, other than CBP members.	MT. Introduce a mechanism to survey users’ needs.
1. Assurances of integrity
	1.1 Professionalism		X
	1.2 Transparency			X			There is no policy of giving advance public notice of methodological changes. The public is not informed of the access of government officials to BOP statistics prior to their release.	ST. Introduce the practice of providing advance public notice of methodological changes. Inform the public that government officials have access to BOP prior to their release by implementing an adequate warning in BEAC’s publications and on its website.
	1.3 Ethical standards		X
2. Methodological soundness
	2.1 Concepts and definitions		X
	2.2 Scope		X
	2.3 Classification/sectorization		X
	2.4 Basis for recording			X			Some transactions are not recorded on an accrual basis.	MT. Take steps to record all transactions on an accrual basis.
3. Accuracy and reliability
	3.1 Source data			X			Respondent firms do not respect the reporting deadline. Source data on imports are insufficient.	ST. Launch information campaigns (meetings, seminars, the media) to encourage the timely return of questionnaires. Collect more reliable source data on imports.
	3.2 Assessment of source data			X			Revisions on imports are not taken into account to consider possible complementary data sources.	ST. Identify complementary data sources to inform compilation of data on imports.
	3.3 Statistical techniques		X
	3.4 Assessment and validation of intermediate data and statistical outputs		X
	3.5 Revision studies		X
4. Serviceability
	4.1 Periodicity and timeliness				X		Periodicity of BOP statistics follows GDDS standards, but dissemination to the public is late.	ST. Reduce the time elapsing between approval of BOP statistics by the CBP and their dissemination.
	4.2 Consistency				X		BOP statistics are not consistent over time. BOP data betray inconsistencies with other datasets, in particular on foreign trade with national accounts.	ST. Institute regular and ongoing cooperation between the DNT (SBDP and SSAM) and the MEF (Customs and the public debt unit) for resolving these inconsistencies. Solve inconsistencies with national accounts on foreign trade.
	4.3 Revision policy and practice			X			Revision studies are not made available to the public.	ST. Publish revision studies
5. Accessibility
	5.1 Data accessibility			X			BOP statistics are not released on a preannounced schedule.	ST. Establish and advertise a preannounced schedule for the dissemination of BOP statistics to the public.
	5.2 Metadata accessibility				X		The GDDS metadata are disseminated only at the IMF website, in English; the French version of the metadata is not publicly available.	ST. Post the metadata in French at the BEAC website and insert them periodically in BEAC publications.
	5.3 Assistance to users			X			There is no public notice of any assistance available to users.	ST. Inform the public of the user assistance offered by the BEAC.

View Table

Table 7:

Chad – Data Quality Assessment Framework (July 2003): Summary of Results for Balance of Payments Statistics

(Compiling Agency: Bank of Central African States – BEAC, National Directorate for Chad – DNT)

Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice Largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria
Element		NA	Assessment				Comments on Assessment	Plans for Improvement and Target Dates
			O	LO	LNO	NO
0. Prerequisites of quality
	0.1 Legal and institutional environment		X
	0.2 Resources				X		Staffing levels are inadequate to ensure prompt processing of forms returned by commercial banks.	Short-term (ST). Add immediate staff to process declarations more promptly; staffing can be cut later when the planned computerization of forms returned by commercial banks is in place.
	0.3 Relevance				X		There is no device in place for measuring relevance.	Medium-term (MT). Introduce a mechanism to monitor and improve relevance of BOP statistics
	0.4 Other quality management			X			No attempt is made to ascertain the opinions of users, other than CBP members.	MT. Introduce a mechanism to survey users’ needs.
1. Assurances of integrity
	1.1 Professionalism		X
	1.2 Transparency			X			There is no policy of giving advance public notice of methodological changes. The public is not informed of the access of government officials to BOP statistics prior to their release.	ST. Introduce the practice of providing advance public notice of methodological changes. Inform the public that government officials have access to BOP prior to their release by implementing an adequate warning in BEAC’s publications and on its website.
	1.3 Ethical standards		X
2. Methodological soundness
	2.1 Concepts and definitions		X
	2.2 Scope		X
	2.3 Classification/sectorization		X
	2.4 Basis for recording			X			Some transactions are not recorded on an accrual basis.	MT. Take steps to record all transactions on an accrual basis.
3. Accuracy and reliability
	3.1 Source data			X			Respondent firms do not respect the reporting deadline. Source data on imports are insufficient.	ST. Launch information campaigns (meetings, seminars, the media) to encourage the timely return of questionnaires. Collect more reliable source data on imports.
	3.2 Assessment of source data			X			Revisions on imports are not taken into account to consider possible complementary data sources.	ST. Identify complementary data sources to inform compilation of data on imports.
	3.3 Statistical techniques		X
	3.4 Assessment and validation of intermediate data and statistical outputs		X
	3.5 Revision studies		X
4. Serviceability
	4.1 Periodicity and timeliness				X		Periodicity of BOP statistics follows GDDS standards, but dissemination to the public is late.	ST. Reduce the time elapsing between approval of BOP statistics by the CBP and their dissemination.
	4.2 Consistency				X		BOP statistics are not consistent over time. BOP data betray inconsistencies with other datasets, in particular on foreign trade with national accounts.	ST. Institute regular and ongoing cooperation between the DNT (SBDP and SSAM) and the MEF (Customs and the public debt unit) for resolving these inconsistencies. Solve inconsistencies with national accounts on foreign trade.
	4.3 Revision policy and practice			X			Revision studies are not made available to the public.	ST. Publish revision studies
5. Accessibility
	5.1 Data accessibility			X			BOP statistics are not released on a preannounced schedule.	ST. Establish and advertise a preannounced schedule for the dissemination of BOP statistics to the public.
	5.2 Metadata accessibility				X		The GDDS metadata are disseminated only at the IMF website, in English; the French version of the metadata is not publicly available.	ST. Post the metadata in French at the BEAC website and insert them periodically in BEAC publications.
	5.3 Assistance to users			X			There is no public notice of any assistance available to users.	ST. Inform the public of the user assistance offered by the BEAC.