The financing of statistical activities is a constant cause for concern on the part of state statistical services. It must be addressed throughout the whole process of design and implementation of National Strategies for the Development of Statistics (NSDS). During the assessment phase, what is required is to identify the financial resources available to a country’s national statistical system, to demonstrate and analyse their main characteristics (sources, amounts, activities covered, mechanisms for the provision and management of funds, applicable conditions, etc.). In the light of the problems that emerge from the assessment, the objectives and expected results are determined, and an action plan covering all the activities to be completed during the period of NSDS is prepared. The action plan is accompanied by a timetable and by a budget prepared on the basis of the costings for the activities and stating the sources of financing (both national and external). Thus, it appears that a financing strategy must be an integral part of NSDS.
NATURE OF THE EXPENDITURE TO BE FINANCED AND SOURCES OF FINANCING AND COSTING (COST ASSESSMENT)
Nature of the expenses involved in the NSDS
In a developing country, the main issues that need to be resolved to make the statistical system perform better, that is to say, to produce relevant, detailed and complete statistics on a long-term basis, that are delivered on time and in formats compatible with good decision-making by economic and social stakeholders, are to improve the governance of the national statistical system, to consolidate the foundations of sustainable development of statistics production, and to improve the quality of statistics production and the use of statistical products.
Resolving these problems requires the preparation of action plans including activities of various kinds that give rise to expenditure that can be grouped together under three main headings: purchases of goods, purchases of services, transfers, etc. Items of expenditure should be identified and costed appropriately.
Items of expenditure must be exhaustively identified and properly costed. The methods used to assess the costs vary according to the nature of the expenditure.
In the case of purchases of goods, state statistics departments have the costs of purchases previously made which are found in estimates, pro-forma invoices or tender documents as the case may be. These costs can be used as a reference for costing NSDS items of expenditure. It is also possible to send potential suppliers requests for information and prices at any time. In the case of purchases of services, contracts signed in the past can be used. For services such as consultancy services, scales of charges for various qualifications (national consultants, international consultants, etc.) can be obtained, particularly from technical and financial partners. Finally, in the case of the transfer of expenditure such as the payment of college fees for the training of statisticians or the payment of contributions to regional or international organisations, costs are communicated by the beneficiary institutions.
In general, great attention must be paid to harmonisation of the methods used to cost action plans for NSDS, especially in countries where statistical systems are decentralised.
FINANCING STRATEGIES AND SEARCH FOR FINANCING
The activities involved in action plans for a NSDS can be financed from national (or internal) resources or from external resources. Usually, a combination of both sources is used. Day-to-day running costs should normally be financed with national resources. Recourse to external resources can be justified in the case of investment expenditure (development of new tools, the completion of national surveys or censuses, training and improvement of human resources, construction of buildings, etc.).
In order to mobilise the substantial financing required for the successful implementation of an action plan of an NSDS, any government of a developing country must balance national and external resources while taking account of the nature of the activities to be financed, development priorities and the country’s financial capacity, as well as the possibilities offered by bilateral and multilateral cooperation. It must thus define a financing strategy, that is to say an “optimal” combination of the possible choices. The proportion of national resources in this combination varies widely from one country to another, and, for the same country, from one period to another. In most of the LDCs, the proportion of financing derived from external resources far exceeds that obtained from internal sources.
A financing strategy must be based on decisions taken at the highest level of government. The development of a national statistical system is, above all, a political matter. It is one of the sovereign missions of the State, which must decide what proportion of national resources will be devoted to statistical activities, and whether or not the system will be developed by having recourse to loans, since it is certain that developing a national statistical system will require medium and long-term investments.
Very few countries have recourse to loans to finance their NSDS, even on very easy terms. When it implemented its 2004-2009 NSDS, Burkina Faso’s financing strategy was based on a combination of resources derived from the national budget, from grants in the form of project and programme aid, and from an STACAP loan from the World Bank at a concessionary rate.
For their part, technical and financial partners have gradually felt the need to refine their financing strategy to respond better to the growing needs of developing countries. The majority of bilateral and multilateral donors, if not virtually all of them, have programming frameworks for their development aid: European Development Fund, UNDAF, country programmes (African Development Bank), etc. Support for statistical development through the financing of NSDS can be mentioned in the context of these programming frameworks.
The search for financing by countries
The NSDS financing strategy must be defined quite early in the process of its preparation, preferably at the end of the national statistical system assessment. The results of the assessment should be shared with technical and financial partners in order to establish a sound basis for discussion with them when a decision has been made on the strategies adopted to strengthen the capacity of the national statistical system, and later when the action plan for NSDS has been prepared. This discussion will be much more likely to succeed if the government has clearly determined the proportion of national resources that it anticipates allocating to the financing of the country’s NSDS. Of course, thorough knowledge of external sources of financing is required, and should be one of the objectives of the assessment.
Since the beginning of the millennium, many developing countries have drawn up and implemented Poverty Reduction Strategy Documents (PRSD) with the support of the international community. In LDCs, PRSD and other equivalent documents now constitute the reference framework for development strategies and policies and for the intervention of technical and financial partners. These documents define priorities, operational objectives and expected results as well as the activities to be carried out, and indicators, in accordance with management principles focusing on results. Statistics should be given a priority status in the development strategy and the NSDS recognized as part of the strategy.
Unless statistics are regarded as a sector to be developed in its own right, PRSD should, in their lists of priority actions, include the main activities of the action plans of NSDS with their costs and timetable; this will facilitate the search for financing for statistical development when the discussions or consultative groups organised to mobilise the financing of the PRSD are prepared and take place.There ought to be a close articulation between the global development strategy and the NSDS. This linkage is unfortunately often missing.
Advocacy in favour of financing NSDS, which takes place during discussions and in consultative groups, is of course important, but it is not enough on its own. The commitments given by donors of funds before and during these events must be followed up, and must materialise, for the purpose of the relevant projects presented by governments over time. There must be constant advocacy, using all available channels, such as groups of technical and financial partners involved in national statistical development (see the example of Mali)( see D. ADVOCACY).
The way in which resources are managed and the expenditure that they should cover vary according to the sources of financing and the financing mechanisms.The case of national resources, and that of external resources, will be dealt with in turn below.
Who and When
National resources destined for the financing of statistical activities are derived from the State’s budget. While national statistical institutes generally have their own budgets, the same is not true of sector-based statistical services whose resources are often mixed up with those of the units on which they depend. For this reason, it is difficult, if not impossible, to know the amount of credit allocated by the State’s budget to sector-based statistical services. To face this difficulty, the preparation on a yearly basis of a consolidated budget integrating the needs of the whole NSS (NSO and other official data producers) would help.
In the case of national statistical institutes, the units handling their administrative and financial affairs are responsible for the preparation of a draft budget every year under the authority of the institute’s director. This must observe the various stages of the timetable of preparation of the State’s budget and take account of the recruitment of personnel.
Generally speaking, the resources allocated to the official statistical system from government budgets will experience great difficulty in increasing more rapidly than the country’s Gross Domestic Product over a very long period, regardless of the priority the government places on the development of its statistical system. One way to increase the effectiveness of funds targeting statistical activities would be to reform the structure of those statistical units in line ministries that support the central statistical body so as to increase their profile and attract a suitable and dedicated budget.
National Statistical Offices (NSOs) that enjoy a level of independence — a trend that seems to grow, in particular in Africa — should consider signing performance contracts with their government to guarantee a pre-defined volume of financial resources over a period of several years. Such an arrangement would require that the NSO deliver well defined statistical outputs. These NSOs should take full advantage of their independent status by offering paid services to third parties, while not losing sight of their public service mission.
In the case of a sector-based statistical department that does not have its own budget, its manager is responsible for arguing for its needs to be taken into account at the time of preparation of the draft budget of the unit to which it is attached.
The State’s budget is voted on by Parliament each year. In the case of investment expenditure, public investment programmes are also drawn up over two or more years. Increasingly, programme budgets are also prepared to meet the needs of results-based management. Finally, other tools such as Medium Term Expenditure Frameworks (MTEF) are also deployed.
In developing countries, and particularly African countries, preference has been given to the preparation of medium-term expenditure frameworks, on the one hand, at the beginning of the millennium, by management focussing on the results of development, which boosted planning activities and which was behind the Poverty Reduction Strategy Documents (PRSD), and on the other hand, by budgetary aid, which has developed since then and which is strongly encouraged by the Paris Declaration on Aid Effectiveness.
According to the World Bank, “an MTEF comprises a global financial budget set at the highest level by central departments, basic discussions to estimate the current and medium-term costs of public policy options, and finally, a process of adjustment of costs and available resources.” Thus, an MTEF makes it possible to reconcile budgetary allocations, the objectives to be achieved and performance measurement. It is therefore possible at all times to balance resources and expenditure to achieve a fixed objective. Usually, a distinction is made between global and sector-based MTEF, the latter being drawn up at the level of ministerial departments, which are given responsibility for their management. In the context of implementation of NSDS, a “sector-based” MTEF can be prepared for statistical purposes for the implementation of an action plan, including the activities that must be carried out by all public statistical departments, regardless of their supervisory ministry. Unfortunately, few examples of such MTEF are really prepared in developing countries. The high technical and political status conferred to the MTEF by the Bretton Woods institutions and the Ministries of Finance suggests that the explicit inclusion of the NSDS and its costs in a METF should be pursued.
Having regard to the limited resources that developing countries devote to statistical activities, the deployment of NSDS remains largely dependent on external resources. The financing of statistical activities by technical and financial partners can take various forms: project aid, programme aid, general budgetary aid, and sector-based budgetary aid.
Project aid is where a donor makes resources available that are intended exclusively to finance the whole or part of a statistical operation (survey or census) or an activity aiming to increase capacity (for example, training or the reinforcement of human resources). Programme aid is wider in scope. It can target the development of an area of activity other than statistics (for example, health, education, agriculture) while reserving part of the funds for statistics.
Certain donors provide budgetary aid which involves making financial resources available for the State’s budget to help it to achieve certain development objectives that can be general, such as the improvement of governance (general budgetary aid) or sector-based (sector-based budgetary aid).
The completion of national surveys and censuses demands substantial finance, which often makes it necessary to have recourse to several donors. For this purpose, certain donors agree to pool their financial resources to facilitate the management of expenditure. Such funds are called “basket funds”.
Basket funds can also be put in place to implement NSDS (the case of Rwanda). The setting up of a basket fund usually involves the signature of an agreement or memorandum between the government side and the donors concerned. These documents contain the rules of the game: the objectives to be achieved and the results expected, the respective contributions of the donors (or even of the government), the mechanisms for managing and disbursing funds, the system of monitoring, assessment, reporting and audit, etc. The management of basket funds may be entrusted to one of the donors chosen by agreement, or to the country’s central statistical organisation, under the supervision of a steering committee. A procedures manual is drawn up and programmes of work are periodically prepared, implemented and assessed. The arrangement of a basket fund makes it possible to reduce administrative management costs by unifying accounting procedures and by having a single reporting system.
Specific financing instruments have been set up multilaterally. This is the case of the World Bank Trust Fund for Statistical Capacity Building, which finances the preparation of NSDS and other activities to build statistical capacity in the form of donations. The World Bank’s STATCAP programme is intended to finance the implementation of NSDS with loans at very low interest rates. More recently, the Statistics for Results Facility has been created to help countries that have a suitable and partially-financed NSDS, to receive donations to implement it. A first group of pilot countries has thus had the benefit of substantial financing.
Finally, since the African Charter on Statistics was adopted in 2009, some African countries have decided to set up statistical development funds (not operational yet). These funds could be in a number of forms: the setting up of an appropriations account at the State Treasury that is managed in accordance with the rules of public accounting; the creation of a public administrative institution of modest size to manage funds made available by the State in the form of grants, or even allocated revenues, as well as contributions of funds from donors that agree to participate in the provision of such funds.
It therefore appears that there are a wide variety of methods of managing finance and expenditure. These mechanisms are not mutually exclusive. Frequently, they are used simultaneously to good effect.
The management of financing from a national source and of the expenditure relating thereto is governed by the rules of public accounting. Disbursements of funds are managed by departments of the State Treasury, while expenditure is managed in accordance with the procedures applicable to public finances, and in particular the public procurement code.
In the case of external sources of finance, management procedures vary according to who the donors are and what financing mechanisms are put in place. In the case of project aid and programme aid, the management of resources can be entrusted wholly or partly to an external autonomous management unit, or to one that is set up inside the beneficiary institution. The management rules and procedures are laid down by the donor. According to their complexity and the extent to which they are under the control of national managers, observance of applicable time limits will be more or less assured.
As its name indicates, budgetary aid is paid to the State’s budget beneficiary. The Paris Declaration on Aid Effectiveness adopted in 2005 recommends that donors should define their aid in accordance with the priorities laid down by the beneficiaries themselves, and that they should use the latters’ internal procedures. They should therefore harmonise their interventions. Disbursements of budgetary aid are often subject to conditions, non-compliance with which can have negative consequences for the statistical activities that would be financed in this context.
As for basket funds, they can be managed like a Trust Fund or as a component of global budgetary aid.
Regardless of the methods used to manage financing from external sources, it is necessary for the managers of public statistical offices and/or of NSDS to have a thorough knowledge of the aid programming cycles of the various donors of funds, in order to benefit from this type of financing on time and to the maximum extent.
Difficulties encountered and solutions
In many developing countries, the results obtained in the area of financing and the implementation of NSDS are rather disappointing, which explains why it has not been possible to complete a number of activities included in the action plans within the time agreed. There are several possible reasons for this, in particular the unrealistic nature of the action plans, which plan too many activities having regard to the human resources available, and the absence of an advocacy strategy and genuine financing strategy.
Ideally, the financing strategy for NSDS should be discussed immediately after the adoption of the survey report and during the period in which the strategies are drawn up, and thus well before the preparation of action plans, properly so-called. In general, it is once the action plans are approved by the government that the statisticians get down to organising a conference of donors of funds. To date, it has not been possible to organise many conferences of this kind, and the few that have been organised have had mixed results, with donors being more interested in Poverty Reduction Strategy documents. A strong linkage between PRSPs/National Development Plans and NSDSs would increase the visibility of the NSDSs an ease the mobilization of resources from the donors.
Another difficulty encountered in the financing of statistical activities arises from the weaknesses in the system of monitoring and evaluating the NSDS. The period covered by NSDS (three to five years, or even longer) is too long to enable the State to enter into specific financial commitments. It is therefore important to draw up annual national statistical programmes that will regularly be the subject of implementation reports (see IMPLEMENTATION). Although derived from the action plans of NSDS, such programmes must only include activities that have already obtained financing or that are capable of securing financing in the course of the year under review. In this respect, a group of technical and financial partners supporting statistics is the appropriate framework within which to combine or supplement the financing already available from national or external resources, and at the same time enables a continued dialogue with national decision-makers on statistical development. The relatively successful experience of Mali is an example of this. (see examples of good practices below).
Along with inadequate human resources, difficulties encountered in the financing and budgeting of statistical activities are among the main causes hindering NSDS implementation. Particular emphasis must be placed on the following points: the necessity for a realistic financing strategy; a proper connection between NSDS and National Development Plans; sound control of financing strategies, programming frameworks and donor disbursement procedures; the importance of constantly arguing in favour of the mobilisation of funds; and monitoring, assessment and reporting of the financial performance of the action plan, regarded as an integral part of the global system of monitoring, assessment and reporting of NSDS. Finally, the possibilities offered by the new donor aid procedures such as budgetary aid should be explored in detail, while remaining conscious of their constraints and limitations.
Admittedly, the synchronization of NSDS plans with each donor programming cycle is quite difficult. It might take one or two years before approval is given to a formal request and several months before the funds are released.
A strong political commitment in favor of the NSDS and a more performing NSS should lead to an increasing portion of the budget financed by the country and to more independance vis-à-vis the donors. The preparation of a new NSDS should target a rise of the fraction of the NSS financed by the country.