• The Need for SIPS: Facing a Major Issue in Public Finance Management
• The SIP Formulation Process for Agriculture

PART II. APPLICATION OF THE METHODOLOGY TO THE FORMULATION OF THE ASIP FOR THE MOUNTAIN AREAS

• Foreword to Working Paper 1 of the FAO Formulation Report on the ASIP for the Mountain Areas
• B. Functions, Objectives, Tasks and Personnel Requirements
• C. The Recurrent And Capital Budget: Procedures, Issues And Improvements Required to Become an Effective Channel for Disbursement of Donor Support
• D. Recurrent and Capital Budget: Estimate of the Model ASIP Budgets of the Mountain Districts and of the Range Management Division of the DLS
• E. Evaluation of the Projected MOA Expenditure in the Mountain Areas under ASIP

The 1996 World Bank-OED report on the experience of fifteen years of lending for structural adjustment by the Bank identifies the historical trends in the conception and design of such operations, and provides a basis to foresee where more emphasis ought to be placed in current and future ASIP design by Governments and Donors alike. The OED report provides general confirmation of the approach taken in Lesotho, although the latter differs with respect to the emphasis given to some important aspects, in particular to issues in public finance management.

The second part deals with the application of the concepts and methodology to the design of the ASIP for the Mountain Areas in Lesotho. It presents the main features of the analytical work done to link goals, policies and strategies of the Government with functions, objectives, targets and tasks of the administration, and to link these with the budgeting and financial reporting process. The implicit suggestion is that some major difficulties of current ASIP implementation in other countries are due to inadequate integration of the substantive and policy aspects of ASIP design with the institutional and procedural aspects at the time of the programme formulation and appraisal.

The applicability of the methodology depends on the progress already made by a country along the path of structural adjustment, and on the complexity of the public administration. Lesotho has already made considerable progress in bringing public expenditure under control and its administration is relatively simple. At the time of writing, not all proposals presented in the second part of this paper had been discussed and agreed with the Government, and therefore the presentation reflects the personal views of the Author more than those of Lesotho officials in several respects. Nevertheless the Author believes that the presentation is of general interest as an example of how to handle the typology of problems to be faced by ASIP designers in similar, and in more complex circumstances as well.

Specialists in Institutions and Management, and specialists in Extension, and in Monitoring and Evaluation may find that the application to Lesotho misses parts of what they would consider important elements of design. Though some of this may be the fault of Author, the practical work in Lesotho has been inspired by the principle: design what can be reasonably expected might be done, rather than all that should be done, and never forget what can be done in recommending what should be done.

PART I. ASIP: CONCEPT, DESIGN, AND METHODOLOGY OF EVALUATION

Summary

(i) Structural Adjustment Programmes (SAP) have been implemented since the early 80s in many developing countries . In a first phase, SAPs were designed essentially to redress serious macro-economic imbalances, rationalize the price system, fight inflation, curb excess demand for foreign exchange, and bring the budget deficit under control. In a second phase other objectives were added to introduce radical changes in economic policies, aimed at dismantling the role of the State in the production of goods and services, and in the control of the market mechanisms. The theory behind the latter approach is that interference with market forces leads to sub-optimal allocation of resources and obstructs private initiatives, preventing the unleashing of the energies and imagination necessary for vigorous economic development; that a combination of subsidies, Government spending on activities more efficiently undertaken by the private sector, and inefficiencies of public enterprises absorbs resources which could be put to better use, thereby resulting in less investment, distorted income distribution, and non-sustainability of Government finance; and that the economic returns from setting the system of incentives right and improving institutional capacity would far outweigh the returns from incremental investment made under a less enabling environment. Macro-economic studies in many countries have shown positive correlations between the introduction of liberal policies, contraction of Government spending, and general economic growth, although the relationships between specific policy change and specific production responses have been difficult to quantify. In certain cases, liberalization has actually resulted in less production. SAPs have been generally successful in bringing about significant reductions in the overall budget deficit, along with the implementation of important policy changes.

(ii) Public Expenditure (Sector Investment) Programmes (PEP, SIP) have attempted to link policy change with improvements in the structure of Government spending in specific sectors of the economy, and, increasingly over time, have supported institutional reform in addition to capacity building. The focus of SAPs and PEPs has essentially been on the macro-economic linkages of policy change and production responses. Donors’ support has generally been justified by the need to fill a balance of payments gap resulting, in the short run, from the implementation of policy reforms. Quick disbursement of structural loans was expected as a result of tranche financing triggered by compliance with policy conditionalities, an expectation which did not often materialize.

(iv) First and foremost is the need to recognize the full implications of the nature of SIPs as exercises in public finance planning and management, which must comply with the basic rules of public finance management. In this connection, the adverse impact of Donor project finance on the system of incentives governing the allocation of domestic resources available from the revenue account ought to be fully appreciated. The sustainable ceiling on all public expenditure which can be incurred on recurrent account in a sector is a key variable that needs to be known in advance, and allowed to play its basic regulatory function on expenditure related to all Government activities, irrespective of the source of finance. Donors’ support for SIPs must accept to include finance for eligible incremental recurrent expenditure on a medium term basis, as a contribution justifiable in its own right . However this requires that the "justifiable increments" can be estimated, that is after current expenditure has been pruned of all outlays which do not conform with the agreed SIP goals, and with the consistent set of strategies and policies adopted to satisfy the goals, including the least cost condition requirement.

(v) Consequently, SIPs must be built on the basis of the necessary restructuring of all the ongoing Government activities in the sector (including project-financed activities) which need to be evaluated in the light of the goals, policies and strategies elaborated for the sector. This must be a detailed assessment, the end product of which is a precise definition of the functions and objectives (targets) of each unit of the Administration and of the tasks of the staff responsible for the operating units within the Administration. Upon these, a realistic estimate of the cost of the staff, and of the means which they require to implement their tasks, can be arrived at, and the eligible increments estimated. Failure to do this prior to SIP final design could lead to misunderstandings about implementation between Governments and Donors and among staff in different positions within the administration. This is a prime cause for Donors, and certain quarters in Governments, to overestimate "the borrower’s ownership" of the SIP, and to an understandable lack of confidence by Donors in what is the acceptable gap in recurrent expenditure finance that they should be willing to fill.

(vi) The third aspect is the need for a genuine effort to (i) understand the mechanisms and procedures of a Government budget process (preparation, approval, control, and reporting) and of the problems encountered by staff in complying with the procedures; (ii) agree on the necessary minimum streamlining and modernization of the procedures, compatible with what the accounting staff can be expected to handle effectively in practice; and (iii) design a system whereby Donors would be able to accept Government (reformed) procedures as a channel for their financial support . This would make it possible to reduce the scope for "project" financing with separate reporting, accounting and audit procedures to those large-scale investments in capital formation, which by their very nature (no recurrent costs involved, long implementation period, need for special procurement methods) should be properly handled in an autonomous way.

(vii) A fourth aspect is the need to recognize the full implications of the political nature of decision making on the allocation of resources within a budget. The overall budget ceiling can be determined to a large extent on technical grounds, and effective rules can be enforced to keep within an overall target ceiling . But the allocation to different sectors, and to different activities within a sector, is the result of a bargain among different interest groups with political power. The analytical tools which can be used to design a consistent public expenditure programme, per se, are not enough to ensure that the allocation of resources will comply with stated Government objectives. At best they can provide some guidance to political discussions and decision making. To this effect, they must be designed so that they can be easily understood by the public and by politicians. Institutional development aimed at a closer relationship of the civil society with Governments can help to reduce the margin for possible divergence. Decentralization of the public administration is one important step in this direction, but certainly not a sufficient condition for substantial progress. The democratization process of developing countries will increasingly associate different institutions of civil society, in different ways, with the decision-making mechanisms of Governments. This is a process which will follow its own historical pace, and which can be facilitated by Donors’ support, provided there is sufficiently strong local political pressure to persuade Governments that it is legitimate for Donors to play that role, and to utilize official channels for their financial support of civil societies institutions other than Government.

(viii) It is suggested that a more effective way to improve compliance with Governments’ stated goals, policies, and objectives in the allocation of resources would be for Donors to devise means to co-finance a Ministry budget pari passu with the Treasury, making use of the same procedures as Treasury, and shifting the detailed appraisal of their contributions from the stage of the formulation and negotiation of several year programmes, to assistance and participation in the preparation of the annual work plans and budgets of the different sections of Government whose activities they are willing to support under a SIP. From the point of view of political ownership of a SIP, the moment of truth is the approval of a budget, rather than that of a multi-annual Plan. Such a move would help the monitoring of progress; actual progress on the implementation of the annual work plan would be added to the list of performance indicators which have been conceptualized for the SIPs so far. These essentially concern the schedule of implementation of policy reforms, including the removal of concrete constraints and of regulations undermining the effectiveness of legislative action, but they have generally not gone much beyond that so far.

(ix) The concern must be recalled, which emerges from SAP/SIP evaluation, that certain declared major objectives of those programmes, such as poverty alleviation and household food security, do not automatically result from adopting liberal policies. In many ASIP formulations, the shaping of Government strategies, policy frameworks, and allocation of resources, tend to be influenced by the non-critical and superficial association of the objective of agricultural diversification with growing export crops (which disregard the significance of foodcrop production) and by the adoption of Western recipes for increasing land and labour productivity, with remarkably little attention being given to specific socio-economic and ecological conditions. There is an obvious danger of dogmatism substituting for thorough and objective analysis. In this connection, a word of caution is in order about the consequences of possible manipulation of the economic models used in sector studies, particularly when these are based on limited and unreliable data. These models can be used to make apparently convincing cases about measures which ignore local constraints and local opportunities, run contrary to accepted Government goals, and sometimes even to common sense . The danger lurking behind the use, in policy/strategy elaboration, of complex models which are non-transparent for politicians and for the public, is that the economists and the statisticians who work out those models can end up playing (albeit indirectly) a decisional, rather than an advisory role. They become, in effect, politicians in disguise - a role which is incorrect for advisors’ to play, not only on general grounds, but also on the grounds that they would never pay for their errors, as politicians may sometimes do.

(x) Finally, it is important to appreciate that if ASIPs must be implemented by Governments and involve the entire set of their activities. Reporting procedures on implementation performance monitoring, financial control, and audit need to be designed within the scope of what Government administrations can realistically be expected to undertake effectively, rather than on the basis of all the information which project/programme designers/evaluators naturally would like to obtain. The key to effective design is to fully appreciate what can be done, and never forget what can be done in recommending what should be done.

The Need for SIPs: Facing a Major Issue in Public Finance Management

The theory of Public Finance recognizes three basic functions which Governments perform by making decisions about the budget policy. Through the fiscal instrument Governments: (i) secure economic stability of price and employment (the stabilization function), (ii) make adjustments in the allocation of resources to different activities in the national economy (the allocation function), and (iii) change the distribution of income and wealth (the distribution function).

Stabilization Policy Issues

The basic principle of public finance is that the budget is unum. Over the years, in many developing countries, and particularly in those at the lowest end of the scale in per capita income and institutional development, the share of Donor project finance in public expenditure has soared, contributing in no small way to non-compliance with the principle. As a result, neither the stabilization nor the allocation function of the fiscal authorities has been properly carried out. The former because the fiscal authorities ended up not knowing the aggregate amount of commitments entered into by Government with regard to recurrent expenditure; the latter because they do not know on what resources they can count for the long run to carry out Government objectives. In those circumstances it becomes very difficult to set the overall budget ceiling: indeed often different accounting and reporting procedures of Donor supported projects make consolidation so complex that the full extent of the expenditure cannot be estimated, and therefore a proper allocation of expenditure to sub-sectors cannot be planned.

The problem would be less serious if project finance would contribute only to fixed capital expenditure, say road or dam construction, for example. Then the issue would be how to ensure that such expenditure is consistent with the resources expected to be available in future on recurrent account for maintenance of the new public assets. However, most agriculture, natural resource management and rural development projects do include large amounts to finance incremental recurrent expenditure, in addition to financing fixed capital formation. The matter has been further complicated by the introduction of yet another practice, widespread in Africa, which is contrary to basic principles of public finance management: the division of responsibility for budget approval and control between the Ministry of Finance and the Ministry of Planning.

The need to redress what is essentially a pathological situation in the public finance of such countries has been felt for some time, but has seldom been fully stated, nor has much been done about it, particularly in sub-Sahara African countries. From the point of view of the fiscal and monetary authorities, the main justification for the redressing is for them to be able to operate a consolidated budget: in simple terms, to know what they are doing. For them, this ought to be the prime reason for the SIP approach as the logical follow up of the SAP approach. The latter essentially dealt with the need to cut overall Government spending, the former attempts to rationalize the entire Government spending and the underlying institutional arrangements in one sector of the economy. The consolidation of the budget is as important to a SIP as the definition of Government objectives and the related policy changes, in as much as effective policy implementation requires the Government to use effective tools, the budget being the main tool through which Government action materializes. This process obviously requires Donor coordination, but this is a consequence of the need for SIPs, rather than an objective of undertaking SIPs per se.

The ceiling imposed on the expenditure of Government performs two distinct and equally important functions. The first is to manage the monetary economy, since the budget deficit is a major source of inflationary pressure (the stabilization objective). The second is to impose the application of priority criteria in the allocation of expenditure to different Government activities (the appropriate allocation objective).

Incremental extra-budgetary resources made available to Government through project financing may add slightly to inflationary pressure, to the extent that they are provided as grants, and that the projects so financed do not require Governments to contribute a share of cost from the revenue account . Extra-budgetary resources, irrespective of the condition under which they are made available by Donors, dilute the second function of the budget ceiling, in as much as the incremental project finance obtained from Donors does not have an opportunity cost in terms of the alternative activities that must be forfeited to implement the project. In fact, the process by which the budget is drawn up becomes the result of two separate and essentially unrelated functions. The development side of the budget is the result of an attempt at maximizing the flow of external project finance obtainable at the lowest possible cost and with the lowest contribution of domestic resources to project cost acceptable to Donors. Consequently, decisions about the objectives and the design of projects are often dominated by Donors, and Donors’ views are not always coordinated with one another and with Government views, in a consistent set of policies and objectives. The second function governs the allocation of the bulk of resources made available from the revenue account: since these do not have to be allocated in competition with development objectives, line Ministries are under little pressure to close down low priority activities, to re-allocate personnel to more important activities, to increase the productivity of the public administration, and to resist requests by politically powerful claimants. In short, the critically important allocation function of the budget authorities (preparing a budget by selecting the activities which are in accordance with Government objectives, policies, and strategies in order of priority) is very loosely performed in the absence of the financial discipline imposed by respect of a budget ceiling applicable to all Government activities.

An important side effect of project finance is related to the usual Donor requirement that separate accounts be kept for their contributions and that procedures be adopted which are often different of the Government established procedures. While the need for separate books is a normal requirement of any financial institution, the need for different procedures is less evident. Few Donors have actually made an effort to assess to which extent Government procedures could be acceptable to them, and thus define a basis for negotiating adjustments. A widespread perception is that Government procedures are cumbersome, obsolete, and in need of radical change. There is certainly need for modernization of established procedures (effective computerization, extended to all line Ministries, is an obvious one in almost all countries), but the main need in most ex-colonial administrations in Africa, for example, is for proper compliance with established practices, not the introduction of radical change.

As a result of the multiplication of accounts and procedures, a strong competition has been set in motion between Government and Donors for a very scarce resource in those economies; namely, accounting skills. The ability of Donors to offer better pay gave them the upper hand in the competition, with the result that Governments are short of accounting and auditing staff, reporting on Government accounts is inevitably delayed, auditing is late and often inadequate. All this has strengthened the idea that Government accounts and accountants are not to be trusted, and the vicious circle is complete. Of course, scarcity of accountants is not the only reason for the often poor state of Government financial control and reporting, but it is certainly a situation which must be remedied as a pre-condition of any possible improvement.

A major issue is the sustainability of public expenditure. Recurrent expenditure is a key variable of sustainability. The function of the Economic Planning Authority and of the Ministry of Finance is to indicate the ceiling on expenditure that will be allocated to the each line Ministry in the medium term. The line Ministries’ function then is to determine what activities to retain, and to estimate the corresponding budget on recurrent expenditure account. The sustainability principle demands that items proposed for the capital budget should be eligible for finance only if the incremental recurrent budget required for the operation and maintenance of the new fixed assets can be accommodated within the expected medium term ceiling on recurrent account, or that alternative systems of cost recovery from users can be effectively introduced. Before an investment is accepted in the capital budget, a firm commitment ought to be entered into by line Ministries about the specific phasing of the cuts of recurrent expenditure on lower priority ongoing activities which may be required to compensate the incremental expenditure on recurrent account expected as result of new investment. It is a commitment the implementation of which must be closely monitored by the central fiscal authorities.

Allocation and Distribution Policy Issues

A major objective of SIPs is the rationalization of public expenditure. It is important to clarify the meaning of the word "rationalization ". All decisions about public expenditure are rational decisions, in the sense that they respond to requests to satisfy public wants. The issue, of course, is who is the public. Indeed, a major problem in drawing up a rational public expenditure programme, i.e. a programme which is consistent with the declared goals, strategies and policies of the Government and with the available resources, is what public administration analysts call "the tragedy of the commons". This is the fact that lobbies (local pressure groups, and, one could add, Donors) have power to get allocations of public funds to implement their preferred projects in excess of what is socially optimum. Solutions to "the tragedy of the commons" involves, besides transparency of public finance information and public discussion of this, strict rules about the ceiling on expenditure (and inclusion of all expenditure in the ceiling), criteria to establish what is the socially (near) optimum claim on public resources of each activity to be included in the budget, and practical ways to measure the implications for the public of selecting one activity and dropping another. Public discussion about such trade-offs would possibly moderate the influence of pressure groups. However, this requires that criteria be simple and that the logic of quantification be relatively manipulation-free and be easily understandable by decision-makers and by the public. Naturally, good analytical tools per se are not sufficient to bring about the rationalization of a government expenditure programme. Tools such as PEA (public expenditure analysis) even when formally adopted by planning units, are seldom internalized beyond those units, and end up not to play the role they could play. The reason is that the rationalization of government spending is a political, not a technical process: without institutional changes which would modify the system of incentives in making decisions about public expenditure, no amount of analytical work, however valid and however convincing, will work in practice.

The allocation function of the budget policy must provide an answer to the major question as to which activities are to be retained by Government and financed by the budget. The public finance theory of welfare economics distinguishes activities which satisfy private wants, social wants, and merit wants . Private wants refer to goods and services which are subject to the exclusion principle, i.e.: the owner/supplier has ownership right to the good/service he sells to consumers, who can acquire them by paying the market price. Ownership is the critical factor, private wants refer to goods which are demanded by individuals, are used exclusively by those individuals who buy them, and for which consumer preferences are set by the market pricing mechanism. Social wants refer to goods and services which are consumed equally by everybody: these are not subject to the exclusion principle, consumer preferences are not determined by pricing, and therefore should be supplied by Government. Merit wants refer to those goods and services which society (through its own system of eliciting public preferences) decides should be consumed by special (meritorious) groups of citizens beyond what they could afford to consume at private market prices. The concept of merit wants is closely associated with distributional issues. The theory recognizes, furthermore, that the borderline between social and merit wants is not always precise, and that some merit wants feature elements of social wants as well.

Having made the distinctions, the theory of public finance of welfare economics does not provide guidance as to which category of wants should be satisfied directly by the public administration, and which ones should be satisfied by private enterprise. In the course of time, development economics practitioners have used a very broad interpretation of the rationale for activities to be financed by the budget, and justified Government investment in activities aimed at fostering production irrespective of the distributional aspect, to the point that production per se has been treated as a social want . At the beginning, structural lending by the WB did not question this interpretation, upon which most projects actually financed by the Bank in agriculture and other sectors had been formulated. This has drastically changed with the SIP approach, to the point that a reverse trend is noticeable these days, which runs the risk of making the opposite error, by disregarding altogether merit wants (and distributional issues) in public expenditure programmes design, except for those which would qualify under the concept of "safety nets".

The theory underlining structural adjustment summarized at the very beginning of this paper calls for social wants to be satisfied by Government, and for private wants to be satisfied by private enterprise. The emphasis on policy elaboration, deregulation and privatization in SIP design responds to the need to correct obvious imbalances currently embodied in the budget structure of many countries entering the SIP formulation process. The extent to which merit wants should be satisfied through the budget, however, is a much more complex issue: in the aggregate, this is determined, on the one hand by the ceiling imposed on the budget in accordance to the principle of sustainability (the stabilization function of the budget policy), and, on the other, by the judgements or estimates of what would be satisfied as private wants by the market. The allocation of the balance among different merit wants is subject to political decisions (the way societies elicit their preferences), about what constitutes merit and about relative merits. These decisions would be more or less democratic, depending of the stage of political development of a country.

In agriculture, typical activities which satisfy social wants would be: setting the rules of the game, enforcing the rules of the game, reporting on the state of the sector, keeping abreast of relevant technology development, providing guidance to farmers as to where they can obtain valuable technical advise. Most other activities of Government in agriculture satisfy merit wants, and this raises the difficult problem of estimating what the market would supply in the absence of Government activities, and of deciding what share of activities such as research, extension, and agricultural education, for example, ought to be classified as a merit or private want in relation to potential beneficiaries. For example: well endowed farmers should pay for the extension services which they wish to obtain by acquiring those services from the private sector (acquisition of technology as a private want). For Government to provide extension services to poor farmers, on the other hand, would satisfy a merit want with public funds. Self-targeting of activities to satisfy merit wants can be obtained by selecting the content of the activities which would interest mostly the target groups (for example: extension of improved animal draft harnessing would interest primarily poor farmers; extension of tractor operation and maintenance technology would be of interest to well-to-do farmers). Some Government activities can be justified by the social wants featuring in merit wants. Consider, for example, vaccinations against major epidemic diseases: although demand by well-to-do livestock owners could be satisfied by private enterprise at market prices, most Governments subsidize vaccinations to ensure that animals belonging to poor livestock owners are also vaccinated, whenever total coverage is technically required for effective disease control. By so doing, Governments treat vaccinations as a merit want. However, all livestock owners, and indeed the economy as a whole, have a precise interest in total prevention of outbreaks of major animal epidemics; in other words, good animal health condition benefits are consumed by everybody, which is the element of social wants featuring in the vaccination activity.

Implementation of Policy Reform

So far, all structural lending by Donors has included policy reform conditionalities. These were initially concerned with the management of the macro-economy and therefore primarily with the stabilization function of the budget, and they have been often complied with by Borrowers. More recent operations have dealt as well with policies concerned with the allocation function. WB experience suggests that Government compliance with more complex agreed policy conditionalities has been generally unsatisfactory, despite statements made to the Board that Governments were "fully commitment" to the loan objectives. This should come to no surprise. In terms of Mintzberg’s theory of organizations, such statements assume that Governments operate under a single typology: passive external and authoritarian internal coalition. This is only one of the many possible situations analyzed by Mintzberg, and possibly the least likely. In fact, no Government can be "fully committed", for the simple reason that no Government, actually no organization, is monolithic, nor is free from external influences and circumstances. Governments are made up of individuals, each pursuing his own objectives, and following his own interpretation of the best way to do his/her job. Governments are subject to external influences, and must respond to changing external circumstances. Individual officers within Governments, Ministers and/or Principal Secretaries, can be fully committed, and often they are. But they cannot promise more than they can deliver in practice. What normally happens is that programmes are elaborated in response to initiatives and with the full support of some influential people in Government, who are confronted with resistance and opposition from within and without the administration. Donors then play the role of countervailing such resistance and opposition, a role which may or may not be determining, depending on circumstances which are not easily predictable. In any case, this is an unhealthy situation, as long as Donors are, appear, or are made to appear the only or major supporters of change, since sustainable change requires a solid domestic constituency. The risk of non-success with policy and institutional reform depends entirely on the balance between the forces acting for change and the forces acting for conservation in the domestic political arena, a situation which is very seldom analyzed in institutional assessment or sector studies.

The above considerations suggest that the rationalization of the current non-project financed expenditure, and the introduction of the institutional changes required to modify the system of incentives in deciding about public resource allocation, ought to be accorded high priority in the SIP formulation process. Progress in pruning and rationalization of the recurrent budget would be an important indication of the capacity of a Ministry to implement a SIP . In several cases "successful" PEP implementation resulted in lower capital expenditure in agriculture, but it is not clear what the impact has been on the recurrent account of the budget.

The SIP Formulation Process for Agriculture

SIP Objectives

The objectives of a public expenditure programme such as a SIP can then be summarized as follows:

• ensuring financial discipline;

• ensuring the consistency of the budgetary allocations to all individual Government activities with the goals Government wishes to achieve and with the sector strategies and policies adopted to achieve those goals;

• enhancing technical efficiency in the use of public funds.

The first objective sets the general constraint under which the programme must be designed. The second calls for the prior definition of goals, policies, and strategies, and for the establishment of criteria to apply in the selection and prioritization of the activities eligible for budgetary allocations. The third objective sets a general condition aimed at improving the productivity of Government services, as a result of which more room would be made for more output/activities to be undertaken within the same budget ceiling. The relationship between the first and the third objective is very important. In practice, it is simply not possible to estimate the financial space for undertaking new activities, or for continuation, beyond the termination of Donor support, of eligible activities currently financed by projects, without having first examined current Government spending, assessed the efficiency of the administration, pruned the unproductive branches, and having identified the means to increase efficiency and to reduce costs. In so doing, the second objective will obviously be brought to bear. Thus the process is closely interlinked.

It is recognized that SIP formulation is a time consuming, long and costly process. This results from the need to gather a great deal of information as well as to develop an understanding of the functioning of the public administration, and to build up consensus about the reforms to be undertaken among several layers of staff in the Government administration, and among institutions in the civil society (Parliament included). In relatively large countries and complex political situations, the time and the resources required to gather information, build up consensus, reach decisions, and mature commitment on institutional reform, tends to discourage Donors under pressure to invest. Then the process runs the risk to be seriously contaminated by shortcuts and compromises, the result of which is that the most needed impact on budgetary procedures, and on searching for acceptable ways for Donors to co-finance the budget, does not materialize. The history of SAPs, PEPs, and SIPs financed by the WB, as reported in the WB’s own evaluation, is indeed a history of compromise. It suggests that, in many cases, the interplay of pressure to lend on Bank Staff, and pressure to borrow on their principal counterparts in Governments, combined with mutual discouragement about the complexity of the task by lenders and borrowers alike, have plaid a more important role than cultural, technical, or ideological constraints in the failure to come to grips with the essential aspects of the public finance pathology.

Steps in SIP Formulation

The first steps of SIP formulation include: (i) the definition of the objectives (goals) which Government wishes to achieve in the sector, (ii) an in-depth analysis of the technical and economic potential and constraints of the sector economy, which sets the background for elaborating: (iii) the policy framework to be adopted to achieve the goals, including the definition of the role of Government, private enterprise, and civil society institutions, (iv) the sector strategies by which the policy framework will be implemented, and (v) the criteria for screening Government activities for their consistency with the goals, policy approach, and strategies adopted .

The technical and economic sector analysis, while covering the necessary macro-economic aspects, must by derived from a good understanding of the producers’ problems, and of their survival and development strategies. The necessary balance between macro-economic and micro-economic aspects, and socio-economic considerations is important. Political and institutional assessments are often missing or superficially dealt with in sector studies. A critical review all current Government activities in the sector must be undertaken, so that a sound basis is provided for the next steps of the process. These include: (i) a review of all current expenditure, Government and ongoing projects, for conformity with policies, strategies, and priority criteria, (ii) the assessment of institutional capacity, including the inventory of personnel and of procedures, (iii) planning of institutional reform (structure, establishment, procedures) required to implement agreed policies and strategies, including cultural change; and (iv) the estimation of a consolidated budget, which would include all current activities to be retained under the SIP. Only at this stage it would be possible to discuss the scope for the Government to assume further commitments on recurrent expenditure within the medium-term budget ceiling, and to plan further expenditure for which Donors finance could be negotiated.

The last steps would then include: (i) the projection of the Government expenditure, on recurrent and capital account for the time period to be covered by the SIP, such expenditure covering all current activities eligible for retaining under the SIP, plus all incremental activities planned, up to the point when all available resources, national and foreign, are absorbed; and (ii) the preparation of a financial plan, in accordance with the (reformed to the extent necessary) Government budgeting reporting and accounting procedures, which would allow Donors to co-finance the budget along with Government own resources, on the basis of the same set of procedures.

Once a consistent set of goals, policies and strategies has been elaborated and agreed upon at the level of the policy makers, it is essential that these be translated into specific functions, objectives, and tasks of the line Ministry Departments, Divisions, and individual Operating Units. This suggestion is not dictated by the belief that management by objectives should apply as the principle governing a public administration: matters are much more complex. Nevertheless, no effective management is possible without a clear definition of functions tasks and objectives, nor is there another way to ensure that the implications of the policy decisions taken, and of the strategy options selected, have percolated through the different layers of the administration, so that all staff understand what they are expected to do and why. Delaying the elaboration of the specific operational implications of policy and structural change to late phases of the SIP process leaves the door open to multiplicity and ambiguity of interpretations, puts the "ownership" process in jeopardy, and undermines effective management of the reforms. Unambiguous interpretations of an agency’s objectives and tasks are essential to develop a sense of mission and to bring about a unified culture within an agency .

Staff workshops are suggested as the best way to bring about staff awareness and ownership of new objectives, policies and functions. The effectiveness of such workshops would be greatly enhanced is they were generally better structured, if their objectives were set up on a realistic basis, and if participants could obtain a prior written brief describing, in addition to statements of general principles, the new detailed tasks proposed for them, as a basis for obtaining comments and suggestions. Some methodological approaches call for staff to participate in the elaboration of the structural changes of an organization, rather than in discussing proposals, in the belief that this would bring about more "ownership" of the reforms, and less resistance to change. Many Donor reports too often associate internal resistance to change almost exclusively with ideology, or with the cultural features of an organization. While ideology and institutional culture are important, and participation is surely a way to influence cultural change, a good deal of staff (and middle management) resistance may well originate in simpler and more concrete preoccupations, such as fear of loosing a job, incapacity to re-cycle oneself in a different context without guidance and training, or confusion about what one’s new role would eventually be. Providing guidance and a clear sense of direction are important features of leadership and management, and neither does necessarily imply imposition.

A set of criteria and sub-criteria must be designed to facilitate a transparent discussion of which activities should be retained by the line Ministry under a SIP, particularly to satisfy merit wants. Criteria must be designed so that they can be used to assess the eligibility, efficiency, and priority of ongoing activities, and to suggest continuation, enhancement, modification of approach, or discontinuation, according to the particular case. Criteria ought to be simple, understandable by Government staff at all levels, as well as by politicians and the general public. New activities to be undertaken under the SIP will be subjected to the same tests. In establishing criteria, we encounter a major problem in SIP design which is the need to distinguish between screening criteria (to establish conformity or otherwise of an activity), efficiency and sustainability criteria (to establish if they are designed to improve the productivity of the administration and how their financing would be met), and priority indicators (to help making judgements and facilitate discussions about which one is more important, among many activities which pass the conformity and the efficiency tests). This matter will be discussed later on.

Once the criteria have been established, the objectives and functions of each Ministry Department, Division and subordinate Operating Unit, and the tasks of the individual posts presently established in each of them must be assessed and, if necessary, re-defined in such a way as to be consistent with the objectives, policy framework, and strategies adopted. The established posts (grades and numbers) must also be re-assessed for consistency with the tasks assigned to each unit, possibly implying changes to the establishment list. As a result, changes to the present structure of MOA, at central and decentralized level, are likely to be called for. Having done that, the vacancies under the revised establishment will be filled by new recruitment, by way of redeployment and training of staff (wherever possible), and surplus staff transferred to other Ministries, disposed of as a result of privatization, or gradually laid off.

By the above process, the Budget Office of the line Ministry would be in a position to correctly evaluate the establishment and to estimate the cost of the personnel required to undertake the programme. A "model" SIP budget can be estimated on this basis, and a programme elaborated so that the pace at which the Ministry will be able to switch from the current situation to the one expected in the future can be negotiated. The Budget Office can then turn to the other items of the recurrent budget. Scarcity of domestic resources combined with failure to prune low priority activities often results in Government staff been paid but given little means to carry out their duties, with the result that the productivity of the administration is extremely low. The marginal returns from increasing the allocations to high priority activities in a way which enables staff to perform effectively are obviously very high, possibly much higher than investing in the expansion of Government fixed assets. This is an important consideration for Donors to bear in mind in shaping their contributions to SIPs. A realistic estimate of the recurrent expenditure of the SIP budget (other than personnel remuneration) would include, in addition to amounts related to the new activities to be undertaken, adequate amounts to pay for all the means which the retained staff would need to carry out their tasks. The basic principle of enhancing Government productivity ought to be observed. This principle implies that staff entrusted with tasks for which it is likely that inadequate resources would be allocated, should simply not be retained. If an activity is eligible under a SIP and has sufficiently high priority, it should receive full allocation for its successful implementation. Otherwise it should be dropped from the list of the activities foreseeable for the medium term, and no resources, staff and/or other means, should be spent on that activity.

In the process of re-defining, re-estimating and evaluating budget allocations, the application of general yardsticks, structural ratio analysis, and other indices utilized in PEA, however analytically significant, is evidently not sufficient. The objective of the exercise is not so much to arrive at a reasonably approximate estimate of the amounts to be budgeted and/or to show that the structure of expenditure has improved or, eventually, that the incremental costs are economically justified: the estimates will obviously need continuous re-adjustment and re-estimating in course of time. The fundamental objective must be to link the targets and the functions of the operating units of the administration (and the tasks of their staff) with the resources made available and the cost of implementing those functions/tasks and of achieving those targets. In other words, a transparent budgeting process must become simultaneously a tool for detailed planning and for monitoring of operational activities. Only then, reporting on the financial performance of the operating units could be meaningfully associated with reporting on their physical progress in implementing the work plan, and thus become a useful management tool, which is not often the case in most administrations. Officers reported disinterest in financial reports often originates in the lack of a precise relationship between work plans and budgets at the level of the operating units for which they are responsible.

Thus a precise definition of the objectives, targets, functions, tasks, output and outcome of the different units of the administration must be arrived at before the SIP is finalized. By so doing, the discussion about a SIP would be brought down to earth, away from the realm of opinions and generalities, where multiple interpretation is the rule, and dug into the concrete typology of what people are actually expected to do. It is suggested that this would be a more effective way to promote and test the "ownership" of the programme, so that action would hopefully follow.

Criteria are used to broadly define the nature of the content of a SIP (i.e. the typology of activities retained). The next step is target-setting. Performance monitoring indicators are then designed to measure the degree of success in SIP implementation (ex post evaluation). A precise definition of objectives targets and output of Government units would greatly help to improve the design of the SIP performance monitoring indicators. Performance indicators deal with two types of output: (i) non-quantifiable output, such as policy changes introduced and administrative action taken (for example: price of maize liberalized, measures taken to privatize Government dairy farms, import permits abolished); and (ii) quantifiable output, such as targets achieved under the work plan of the operating units (number of seeds multiplication groups trained, number of vaccinations carried out, for example). A SIP implementation schedule ought to include both, although the latter must naturally be fixed on an annual basis, and constantly reviewed in the light of the implementation problems faced by the annual work plan. Flexibility is very important and is naturally provided by the process of annual budgeting, complemented by the role that information on current implementation problems will be allowed to play in drawing up the next year’s plan of work and budget.

Words such as output and outcome are too often used as synonymous, which they obviously are not, a confusion which has a parallel in the expression "monitoring and evaluation". The distinction (output refers to products, outcome refers to impact, monitoring the output production process is one thing, evaluating the impact of the output produced is another) must be clearly kept in mind in the design of the performance indicators, of the monitoring system, and of the evaluation system.

Whenever the assessment of impact is brought to bear on the design of performance indicators a switch is made from output to outcomes and from the monitoring function to the evaluation function. Typical outcomes corresponding to non-quantifiable outputs would be, for example: the number of Government dairy farms actually privatized, the price of maize in the market as a result of liberalization, the change in quantity and price of imports as a result of floating the national currency, changing the level of import duties, or abolishing import permits. Outcomes corresponding to quantifiable output would be more complex to evaluate: for example, the evaluation of the seed multiplication groups trained involves assessing the quantity and quality of improved seeds they have produced, the market uptake and the price at which the seed crops have been sold, and eventually the way farmers have made use of the improved seeds they have bought. Mixing output and outcome in the design of performance indicators may lead to confusion, and may result in designing complex indicators which cannot be adequately coped with by an administration. The time when performance indicators must be available is also critical. With respect to output, information is essential to prepare the next year’s programme of work and budget. Some information on outcomes may be useful at the same time, but not essential. In agriculture, most information on outcomes can only be meaningfully collected over a period of time. In those cases, qualitative information, particularly the views of the clients of Government services, would be particularly useful, besides providing an important institutional development on the path towards democratization.

Issues in SIP ex ante Evaluation

An effective public expenditure programme must comply with a number of conditions. It must be within the ceiling of available resources (revenue account allocated to the Ministry plus donor finance). The recurrent budget component must be within the ceiling of the resources that the MOF could be reasonably expected to be able to allocate from the revenue account in the long run on a sustainable basis . It must be conducive to the achievement of the overall Government policy objectives. It must be the result of selecting the least cost solution among options involving different costs. It must be compatible with the output targets the Administration is expected to achieve. It must result in improvements of the productivity of Government staff. It must show an appropriate balance between the cost of the general administration and the direct cost of providing services to the public. Incremental expenditure related to activities the aim of which is to increase taxpayers’ income, must be commensurate with the expected outcome in terms of taxpayers’ income growth. Within the latter category of expenditure, an appropriate and justifiable balance must be struck as between expenditure allocated to different sub-sectors (activities). Progressive or regressive income transfers resulting from the programme must be analyzed and evaluated.

As suggested earlier, criteria and indicators which can be applied to construct and evaluate a budget fall into two different general categories. The distinction between criteria and indicators lies in that criteria are absolute: they consist of questions to which one can answer yes or no, although in several cases an element of judgement would be involved to tilt the scale. For example, an activity does or does not conform with an objective of Government policy, does or does not imply a continuing recurrent cost for the budget, can or cannot be better implemented by the private sector, is or is not designed as a least cost solution. Sub-criteria can be devised to further test if the answer depends on the way an activity is undertaken. Conformity criteria indicate whether a proposed Government activity (and therefore a budget item) fits with social preferences as expressed by the policies and strategies framework.

On the other hand, an indicator is a measurement which has a relative value. Per se, an indicator does not say if an activity should be undertaken or not, neither it says if an activity should be preferred to another. Indicators apply to activities which satisfy merit wants, in as much as ways can be devised to measure the benefits of satisfying those wants. In this connection, an indicator provides a useful basis (among other considerations, and/or by subjectively weighing different indicators) upon which to make a judgement about the relative importance of one activity versus another. Example of indicators are: net impact of an activity on the Government budget, activity effectiveness ratios, economic cost/benefit ratios. None of these should be used as tools to directly rank individual activities in order of priority, but they offer useful guidance to making informed judgements about priorities.

As observed in paragraph 10, it is useful to distinguish between different types of criteria and indicators:

• policy consistency screening criteria: these are derived from the goals, policies and strategies of Government, and test the consistency or otherwise of an activity with those. They must be elaborated ad hoc, depending on the policy framework adopted. For example: poverty alleviation, natural resource conservation, privatization may or may not be goals/policies adopted by Government. Donors concurrence with a Government’s goals, policies and strategies must be discussed at the time of the formulation of a SIP general framework: as a result of these discussions, Donors can decide whether they will support the SIP.

• efficiency and sustainability screening criteria: these aim at ensuring improvement in the productivity of the public administration and at exploring the financial sustainability of the activity under test. These criteria ought to apply to all types of public expenditure programmes, irrespective of the goals and the policies adopted. Example of efficiency and sustainability screening criteria are: is an activity planned as a least cost solution? can the activity be more effectively implemented by another institution, private or public? would the activity require incremental recurrent costs? if so, how would those be financed?

• priority indicators: these are the most difficult to elaborate since to a very large extent they involve subjective judgements. Further distinctions are required here, as between (i) activities which provide services the outcome of which is presumably quantifiable, and activities which do not, and (ii) activities, the impact of which on the distribution of income and wealth is neutral, progressive or regressive.

Efficiency criteria raise the issue of how to measure the productivity of the Administration. In this connection, the distinction between output and outcome of Government activities is useful. Many products of the public administration do not lend themselves to measurement and quantification: for example, the regulatory and policy making functions of a line Ministry; improved accounting and financial reporting; transparency and better governance. Other products do, but the meaning of the quantification is not a sufficient measure of productivity in all cases: for example, one can measure the number of visits to farmers of an extension service, but this does not say much about whether they have been useful or otherwise. Outcomes are also not always easily quantifiable: the difficulties encountered in attempting to measure specific supply responses to policy change is an indication of this. Activities such as agricultural extension, research, animal health services, for example, may produce outcomes which are quantifiable in theory, but only to some extent in practice, partly because of the difficulties of collecting homogenous series of data, partly because of the difficulty of isolating the impact of the specific government activity from that of other factors, including Government’s own policy reforms.

The least cost solution criterion (inclusive of the answer to the basic question of whether Government should undertake the activity directly or through contracts with private enterprise) is not subject to uncertainties about the measurement of outputs. It remains, therefore, the prime efficiency criterion. The danger exists, however, that compliance with this criterion could lead to underestimating the resources required to carry out the activity effectively. This is a common error in which central financial authorities indulge when cutting budget proposals by Line Ministries, without due regard for the implication of the cuts, and by so doing setting in motion a vicious circle which accentuates the inefficiencies of the administration, rather than enhancing it. A way to check if the budget is likely to increase the productivity of the administration is to compare the cost of staff to the cost of other means made available to staff to implement their tasks, and to assess the adequacy of those allocations. At the same time a judgement would be required as to whether the staff assigned to the tasks is adequate, insufficient, or vastly in excess.

While the use of such criteria can be made mandatory by the central financial authorities, the use of indicators in budget optimization should be intended simply as a way of narrowing down the scope for making decisions exclusively on the basis of judgements, not as a way of substituting for judgements. There is nothing wrong in making use of judgement in the allocation of resources under a public expenditure programme: the budget process is a political, not a technical process, and political processes are essentially based on judgements about opportunities. Screening criteria are important to facilitate compliance with the necessary rigour in respecting the budget ceiling. Priority indicators can at best be used as general guidance in political discussion. Pretending, for example, that ERR analysis, or environment impact analysis, should be used to rank activities in order of priority would be not only technically unsound, but also politically wrong.

One often meets with a misconception about the meaning of prioritization in the present context, in as much as it is regarded as a technical matter, in which objective solutions can be provided by adequate analysis. However, whether objective solutions can be really arrived at is far from evident in many cases, and in many other cases it is evident that this is impossible. Furthermore, the role that solutions alleged to be objective can actually play in decision making about allocating a budget ceiling is far from established in practice. In real life, prioritization is the result of bargaining among different forces which determine the final choice. The priority to be accorded to expenses incurred to improve financial control and reporting, for example, cannot be determined by means of objective formulae. Even if it were possible, it would not be a major cause of decisions. The decision would be the result of discussions between the financial authorities (and eventually Donors) and the line Ministries who would have to decide from which other activity(ies) they must withdraw resources to accommodate those expenses, on the basis of judgements, opportunities and the subjective evaluation of the trade-off. This would also be true in the case of activities which lend themselves to be examined on the basis of reasonably objective yardsticks. The attempt to construct general equilibrium models to "optimize" the process would be largely a waste of resources, produce highly debatable results, shift a political discussion which is normally on very concrete, simple terms onto the realm of esoterics, and risk transforming the technicians engaged in "process optimization" into non-politically responsible politicians in disguise, through the influence they can exert on those who recruit them, generally influential Donors.

If the question of the priority to be accorded to activities which do not result in measurable outcomes is essentially a matter of judgement, some analytical tools may help to focus the political discussion about the priority to be accorded to those activities which aim to have an impact on the public economic conditions. Taxation, or any form of revenue producing activity, is an obvious case, but we are not dealing with the revenue side. In the agricultural sector, the question is what kind of analysis should be used to guide the political confrontation in deciding about activities such as crop production, agricultural research, extension, conservation and forestry, etc.

In the latter fields, judgements would have to satisfy, inter alia:

1. the point of view of the Exchequer, who would like to know what would be the ultimate effect of such activities on the financial conditions of the Government;

2. the point of view of the tax-payers, who would like to know what the net gain of the public would be (would they pay more or less than they will benefit?) and whether there would be transfer of resources among different types of tax-payers (who will pay, who will benefit?);

3. and the point of view of the Economic Planner, who would be interested in what impact the implementation of the activity would have on the national economy.

In this respect, the following indicators are discussed:

• net impact on Government budget

• effectiveness analysis

• cost benefit economic analysis

Net impact on Government budget. Expenditure analysis ought to be done in terms of the programme net effect on the budget, that is taking into account potential increases in tax paid, after deducting the tax component of government expenditure, if relevant, and the incremental government revenue expected as result of the programme activities, if any. A synthetic index can be calculated to show the net effect.

Effectiveness indicators. In appraising Government expenditure in activities aimed at bringing about specific benefits to specific groups of people, it is important to devise simple indices which would to some extent combine the benefit approach of public finance theory with the income distribution effect of budget policy. Effectiveness indicators can be estimated to show what benefits can be expected by the tax payers in exchange for the resources withdrawn from them to pay for activities which bring about those benefits. From the point of view of the clientele of Government services (the taxpayers), in agriculture the benefits are measured by the expected increase in the net income of those farmers who participate in Government sponsored programmes. By comparing the present value of the stream of incremental net farmer income to the present value of the total Government expenditure incurred in the related activity, a measure of the effectiveness of the activity is obtained. The balances would show the taxpayers’ gain or loss. Naturally, benefits may occur to taxpayers who are not the same people as those who bear the costs, resulting in a transfer of resources, so that the progressiveness of the public expenditure programme can also be evaluated. The ratio of the two present values indicates to which extent the incremental resources generated by the tax-payers as a result a Government activity exceed the resources withdrawn from tax-payers by Government for that purpose. The ratios (indicators of effectiveness) can be estimated for different activities to provide an insight as to how effectively public money is used in each activity for the benefit of the public, and to evaluate the balance of the ASIP structure by activity from the point of view of the clients of Government services. Ratios greater than one would suggest a public gain, whereas activities showing effectiveness indicators of less than one would need solid justifications on other grounds. Clearly such indicators are not measures of economic returns on Government expenditure, although in some cases they can be used as proxies for them. These indicators are subject to the validity of the assumptions that benefits are entirely due to the related activity, an assumption generally made by all project analysts, but one which in practice is seldom entirely justified.

Economic cost/benefit analysis. This ought to measure the incremental cost versus the incremental benefits of Government activities, costs and benefits having been measured in economic prices. There is no doubt that conventional cost/benefit analysis must be undertaken when relatively large scale projects, involving expenditure on fixed capital formation, are included in an ASIP, irrespective of whether they are meant to satisfy social or merit wants, and provided that economic benefits are quantifiable without excessive use of assumptions and proxies. The need to undertake such analysis in the case of activities which involve large shares of recurrent expenditure relative to fixed investment, is less obvious . An overall economic analysis of the entire SIP has hardly any meaning, since it would mix the cost of activities producing measurable output and outcomes with those which do not. Limiting the analysis to aggregating only the costs of those activities which produce measurable outcomes is a partial solution, also of limited meaning, and possibly a source of confusion, since it will hide cases of cross-subsidization which ought instead to be singled out and discussed in detail in dealing with public finance issues. The economic evaluation of single activities would be more meaningful, were it not for the issue of separating supply responses to policy change from supply responses to Government spending on the activity under evaluation. Furthermore, it is very difficult in practice to estimate what the incremental recurrent cost of individual activities would be: given the way accounts are normally kept in Governments, the analyst can project the future expenditure on specific activities, but not figure out the basis from which to calculate increments with any degree of precision. Finally, the interest in cost/benefit analysis of individual activities depends on the belief that c/b ratios could be used to rank activities in order of priority. Generally speaking, and notwithstanding some opinion to the contrary, this is not a legitimate use of c/b ratios: at best they can be used to identify macroscopic cases of sub-optimal allocation of resources. To do that, however, it is likely that the effectiveness indicators would be sufficient.

The suggestion, therefore, is not to spend undue resources in carrying out economic evaluation of specific Government activities to be undertaken under a SIP, when these do not involve a very large share of fixed capital investment in the total cost, when the activities have passed all tests of conformity and efficiency criteria, when input and output prices include few subsidies, and effectiveness indicators are greater than 1: under those circumstances little extra knowledge would indeed be acquired by adding economic evaluation by cost/benefit analysis. This analysis would be useful when all those conditions are met, but the effectiveness indicators are less than 1, provided that benefits can be adequately estimated . Other tests, such as environmental assessment and related criteria, ought to be part of conformity criteria, if included in Government objectives; otherwise they can be introduced at appraisal as independent adds-on criteria, if so required by Donors. Some Donors may insist on economic evaluation by cost/benefit analysis of their contribution to a SIP: this could be done at the time of appraisal, accepting all the limitations that such analysis faces when applied to the evaluation of so called "soft" projects, and the implication of having to make a great deal of assumptions, in addition to the points made earlier.

Some Differences between SIPs and Projects

It has been noted that, as a result of the inertia built into the frame of mind of officers and Board members used to think in terms of project financing, Donors have the tendency to projectise SIPs. The following is an attempt at identifying some important differences between the two approaches. They are:

With respect to policy formulation and implementation:

in project formulation and appraisal changes in policy are defined as measures required to ensure that expected benefits will be actually achieved: they are presented as issues, requiring assurances by Borrowers. SIPs view policy changes as objectives per se, include a precise schedule of policy implementation to go pari passu or, even, to precede, disbursement of funds earmarked to finance service or fixed capital formation activities retained under the SIP.

With respect to public expenditure:

1. projects deal with incremental costs; SIPs deal with the total public expenditure;

2. projects are costed in line with the concept of components, SIPs must be costed in accordance with the framework of the Government budget;

3. project components put together selected activities to be performed by different units in a Government structure, and require "coordinating committees" for their implementation. SIPs do not have that requirement: coordination is achieved at the time of the annual Work Plan and Budget preparation; implementation supervision is ensured by the regular management structure of the administration;

4. projects require separate accounting and separate reporting procedures, the latter to aggregate expenditure undertaken by different units. SIPs do not have that requirement since they are costed by Government expenditure centre;

5. projects are conventionally designed with a view to foreseeing all details of multi-annual flows of activities, targets, finance, and procurement. Within a pre-defined typology of activities and an agreed envelope of Donor financial contribution SIPs would define specific activity targets and related financial requirements every year at the time of the annual budget formulation and approval.

6. in agriculture and rural development, projects often tend to be multi-sectoral, involving not only different units of a single branch of the administration, but also different branches of the administration, which in turn requires setting up "inter-ministerial coordinating committees". SIPs deal by their nature with a single sector, and hopefully with only one Ministry.

SIPs are broader in concept than public expenditure programmes, inasmuch as SIPs ought to include the contribution of the private sector, and the role of civil society institutions, besides that of Government. These would be taken into account in different ways:

(i) with regard to the private sector:

1. through the Government budget: efficiency criteria require that private enterprise be called upon to supply public services, whenever these could be supplied at lower costs, and/or more effectively, and/or whenever the public intervention is envisaged for a limited period of time. In these cases, finance for the activity could be included in the public expenditure programmes, but implementation would be contracted out;

2. through private contributions to financing a public function undertaken by the private sector and recognized as a SIP activity: this requires specific agreements between Government, the private sector implementing agency, and the private sources of finance: the activity would be included in the SIP, but not in the budget;

3. through private investment in productive activities induced by the more enabling environment which is expected to result from the implementation of policy reforms: these would inevitably remain at the level of planning and of forecasts: strictu senso, they would not be part of SIPs. Progress would be followed up in the context of the regular reporting on the state of the sector economy.

(ii) with regard to civil society institutions:

by way of the institutional arrangements devised to associate, under a SIP, the representatives of civil society institutions in discussions about policy design, public expenditure planning, and, more closely, in the continuous evaluation of government activities.

Developing Rural Financial Markets

(i) The design of the ASIP for the Mountain Areas of Lesotho included a proposal for institutional adjustment and capacity building, an essential component of ASIP design necessary for the successful implementation of the programme and to provide an acceptable unified channel for Donors financial support. Lesotho was considered ripe for an ASIP approach, in view of the progress made, under previous adjustment operations, in the overall rationalization of Government staffing, in bringing recurrent Government expenditure under control, and in view of the very limited number of still ongoing projects financed by Donors in the agricultural sector. The institutional aspect was limited to the analysis of the three Mountain Districts of Mokhotlong, Thaba Tseka and Qacha’s Nek, and to the Range Management Division of the Department of Livestock Services.

(ii) The technical and socio-economic content of the programme was elaborated in a set of documents dealing with agriculture, livestock, and natural resource conservation, along the lines of conventional project preparation. The mission did not address the problem of developing rural financial markets, the subject of an IFAD project which is just starting to operate in Lesotho. The institutional issues to be resolved concerned the adjustments in structure, staffing, budgeting and financial control procedures, as well as the arrangements through which farmers, private enterprise, and institutions of the civil society would play a role in ASIP design and implementation. Important aspects such as those related to procurement and disbursement procedures were not dealt with in details, but left to programme appraisal, pending discussion with the Government and finalization of the ASIP formulation.

(iii) The mission proposal was elaborated with a view to achieving consistency, to the extent considered practical, of:

• stated Government goals with the policy approach;

• goals and policies with strategies, the latter affecting:

• production

• environment

• administration

• policies and strategies with the activities to be undertaken (retained) under the ASIP:

• by Government directly

• by Government through contracting to private enterprise

• by institutions of the civil society on their own;

• the activities retained with the recommendations of the technical and socio-economic studies;

• the activities to be retained by Government with the functions of different sections of the administration;

• the structure of the administration with an unambiguous definition of the functions of each section of the administration;

• the targets of each section of the administration with their staff strength (quantity and quality);

• the procedures for the elaboration of the work plan with those related to the approval of the corresponding budget

• the delegation of authority to commit Government expenditure with the lowest level of the administration at which accounting and financial reporting can be effectively kept

• the responsibility for implementation of the work plan with the level of the administration directly involved in such implementation

• reporting on the implementation of the work plan with the level of responsibility for its implementation

• monitoring of implementation performance with the level of responsibility for general management of activities implementation.

(iv) The below table shows the logical sequence of the formulation process linking Government goals, policies, and strategies with the activity screening criteria, the definition of functions, objectives and tasks of various layers of the District administration, and the relationship with the cycle of preparation of the annual work plan (and related performance monitoring) and with budgeting, release of funds and accounting procedures.

( see Table in PDF format)

(v) District Sections are proposed to become budget preparation and expenditure centres, under the supervision of the District Agricultural Officer and under the control of the District Senior Accountant; it is also proposed that Operating Units be formalized and entrusted with initiating proposals for the work plan, which would be processed through the Section Heads, vetted and coordinated by the DAO, and approved in Maseru. Under the supervision of the Section Heads, the Operating Units would be responsible for implementing the work plan after approval, and for reporting on the progress of implementation. Monitoring of work plan implementation would be the responsibility of the District Agricultural Officer, who would control the Management Information System, MIS. The Operating Units would be responsible for collecting the performance monitoring information the accuracy of which would be checked and inputted into the MIS by the Section Heads. The budget proposal would be initiated by Section Heads after approval of the work plan of the respective subordinated Operating Units. Budget proposals would be vetted completed and coordinated by the DAO, and approved by MOA after checking consistency with the approved Work Plan. The District budget finally issued in the Vote Book would be broken down by District Sections. The Quarterly Warrant issued to the DAO would be sub-warranted to Section Heads, but the DAO would retain his/her responsibility as the Chief Accounting Officer at District level, and the authority to request virements. Financial information of the Status of Fund by Section would be issued monthly by the District Senior Accountant to the DAO and to Section Heads.

(vi) Clients’ participation and the role of institutions of the civil society and of private enterprise would be ensured in several ways, including: a demand driven approach to planning agricultural services, through the "needs assessment workshops"; support to interest group formation; emphasis on the Machobane approach and contracting farmers’ extension in the Machobane system to NGOs; encouragement to small traders to provide agricultural inputs in remote villages through the operation of a revolving fund; support to village paravets; innovative farmers representation in Adaptive Research Committees at national and District level; support to the Wool and Mohair Growers’ Association, promotion and support of Grazing Associations to take over range management functions while protecting the rights of non members; contracting conflict resolution activities in RMAs to specialized NGOs; close association with the Village and District Development Councils in the field of conservation and forestry; beneficiaries participation in the evaluation of the impact of programme implementation.

Definitions

Some of the terms used may be interpreted to mean differently by different readers. The following definitions specify the meaning in which they are used in the present context:

Budget programmes: are defined in the Estimates: Programme 1 = Administration, Programme 2 = Livestock Services, etc. MOA Departments are entrusted with the responsibility to implement a Budget Programme, but some Departments may be implementing more than one programme (Department of Livestock Services implements Programme 2 and programme 9, Range Management, for example).

Sections are the administrative units of the District which correspond to the MOA Departments (Crops, Livestock, Extension = Field Services, etc.). They are responsible for implementing one or more of the Budget programmes.

Operating Units are defined as the subdivision of the District Section having a specific sub-programme to implement. For example: in Crops, staff assigned to the Seed Multiplication sub-programme make up an Operating Unit.

functions: the activities of a sub-division of the Administration. These are:

• either services provided to the clients (output), (example: provide veterinary services: vaccinations);

• or internal activities governing the relationships between units within an agency, between sections within a unit, or between an agency and its clients (example: issue sub-warrant, supplies/reports, store inspection, etc.).

objectives: the immediate result (output) of the service activities of an operating unit. This is a restrictive interpretation of the word and does not refer to the impact of the activity (outcome). This definition is adopted as it allows for a non-ambiguous distinction between monitoring and evaluation.

targets: measure the planned output of an activity of the lowest level of the administration organized as a group, i.e. the Operating Units: objectives become targets when the annual work plan is approved;

tasks: the activity to be performed by the staff of a Section or Unit. Functions refer to units, tasks to individual posts. Some tasks do not refer to services, but to the internal relationships within the Administration.

monitoring: involves the following activities:

• measuring the output of the activities of the operating units

• controlling that the activities of the operating units are implemented according to the work plan

• discussing implementation problems with the management of the operating units and advising about their solution

evaluation: assessing (in quantitative terms if possible, but essentially in qualitative terms) the impact of the activity of an operating unit (outcome): that is, the effect on the units’ clients (behaviour, production, etc.).

The above definitions imply that (most) service activities included in the ASIP will produce measurable results (output).

Efficiency of the public administration is indicated by the total (net) cost of providing a service: it relates costs to output (measurable targets).

Effectiveness of the public administration is the relationship between the cost of undertaking a task (or set of tasks) aimed at obtaining an impact on a target population and the impact actually obtained (outcome).

Management Criteria Applied for the Definition of Functions (Sections), Objectives (Operating Units), and Tasks (staff posts) and for the Layout of the Structure of the District Administration

In elaborating proposals regarding the future structure and staffing of the MOA units, to the extent possible, the following management criteria have been applied :

for defining functions and objectives:

• functions of Sections and Operating Units are defined in a way as to avoid possible multiple interpretations by managers and front line operators

• overlapping of functions of different units has been avoided

• setting of contradictory functions has been avoided

• the function of operating units is to provide services to clients: Government clients are people, generally speaking, the tax-payers: their views are important

• the functions of each operating unit is defined in a simple, concrete and precise way, to ensure that they are clearly understood by all operators

• objectives are set in terms of results, i.e. of measurable outputs, so that they can be translated into targets by the Work Plan and targets achieved can be measured and achievements reported;

for the layout of structures:

• linkages between different units are defined as a result of precisely defined respective functions and precisely defined tasks of their staff

• technical service tasks relating to one subject matter are under one supervisory authority

• communication is essential within and among operating units and between the agency and its clients

• coordination is a workplan and budget formulation matter

• implementation and performance monitoring is a matter of management

• coordinating committees are useful at the planning and budgeting stage, at the stage of implementation they generate confusion of roles, leading to inefficiencies

for defining tasks:

• tasks assigned to one unit/post do not overlap with tasks assigned to other units/posts

• each staff member is responsible to only one task manager

• authority to take operating decisions is delegated to the operators who have access to all the information required to make such decisions.

• the staff of each operating unit is responsible for a single task, or for a number of specific tasks which complement each other as required to achieve the objectives of the Operating Unit

• no single task is shared by staff of different units.

From Goals, Policies, and Strategies to Activity Screening Criteria

The selection of the activities to be retained under ASIP has been subject to a set of criteria which essentially test the consistency with the ASIP objectives, sub-sector strategies and related policy guidelines set by Government.

The ASIP objectives (goals) are hereby recalled:

• poverty alleviation

• food security

• employment generation

These objectives were interpreted to mean that the primary beneficiaries of MOA activities are rural people classified as poor, rural people insecure about their food supplies, and unemployed rural people. Consistency with the stated objectives calls for these to represent the primary target groups of the new policy. This is particularly important in elaborating a policy for the medium term, since a positive impact of general economic growth on the target population is likely to be a long term prospect, and, in Lesotho, will largely be achieved through the growth of non-agricultural sectors. In the Mountain areas, some opportunities exist for sustainable development of smallholder agriculture, such as improved wool and mohair production, potato production, fruit trees (especially apple), inter-relay multiple cropping utilizing ash and farm-yard manure, which offer comparative advantages to poor small scale producers. Consequently, activities having a direct impact on the categories of people defined as target groups have higher priority in the activity selection process with respect to other activities.

The sub-sector strategies adopted by the Government to achieve the objectives have been defined in official documents as:

• agricultural policy and strategy development

• privatization and market liberalization

• land reform and natural resource management

• agricultural diversification

• re-orientation of agricultural services

• capacity building.

The relationships between the ASIP objectives and the sub-sector strategies as stated in Government documents are not self-evident. This explains the first of the sub-sector strategies, which is not a strategy in itself, but an indication that such relationships were still to be elaborated.

An important linkage between the ASIP objectives and the sub-sector strategies which has not been focused upon so far in official documents concerns the definition of the respective role of Government agencies, private sector institutions, and farmers, as individuals and as grassroots institutions, in the implementation of the sub-sector strategies. This gap had to be filled by further elaboration, so that the policy framework would be more clearly qualified. Further elaboration was also required with reference to the principle included in the general policy framework published in official documents, headed "efficient resource utilization", which naturally cannot refer only to productive activities, but ought to apply to service activities as well, including those provided by Government. The following set of general policies summarizes, in a synthetic manner, statements included in several GOL documents, and reflects the spirit of the Lesotho ASIP approach:

• divestment of Government activities that can be performed more effectively by private enterprise, including farmers

• mobilization and promotion of private sector initiative to provide agricultural services

• involvement of all stake holders in a participatory way in defining government programmes

• closing or modifying activities which are not technically sound or economically justifiable

• closing or modifying activities which are not socially feasible or undermine the livelihood of the primary target group

• adoption of cost-effective solutions

• improvement of Government resource allocation, budgeting, accounting and auditing procedures

• decentralization of public services

• enactment of laws required to achieve the stated policy objectives, and to implement the strategies and sub-strategies, in accordance with the stated policy framework

• institutionalization of a forum for beneficiaries of agricultural services to participate in the evaluation of the impact of those services.

The above policies reflect the principles of good governance: accountability, transparency, participation, and control. They include principles of equity as well, for consistency with the stated general ASIP objectives.

Criteria Applied for the Selection of the MOA Activities to be Retained under ASIP

Strict adherence to the main ASIP objectives would be too restrictive in defining which Government activities should be retained under ASIP. Clearly, the MOA must retain a number of activities which are not related to poverty alleviation, household food security, or employment creation. Some activities would affect segments of the population which are poor as well as others which are not poor. Other activities may have only a very indirect effect on poverty, food security or employment. Some activities may be conducive to further impoverishment of poor people, particularly in the absence of mitigating measures, and, more importantly, of legislation protecting poor people and effective means of enforcing such legislation.

For the Mountain Areas, the criteria used aim at: (i) identifying all activities to be retained under the ASIP for support through the Government budget, and (ii) suggesting the priority to be attached to each activity, such priority being determined by the extent to which the activity can be linked to the general ASIP objectives, i.e. directly benefiting the target group . The following criteria were applied:

Does the activity contribute directly to poverty alleviation, household food security or employment generation, by providing direct benefits to households that are poor and food insecure??

Does the activity contribute indirectly to poverty alleviation, household food security and employment generation through production of intermediate goods and services (such as seeds and extension advice) consumed by the poor?

Does the modality of implementation of the activity reflect ASIP sub-sector policies, namely:

• Can it be more effectively undertaken by private enterprise?

• If so, how can private initiative be mobilized?

• Does it involve the beneficiaries in the planning and implementation of the activity?

• Does it result in a permanent cost of the Government budget?

• If so, can Government afford to sustain this cost?

• Is it (can it be) designed as a least-cost solution?

• Does it require enacting new legislation to be effective and/or new procedures of effectively implementing existing laws?

Does the activity increase production on a sustainable basis, independently of what category of farmers is most likely to benefit?

Is the technical feasibility of the activity adequately proven?

Is the activity financially sustainable (at both beneficiary and Government level)?

Does the activity support ASIP sub-sector strategies:

• Does it result in privatization and market liberalization?

• Does it improve security of land tenure?

• Does it improve natural resource management?

• Does it contribute to agricultural diversification?

• Does it result in re-orientation of agricultural support services?

• Does it improve Government financial practices?

• Does it contribute to decentralization of public services?

• Does it contribute to capacity building?

Application of the Management Criteria to the Structure of the District Agricultural Office

The function of the District Agricultural Office is to implement the Government programmes listed in the MOA Budget, namely:

In addition, some animal production units of the Livestock Services Department (Basotho Pony Farm Stud) are centrally controlled from Maseru and financed under the DLS budget, others (ram studs) are District activities and are financed under the District budget.

The Range Management Division of the DLS operates in the Mountain Areas directly from Maseru: it finances and controls all the established RMAs, in addition to undertaking range inventory and cattle post adjudication activities.

Re-structuring in accordance with the management criteria set out in paragraph 8 calls for a minimum of rationalization:

The current separation of Livestock Services and Range Management has historical origin, since the RMD was set up as an independent unit under the USAID financed LAPIS project. This arrangement generates lack of clarity as between the functions and the tasks of the Range Management Division (HQ) and those of the decentralized RM sections at District level. The re-structuring assumption is that the Central Unit would undertake all activities which regard range inventories (on established RMAs and new RMAs), cattle post adjudication, related data bank establishment and mapping, drafting legislation with regard to RMA and GAs, as well as the community development work leading to the establishment of new grazing associations in new RMAs. The District staff would service the existing GAs, ensure the respect of the RMA rules by the associations and the non-member residents in existing RMAs, assist in dispute settlement, as well as promote the idea of establishing RMA/GAs in new areas, the latter to the point when sufficient interest has been stimulated for the farmers to request the intervention of the RMD. The RMD would naturally provide the necessary backstopping to the District RM Sections, upon request, or as a result of supervisory inspections, will associate District technical RM staff to the range inventory work on existing RMAs, but the primary responsibility for support to existing GAs/RMAs would be transferred to the District Range Management operating units. This is not the case at present with the result of leaving the function of the District units rather inadequately defined. A more rational arrangement, respecting the management criteria, would be for the Livestock and Range Management Sections to be formally merged at District level, under one Livestock Services Section, which would be symmetrical to the arrangement at HQ. This measure would integrate the RMA activities with the other livestock services offered by the District. Range Management ought to become visibly an integral part of extensive livestock production system, while today it is seen as an independent approach: the proposed structural reform would give a clear indication of that. At the same time, the respective tasks of the Central and District range management units would be precisely defined and understood.

The responsibility for running the Animal Production Units of the Livestock Services Department is sometimes divided: some units are transferred to the Districts (and financed under the DAO budget), others are not, according to their historical origin (GOL or Donors’ financed). Application of the policy criteria calls for privatization, thus the existing arrangement is to be considered a temporary solution. It has been assumed that, on a temporary basis, the units will be administered at District level.

The District Crops Section is responsible for supporting the operations of Technical Operation Unit, which taxes heavily the scanty human resources of the Section. MOA policy is to privatize the tractor services. It is assumed that this will remain a responsibility of the Districts until such a time as the equipment will be disposed of. The District will retain the Mechanization Officer, to run the District service tractor (one tractor is required for general work in the Districts) and to implement the draft animal equipment testing and extension programme proposed in WP 3A (Crops).

Alternatively, and possibly a better solution, all Production Units to be privatized should be retained under the responsibility of the corresponding Headquarters Department, and financed directly from the MOA central budget, until fully disposed of.

Adaptive Research. Under ASIP, three different programmes will be implemented: (i) the testing of new varieties of poplar and willows, which will be undertaken by the Conservation and Forestry Sections (tree nursery), (ii) testing (and extension) of animal draft equipment, which will be undertaken by the Crops Sections (the new task of the Mechanical Officer), and (iii) a more complex programme on crops (food and fodder) and on integrated pest management, involving work on the stations and work with farmers on their own fields. This will need close technical support and implementation supervision of the Research Division of the DFS. However, the day-to-day operations of the research stations and the routine follow up of the on-farm trials programme cannot be effectively managed from Maseru. On the other hand, the Work Plan and the evaluation of the results is a function of the Research Division of the DFS. ARD will appoint a Senior Agriculture Research Officer (SARO) in Thaba T’seka, reporting the Director, ARD, whose task will be to design the research programmes (station and on farm, the latter on a participatory basis) and to evaluate the results. Routine station management and follow up on farmers work must be entrusted to technical personnel, and the natural locus of them in the Districts are the Crops Sections. Staff will be appointed to the Crops Section to undertake this task, under the technical supervision of the SARO, Thaba T’seka, and with the logistical support of the DCO.

The re-structuring assumptions do not involve major changes of the existing arrangements, but some rationalization and clarification.

Planning, Implementing, Reporting and Monitoring District Activities

The District ASIP. Under ASIP, the decentralization policy of the MOA will be implemented further. Within the framework of the overall national ASIP, each District would formulate their own 3-year rolling ASIP, for the implementation of which they would be responsible under the decentralization policy. Farmers demands for services will be obtained through the ‘needs assessment workshops" and other RRA/PRA methods. A 3-year rolling plan would be prepared, with consultants’ assistance, the purpose of which would be to guide the preparation of Annual Work Plans, which represent the basic operational document governing the District activities and the allocation of Budget resources. Annual Work Plans will be adjusted in the light of reports on performance and implementation problems, and of reports from Programme Evaluation, and activity in which programme beneficiaries are expected to play an important role.

The Annual Work Plan. The activities of the District Agricultural Office are set in accordance with an annual Work Plan. This is prepared well in advance and the document used by the DAO as a basis for elaborating the annual Budget proposal which is submitted to the Director, DFS as a separate document. In theory the two documents should be consistent with one another, that is the resources appropriated under the Budget should be adequate and allocated to the activities approved under the Work Plan in a consistent way. In practice, the Work Plan is regularly approved, while the budget proposal is not. DAOs receive the first quarterly sub-warrant from the DFS some time after the overall Ministry appropriation is approved by Parliament, and calculate the total budget available for the year by multiplying the sub-warrant by four. Invariably the warranted budget allocation is below the request, but no adjustment is made to the work plan to reflect the impact of reduced resources, and a revised Work Plan is not prepared.

Currently, the annual work plan is initiated by the DAO, who consolidates the proposals made by the Heads of the Sections. Little guidance is received from the Centre about how to prepare the Work Plan. New policy lines of the MOA and previous year performance are seldom taken into account in the preparation of the Work Plan. The Work Plan submitted by the DAO is actually reviewed by the MOA Departments for the Section of their competence, approved and returned to the DAO by the DFS. In practice the Work Plan of next year tends to repeat the one of the previous year.

Under ASIP, the present procedure would be strengthened: the Work Plan formulation initiative would begin at the level of the Operating Units, under the coordination of the Section Head who would then be responsible for translating the approved Work Plan into a budget proposal. The DAO would retain his/her function of overall coordination and control at District level of both Work Plan and Budget proposals. The annual Work Plan would thus become the essential tool of coordination of the District activities and of implementation of the District functions that it is intended to be. Decentralization offers more opportunity to District staff, but imposes more responsibility too. A more pro-active approach is expected of the District administration as a result of less financial dependence of the Centre. The relationship between the activities of different Sections, their consistency with the MOA policies and with the District approach interpreting those policies, are determined at the time of formulating the Work Plan and formalized in that document. The Plan must become the occasion to discuss the performance of the previous year, to examine critically the reasons for achievements and for lack of success, to review the utilization of the budgeted resources by each units, and to decide which activity should be accelerated and which one planned at a reduced pace. It is important that the preparatory work be done in a participatory way, so that communications among managers is enhanced, conflicts resolved, and a uniform culture develops within the District administration. The instrument of a participatory approach to the formulation of Annual Work Plan and Budget does exist in the MOA District Management Team.

The decentralization of the budget to District level ought to become an occasion for improving the current practice also. This requires a more pro-active role of the Centre, who would issue policy and priority guidelines, as well as comments on previous year performances, and discuss them in the annual Workshops which are currently organized by the Departments with their respective District Section Heads. It also requires initiative from the DAO and the Section Heads to provide solid evidence to forcefully defend their plans and obtain support for the budgetary requests, which, as of next Fiscal Year, will flow directly to them from the PS MOA, rather than indirectly through the DFS.

The definition of the objectives, functions and tasks suggested in the Tables attached to this Working Paper could be used to re-define the activities and the targets included in the Work Plan. The format with which the Annual Plan is submitted does not need to be changed.

The Management Team. At District level, the Section Heads meet regularly to exchange information about the District activities, and to discuss progress and problems encountered in the implementation of the Work Plan . The presentation and discussion of their proposals for the next year Work Plan is part of the Term of Reference of the Management Team. Weekly meetings are called for to arrange the allocation of transport, a requirement resulting from vehicle shortage. With ASIP, a more adequate vehicle fleet would substantially reduce the need for such meetings: in as much as they would still be required, they should be chaired by the Executive Officer-logistics, as part of his routine assignments.

The Senior Accountant, to be posted shortly to each Districts, should be a member of the Management Team. He would provide information about the use of warrant resources to each Section Heads, so that financial information becomes part of the management tools available to Staff responsible for the implementation of the Work Plan. To this end, the splitting of the District Warrant into sub-warrants issued to each Section is advocated in Section C.

The Adaptive Research Committees. At present the Agricultural Research Network (NARN) provides a forum for exchanging opinions and information on the general subject. It is proposed that a sub-Committee of NARN be established to deal with Mountain Agriculture at national level. This Committee would be chaired by the Director ARD; members would include the DAOs, the SARO -Thaba T’seka, FSD, DC, and other MOA department officers involved in research, representatives of NGOs working in mountain areas on agricultural problems, and a representation of Innovative Farmers. The ARD would undertake PRA studies to ensure farmers’ input in the agenda of Mountain research. At District level, the circulation of information about the Mountain Adaptive Research Programme within the District Administration and among interested parties in the civil society will be enhanced with the establishment of a Research Committee in each District. Members would be the Management Group, the Officer in charge of the District Station, the Senior Agriculture Research Officer posted in Thaba T’seka, representatives of NGOs providing agricultural services in the District, and a selection of Innovative Farmers. The role of the Committee would be to create awareness, to disseminate information, to provide advise to the Senior Research Officer, ARD Thaba T’seka, in charge of the programme design and evaluation, and to ensure that farmers views are included in the research agenda.

Implementation responsibility. Front line operators are ultimately responsible for doing the job. In the District Administration, the chiefs of the Operating Units would have the primary responsibility for implementing the work plan: the Seed Multiplication Officer will be responsible for implementing the work plan of the Seed Multiplication Unit, which he/she has initiated to formulate; the Forestry Officer will be responsible for implementing the plan of his/her Unit; the Extension Officer for the routine operations of the Area Resource Centres, and so on. Some activities, however, are to be implemented at Section level, for example, organizing the multi-disciplinary "Needs Assessment" workshops, which would be the responsibility of the Head of the Extension Section. The details are shown by the description of all tasks of the staff posted in managerial position in the series of tables DA...-3.

The work of the Operating Units is supervised by the Section Heads, to whom the chiefs of the Operating Units would report. The overall responsibility for implementing the entire District work plan rests obviously with the DAO, who is also responsible for the work of the Administration Section. With respect to the technical Sections, the DAO would exercise a monitoring function, keep abreast of problems in implementation, and assist Section Heads to overcome the difficulties encountered.

Progress Reporting. Progress reports on the implementation of the Work Plan are meant to be submitted monthly, but the schedule is seldom respected. The format of the reports currently used is adequate, but the information requested needs revisiting since in many cases information on output is mixed with information on outcomes, and the latter cannot be provided in the same way as the former. The format of reporting and its content could obviously be changed to provide much additional information to satisfy the requirement of sophisticated Management Information Systems. Considering the difficulties currently encountered in adequately filling the present forms, submitting them in time, and analyzing the information, any attempt at sophisticating the forms of reporting would probably be self-defeating. With ASIP, responsibility for collecting the basic information on the output of the administration (achievements against set targets) would be entrusted to the chiefs of the Operating Units. Data would be checked by the Section Heads for accuracy and inputted by them into the MIS. DAOs would have to ensure that the forms presently used are properly, accurately and timely filled, and that the MIS is loaded timely.

A simple method of computerizing the information contained in these reports would be introduced, to make possible the storage and the analysis of the information. In designing the method, the data will be arranged in such a way as to supply a summary report on the degree of achievement of the objectives of each Section, in terms of the targets (output) of the Operating Units included in the Annual Work Plan, which is essential information for monitoring performance. The possibility of combining such information with data about the use of funds, without adding complications to the accounting procedures, would be explored. In this connection, however, a note of caution similar to that voiced about the performance reports is in order. Information about the aggregate use of funds by Sections would be relatively easy to obtain if the recommended sub-warrant procedure is introduced: a breakdown by operating unit would involve introducing much more complex accounting methods, which would hardly be manageable. Financial reports by Section would be sufficient, however, to provide a useful management tool to Section Heads. MOA would recruit a Consultant to design the computerization of the progress reports and train staff on the use of the software.

Performance Monitoring. Monitoring of performance is an essential management tool. The prerequisite for effective monitoring is the precise definition of the target objectives (output) set in the Annual Plan. This needs to be checked by the authority which vets and approves the Annual Plan: for the District Plans, the DFS. It would be the DFS responsibility to review the Annual Plans, and to ensure that measurable output targets are properly defined, realistically estimated, and accurately quantified. To do that in a coordinated manner (coordination is essential at the planning level, as much as it is almost impossible to exercise at implementation), a multidisciplinary working party would be called by the Director DFS, in which the Technical Departments will participate, each one for its area of competence. This proposal represents a modification of current practice, whereby the technical Departments review the District submission separately. DFS will also retain responsibility for keeping record of the meetings, along with copies of the computerized performance reports.

The results of the activities will be measured and compared with the targets by way of the computerized analysis of the District Sections and operating units progress reports as proposed above. The DAO (as District performance monitoring officer) will spot check a random sample of reports to control their general accuracy and control the data inputted. The software would automatically show the divergence between targeted output and actual output recorded.

Routine analysis and discussion of the performance monitoring reports ought to be undertaken on a quarterly basis by the DAO so that implementation problems are spotted, discussed, and remedial action taken in time. A more frequent schedule would be difficult to follow in practice. However, spotting of problems and reporting to the DAO on an ad hoc basis would naturally be part of the normal task of all managers and of the monitoring officer. The computerized recording of the information contained in the progress reports would produce cumulative monthly totals of the measurable targets, and print out the column "remarks" on monthly basis (three columns for each quarterly report). The DAO would use the synthetic computer reports for internal monitoring of the Section and Operating Units activities, on a regular basis, and discuss them with the Heads of the Sections.

Under the leadership of the DFS, a multi-disciplinary team, composed of representatives of the Technical Departments of the MOA Central Office, would provide external monitoring assistance. A unit would be set up by the DFS in Maseru for this purpose, staff trained to perform effectively the task, including reporting on the findings of the monitoring team and keeping the records of findings, and of remedial action proposed and taken up to date.

Excerpts from the Formulation Report: Detailed Analysis of District Administration and Crops Sections

The FAO Formulation Report on the ASIP for the Mountain Areas included a detailed analysis for all the major Sections of the District Administration, namely: Administration, Livestock and Range Management, Crops, Extension, Conservation and Forestry, Marketing, in addition to the Range Management Division of the DLS. As an example, the analysis and proposals for two Sections, Administration and Crops, are presented below.

Administration (Programme 1). The functions of the District Administration are:

1. the preparation of the District Annual Work Plan and Budget

1. reporting on the implementation of the Work Plan

2. the administration of the Warrants issued to provide resources for the implementation of the Annual Work Plan

3. certification of approved expenditure under the Warrant

4. record keeping and reporting on the use of fund

5. control of the District Stores and reporting

6. personnel

7. adequate maintenance of public assets

8. adequate maintenance of vehicles, supervision of vehicles operations and drivers.

These activities are not related to the main ASIP objectives but support the entire range of an agency’s scope of work. As a result of the recent Government decision that Districts will become primary budget centres (Department level), primary warrants will be issued to them rather than sub-warrants of the DFS. This decision shows that considerable progress has been made towards a positive answer to question (f) and (g) of the fourth criterion. Improved staffing is needed for the policy to be implementable in practice. The major changes proposed are:

• the appointment of a Senior Accountant (decision already taken);

• the appointment of an Executive Officer, Logistics, responsible for the maintenance of all public assets belonging to the MOA within the District territory;

• the appointment of an Assistant Storekeeper;

• the reduction of Office Attendants to the number actually required to perform the work.

The staffing pattern and the estimated expenditure on salaries of the establishment of the Mountain District ASIP model DAO Office are shown in the following table:

The Senior Agriculture Research Officer and his Assistant (Thaba T’seka Station) would be appointed by the Agricultural Research Division, DFS - Maseru and will not be part of the District Administration.

The staffing pattern and the estimated expenditure on salaries of the establishment of the Mountain District ASIP model are shown in the following table:

Clients’ demand. The basic approach to the Mountain Areas ASIP design is, to a large extent, though not entirely, a demand driven approach. From the institutional point of view the following practices would apply:

1. "needs assessment" workshops: these are specifically designed to obtain insights into the requirements of the rural people; the recent experience has been that "felt needs" are essentially not related to agricultural development, but concern rural infrastructure such as roads, schools, clinics, footpaths and bridges. Under ASIP, the workshops methodology will be modified to enable farmers to better focus on the menu of technological solutions/activities which MOA and NGOs could offer in the agricultural field; and, conversely, for the MOA and NGO officers to understand better the most urgent production problems farmers are faced with and would like to obtain suggestions as to possible solutions;

2. "interest groups": these are formed as a result of needs assessment workshops or other initiatives, and will provide the basis for planning Government or NGOs activities in response to the interests of those groups: they will play a special role particularly with respect to activities in the field of Conservation and Forestry, Crops (Horticulture and Seed Multiplication), Extension (Nutrition and Horticulture), Livestock (sheep dosing campaigns, paravet programme), Range Management (Grazing Associations), and training in Machobane System;

3. the re-defined functions, objectives and tasks of the extension section of the District administration aim at providing a bridge between innovative farmers and SMS and, through them, with the adaptive research programme, thus contributing to focus on clients demand in activities planning;

4. the Mountain Areas Adaptive Research programme has been designed as a two prong approach, which combines on-station research with on-farm trials; participation of representatives of Innovative Farmers to setting the research agenda and monitoring of results have been specifically included in the ASIP design (the proposed Adaptive Research Committees at the level of the ARD and at District level);

5. VDCs and DDCs are been developed by the Ministry of Local Government, although not without hesitation and some ambiguity of approach, into democratic grass-root rural institutions, through which elected representatives of the rural people will be able to influence decisions about the use of resources at local level. Though this will primarily regard the LHWDF, VDCs and DDCs could also influence the work programme of the MOA District Administration, including that sub-contracted to NGOs.

For decentralization to become an effective tool of improved performance, better outcome, and good governance, it would be important to secure an ever increasing participation of the of the Institutions of the Civil Society and of the clients of Government Services in the planning of the services activities which they demand. As indicated above, some institutions already exist to this effect. During ASIP, an effort will be made to discuss, propose and test ways to make participation closer and possibly to formalize and input of clients in the preparation process of the District Annual Work Plan.

Mobilization of private enterprise. The following institutional arrangements exist and/or are proposed to be strengthened/established to this effect:

1. Wool and Mohair Growers Associations already perform an important role in organizing the Government Woolshed operations; they also contribute to some extent to their costs; an ever increasing role to be plaid by these Associations is envisaged under ASIP;

2. Grazing Associations are being developed and promoted as private concerns designed to take over functions so far performed by public authorities, Government and/or Chiefs, in the allocation of grazing rights, grazing control, and provision of basic services, such as distribution of dosing drugs and of supplementary feed, organization of vaccinations, improved animal purchases etc.;

3. contracting farmers’ education in the Machobane system to NGOs: this involves supporting with contracts the strengthening of the Machobane Foundation (to be shortly initiated with IFAD and the Swiss NGO HELVETA with separate funding), as well as contracting other established NGOs which could assist in this field;

4. involving private sector suppliers of micro-irrigation equipment in the Horticulture demonstration and extension programme;

5. the proposed establishment of a fertilizers and agro-chemicals revolving fund to be set up to promote and establish the distribution of inputs by small scale traders in remote villages;

6. the proposed paravet programme aimed essentially at privatizing an enhanced sheep dosing effort.

Clients’ role in Work Plan evaluation. Consideration will be given to establishing a formal channel for beneficiaries of the Government activities to make their opinion known about the effectiveness of Government activities to comply with their demands.

Administration Cultures, Leadership and Training Requirements

The implementation of ASIP would require some cultural changes within the Ministry Departments and at the District level. Government is aware of this, although awareness varies considerably from Department to Department, and from operating unit to operating unit within Departments and District Sections.

At District level and very broadly speaking, several cultures could be identified by an in depth assessment of the Administration. Briefly, the following features have been singled out:

1. people are motivated wherever their tasks are clearly identified and sufficient resources made available for them to perform their tasks (seed multiplication, nutrition extension, for example);

2. people are also motivated wherever the task (even if poorly identified by the agency) is clearly interpreted in the mind of the staff as a precise professional connotation (range management in one District, for example). This applies even when insufficient resources are allocated to perform the task;

3. staff is motivated by opportunities to contact villagers about their needs, and frustrated about the lack of means to respond to requests;

4. lack of interest in the job is dominant wherever procedures are called for which produce no tangible results, either in terms of interest from supervisory units (the annual work plan and related reporting) or in terms of resources allocated to a task (the budget is not related to the annual work plan);

5. staff is inoperative when little guidance is received from supervisory bodies about policy guidelines and about priorities for activities to be undertaken;

6. whenever the resources allocated to implement the tasks are well below requirements, staff disappointment is high and considerable lack of interest develops, to the point of not performing even the minimum of tasks that available resources would allow (transport availability in every District);

7. some staff have a tendency to interpret Government functions in terms of instructions (you must do this, all of you) or of prohibition (you must not do that), rather than in terms of governance (let us agree on what should be done, all of us). Instructions form the Centre to Districts are often ambiguous and generate misinterpretations which tend to reinforce the top-down approach rather than governance;

8. conflict resolution skills are poor among staff, leading to escalation;

9. technicians (crops in particular) view local innovations with suspicion and do not appreciate that there may be "scientific" solutions which do not call for the use of modern inputs and equipment imported from the western menu of technologies;

10. Department Directors, District Section Heads, and staff responsible for operating units do not view the budget and the reports on the use of funds as an important management tool: this leads to accountants working in isolation, to lack of control on the accuracy of financial reports, to poor budgeting practices, sometimes to failure to exploit the flexibility of the system of virement by some managers.

In defining functions objectives and tasks of the Administration and the role of Government and NGOs, the positive and negative features summarized above have been taken into account. Action will be taken to mitigate the negative aspects through ASIP.

Some remedial action would consist in the implementation of ASIP itself, by providing a clear definition of the activities to be undertaken, re-establish an adequate annual planning procedure, provide sufficient budget resources for the implementation of the annual plan, and mobilizing Centre Department support to District activities through issuing proper guidelines for the preparation of annual work plan and the performance monitoring process.

Other remedial action would involve training of staff, and running of re-orientation courses. At District level, however, the DAOs would play a most important role, as they will provide the required consistent interpretation of the ASIP policy, and the leadership necessary to establish a uniform culture in the District administration which would be in line with the ASIP goals and objectives. Thus the selection of the DAOs and their briefing is an essential matter in ASIP implementation.

C. The Recurrent And Capital Budget: Procedures, Issues And Improvements Required to Become an Effective Channel for Disbursement of Donor Support

General Procedures: Need for Unification of Recurrent and Capital Budgeting

In the MOA Recurrent and Capital Budgets are prepared by different Departments, and follow different procedures. This reflects the fact that, while the Recurrent Budget is approved by the Ministry of Finance, the Capital Budget is approved by the Ministry of Planning. Since the Capital budget essentially consists of Donor financed projects, confusion arises on several grounds. Most MOA projects include a large share of recurrent expenditure, which is budgeted for in the Capital Budget, resulting in an underestimation of Government commitments on this account.

The Recurrent Budget proposals are prepared by the Technical Department and coordinated by the DPS: the consolidated MOA proposal is cleared with the PS, who is the chief financial officer of the Ministry, and forwarded to the MOF by the PS. MOF proposes changes (mostly cuts to keep within the ceilings imposed on the Ministry allocation). These are discussed with the MOA and a final version, cleared by the Cabinet, is included in the finance bill for parliamentary approval. Once the estimates are approved, the Vote returns to the MOA (the Vote Book), and funds are released by Treasury with quarterly warrants.

The PS/DPS re-allocate the Ministry quarterly warrants to the Departments. The Department of Field Services, responsible for the District Agricultural Offices, issues quarterly expenditure authority to the DAOs, allocating the Departmental Warrant by ways of sub-warrants.

The Ministry Financial Controller, a staff member of the Accountant General’s Office, supervises the accounting staff posted in the MOA. Accountants are posted in all the Departments and in all Districts, although the staffing is generally inadequate and, at the moment, particularly so in the Districts. Departmental and District accountants keep the ledgers, record the payment orders issued by the Certifying Officers under the warrants for their respective unit. The Financial Controller checks that all payment orders are issued in accordance with the established procedures and that funds are available under the Vote and authorized by the warrants. A monthly statement of the Status of Funds is prepared by each Department accountant for the Financial Controller, who produces the consolidated monthly statement on the Status of Funds. This document records the total of the payment commitments entered into by the Ministry against the Vote and the corresponding Warrant items, showing the monthly balances of un-committed funds.

The Capital Budget is prepared by the Department of Planning, Economics, Marketing and Statistics, and controlled by the Ministry of Central Planning. The DEPMS, in addition to the many functions listed in its title, is also responsible for the administration of Donor-financed projects. Over the last many years, the Capital Budget of the MOA consisted essentially of project financed activities. As these include a substantial share of items of recurrent expenditure, the Government commitment on this account is not consolidated. Elements of project costs, however, are also included in the Recurrent Budget (counterpart costs). Each Donor financed project has its own budgeting, disbursement, accounting and financial control procedures, different one of the other, and differing as well of GOL procedures in most cases.

The DEPMS responsibilities have grown considerably over the years, while its staff has been seriously depleted. At present, its functions range from the elaboration of general policy documents to market analysis, from overall sector planning to drafting legislation, from delivering market information and elaborating agricultural statistics to planning, budgeting, and controlling a fair number of development projects. As a result of having to perform so many diverse functions, and with an incredibly small work force, performance leaves a good deal to be desired.

The MOA will consider separating the policy formulation, legislation advisory, and sector planning functions of the DEPMS (including the related statistical work) from those of preparing the annual capital budget, and of supervising the performance on the capital budget. Consideration will be given to contracting market studies and information services to specialized private enterprise, as it has been done successfully in several other African countries. These measure would allow the limited staff of the DEPMS to concentrate on the their specific functions (statistical analysis, sector planning and policy elaboration).

At present, there is no Budget Office in MOA. Under ASIP, however, "project financing" would be discontinued, and Government expenditure, under recurrent and capital account, would be consolidated, irrespective of the source of finance, and irrespective of which is the financial Ministry of reference. Recurrent costs would be budgeted under the recurrent side of the budget, up to the limit of long-term financial sustainability set by the MOF and the MOP. Items of expenditure which are of a recurrent nature, would be charged to the capital budget only when such expenditure is to be incurred for a limited period of time and this is reflected in limited contractual obligations for the Government (see later: Section D). The preparation of the recurrent and capital budgets would be unified and a new unit (the Budget Office) established under the DPS, who would de facto exercise the functions of Joint-Secretary (Finance), in addition to being in charge of Administration. Staff would have to be recruited to man the Budget Office, and trained in computer use, and in the preparation of adequate documentation supporting the budget proposals. Staff would also be trained in the interpretation and use of the reports of the work plan monitoring units, which would become the basis upon which such documentation would be prepared. The Financial Controller Office would be part of the Budget Office, and assume direct overall responsibility for both Recurrent and Capital budget control.

While past Donor support has stressed the importance of planning (overall planning, sector planning, land use planning, etc.) and considerable resources have been invested in strengthening some planning aspects (land use for example), little attention has been given to less glamorous aspects of the public administration, such as annual budgeting, financial control, and auditing. Yet these are the building blocks of the performance of any organization. They set the limits of what an organization can do in practice, responding to the essential requirements of public finance management and control.

Despite the shortcomings identified, the MOA has made considerable progress in the management of expenditure. This is all the more commendable since it has been achieved against very tight constraints in terms of staff and staff quality, and without the help of the most elementary mechanization. There is considerable need for external assistance in terms of technical advise, equipment, software and training associated with the required improvement of budgeting, performance monitoring, financial control, and auditing in the MOA, which will be subject of other ASIP preparation work. Improvements do not require much change in procedures, but significant support to acquire the necessary equipment, and to recruit and train qualified staff, as well as a firm policy aimed at enforcing existing procedures and at ensuring that qualified and trained staff is retained by the Ministry.

Issues in Recurrent and Capital Budgeting and Expenditure Approval

The annual recurrent budget preparation cycle is initiated well in advance of the end of the financial year, with the Budget Call Circular issued jointly by the MOF and the MOP which sets the guidelines and the key parameters that line Ministries must observe in preparing the submission. These are further communicated within the MOA to the Department Heads, and to the DAOs through the DFS. The Capital Budget is also prepared well in advance, and consists largely of a consolidation, not always complete, of the annual work plan and budgets of ongoing donor-financed projects.

At District level, the Recurrent Budget submission is prepared towards the end of the third quarter of the fiscal year, when the status of funds is reviewed by the DAO and requests for virements are also prepared (see para 89). District Budgets proposals are forwarded for approval to the DFS, Maseru.

The main issues identified with the recurrent budget preparation procedures include:

• the absence of clear Ministry guidelines to orient the preparation of budget proposals by Department Directors and DAOs;

• the insufficient flow of financial information to staff responsible for the operating units that would enable them to understand the relationship between the physical and the financial aspect of the Work Plan and Budget ;

• the absence of a performance monitoring system to orient Department Directors, DAOs and Section Heads in preparing the budget on the basis of past experience;

• the lack of integration of the annual work plans with the corresponding budget;

• inadequate capacity to cost services and other requirements, not only at District level, whenever more than marginal changes from the previous year estimates are involved;

• inadequate skill in presenting the justifications of budget proposals;

• the classification of the Budget needs to be up-dated, simplified, and made easier for operators to use properly.

The Financial regulations and procedures have been the subject of an exhaustive description and review by IFAD-appointed Consulting Accountants . The general view has been expressed in a WB report that some of the regulations "need to be updated to take into account the expanded functions of Government". Nevertheless the same document recognizes that the enforcement of existing regulations is much more important than a revision of existing regulations. It is expected that the MOF will issue manuals to provide firm guidance in this, and to better clarify the budgetary process, the financial control procedures and the respective obligations of operating staff, accounting staff and storekeepers in the process.

The disbursement of the Budget follows the following procedures:

• Salaries and allowances: are paid directly by the Treasury; checks entitled to District staff are collected in Maseru by the District Personnel Officer and are payable by the Bank at the District HQ.

• Other recurrent expenditure items: MOF issues quarterly warrants to release the amounts authorized by the Vote. The warrants are itemized in a way which is described in a later section. They represent authority for the MOA Certifying Officers (the PS, the DPS, the Financial Controllers and the DAOs) to issue payment vouchers within the ceiling of each item written in the warrant. In MOA, the Director, DFS receives part of the Ministry warrant, which is allocated to individual Districts by way of sub-warrants, copies of which are sent to the District Sub-Accountant, who is the Treasury Representative in the Districts and is authorized to sign the checks in settlement of the payments vouchers issued by the DAOs. A consolidated monthly statement of the Status of Funds is produced by the Financial Controller, consolidating the ledgers in which every item of expenditure, authorized under the Vote, and for which authority to spend has been warranted, is recorded. The Status of Funds statement shows the Vote, the warrants received, the cumulative commitments entered into under the warrants, and the balances as of the date. The Status of Fund shows the balance of commitment entered into by the MOA: checks issued in settlement of the payment vouchers and corresponding cash balances are recorded by Treasury. Reconciliation of the MOA and Treasury accounts is a problem: there is serious problem of aggregation. In the MOF the accounts are mechanized, while the MOA accounts are kept manually. In MOF, the software used (GOLFIS) needs a good deal of tuning up, and errors in inputting are combined with errors due to the computer. In MOA, manual input errors are certainly extant, but are not compounded by the errors of machines.

• Capital Budget (Donors’ projects): these follow ad hoc procedures, not always conducive to better performance than those adopted by GOL.

• Capital Budget (GOL funds): the main issue with this procedure is the Central Tender Board Threshold of M 3,000 (about US$ 700). The ceiling has been fixed a long time ago and never adjusted to reflect price inflation. The implications (overburden of CTB staff, delays, splitting and packaging of purchases to circumvent the rule, etc.) are obvious. Government would review the ceiling as a matter of priority: the new ceiling should be of the order of 10 to 20 times higher.

Flexibility in the GOL budget is allowed in two ways. The major way recognized in the Constitution involves a Supplementary Budget Vote. It is an extraordinary measure, not recently utilized, as a result of the rigorous control introduced by the SAP and followed ever since by the MOF. The other is the system of virement, whereby resources appropriated under an item of the Vote can be re-allocated to other items, if expenditure forecast is less than budgeted, and provided the amount subject to virement is forfeited. Virements are normally requested after a review of the Status of Fund to obtain amendments to the 4th quarterly warrant. Approval of the MOF is of course required, and, reportedly, quickly granted in the case of one District. Provided adequate control is exercised, the system introduces an important element of flexibility, which would be better exploited under ASIP, when financial reporting has improved and the performance monitoring system will be installed and functioning.

Issues in Financial Control and Reporting

GOL internal financial control procedures on the recurrent budget are adequate . The cases of non-compliance with established procedures experienced in the past are rapidly diminishing as a result of a serious drive from MOF to eliminate them. Inadequate control is a more serious problem with some Donor-financed projects operating independent accounting, reporting and auditing systems. A key impact point for bringing about further improvement is the inadequate staffing of the Financial Controller office and, in particular, of the District accounts offices.

On the other hand, the control of stores is generally inadequate. This is a particularly serious problem in the Districts, where some items are stored in dispersed, unsecured places, and entrusted to staff who have little training as storekeepers. Furthermore, little control is exercised over them at the moment (see later: revenue accounts). Control will be ensured by regular inspections of the Senior Accountant under ASIP.

Maintenance of public assets, particularly buildings, is poor at the District level. This is partly due to inadequate budgetary appropriation, partly to the difficulty of contracting for works the smallest size of which would require CTB approval, and partly to the absence of managerial staff entrusted on a full time basis with the responsibility for maintenance, which now falls on the shoulders of the DAO like every other function not having a specific technical connotation. The revision of the CTB ceiling, the appointment of the Executive Officer (logistics) proposed in Section B, and an adequate budget would go a long way toward improving the situation under ASIP.

At present the Status of Funds reports are prepared by the Departments accountants at MOA HQ, and the District Accountants. These reports are sent to the Financial Controller for consolidation. Control over the accuracy of the figures is inadequate, either by way of reconciliation with the ledger of the payments vouchers cleared by the Financial Controller, or by spot checking the submission of the Department/Districts accountants. This has serious drawbacks and causes friction with the MOF own estimates. Secondly it is an indication that the Status of Funds is not understood as a management tool, but simply as an obligation to fulfil in order to obtain the continuation of the release of funds from Treasury. Finally, it suggests lack of appreciation for the importance of accuracy in handling public funds.

On the Treasury side, the payment vouchers cleared by the MOA Financial Controller are recorded along with the details of the checks issued in settlement. The reconciliation of the Treasury payments accounts with the MOA Status of Funds and with the Vote Book is not done, despite it being a requirement of the MOF. One reason for this is the poor performance of the MOF computerized accounting system: as the printout is not to be trusted, attempting the reconciliation would only cause conflicts. Steps will be taken by both the MOF and the MOA to resolve the problem, by improving the performance of the MOF software on the one hand, and by introducing computerized accounting and adequate checks in the MOA, on the other. A consultant will be shortly appointed by the MOA, with WB assistance, to report of the computerization of the MOA accounts. Funds are included in the Mountain Districts capital budget for the purchase of the necessary equipment and software and the training of staff in their use on the expectation that a decision about the software to be used will be taken shortly by the MOA, jointly with the MOF.

As part of the overall ASIP preparation, an improved system of financial reporting will be studied and adopted under ASIP. Linkages with the performance monitoring system will be established, with due regard, however, to avoiding undue complications of the accounting procedures. Some suggestions as to the features of the improved system are included in Section B, the suggestion that the District warrant be split among Sections by the DAO and that the monthly status of funds of each Department be regularly communicated to the District Department Heads by the Senior Accountant and discussed by the DAO, the SA and the Department Head concerned).

Internal auditing within the GOL is weak, due to lack of staff in the Auditor General Office. Consideration would be given to the employment of external auditors under ASIP, until the situation improves.

The Budget Classification: Need for Adjustments

The Estimates of the Kingdom of Lesotho, published by the MOF after Parliamentary approval of the Budget, include the current Budget Classification (annex 3). The latest classification includes four sub-heads:

• Personal Emoluments

• Travel and Transport

• Operating Costs

• Special Expenditure

Personal emoluments are subdivided into 6 items and 12 sub-items. The inclusion of the 6th item, "short term local training", under this sub-head generate some confusion. There is room for rationalization of other items and sub-items as well, some reflecting historical changes rather than the logic of a classification.

Travel and Transport: this sub-head is subdivided into 12 items (1 abolished) and 21 sub-items. It includes the subsistence allowances paid to Staff travelling (local and international subsistence are separately accounted for).

Operating Costs: these include 17 items and a total of 51 sub-items. There is need for simplification and streamlining of several of these items and sub-items to ensure that that the classification is clearly understood at all levels, and that entries are correctly done by lower level budgeting and accounting staff. The lack of explanatory notes (a manual would be welcome) is noted. Some items or sub-items have the same or very similar names. Other items reflect the practice of "counterpart fund" contributions to Donor-financed projects, adding to the confusion already noted about the recurrent budget elements included in the Capital Budget. In particular the content of the items: upkeep and running costs need clarification. The apparent duplication of the items included under running cost and under purchases of production materials, goods and services needs to be eliminated (at present the latter sub-head is used in the Districts to purchase extension demonstration materials).

Special Expenditure: this includes 5 items and 17 sub-items, some of which are improperly classified as recurrent expenditure (purchase of vehicles, equipment, arms mixed with ammunitions, for example).

The "Departmental Warrant" form issued to the Districts by the DFS includes a selection of sub-items of the official classification. In the form they are listed in accordance with the codes of the budget sub-head number. In the following presentation the warrant items are re-grouped in a more functional way:

transport:

• vehicle repair and maintenance

• fuel and lubricants

• vehicle hire

• local fares

• horse hire

• power

• communications

• office overheads

• printing

• stationary

building maintenance:

• maintenance of public assets

• minor works

• upkeep

• running costs

• subsistence

• local training

Another item "counterpart contribution" is also currently included in the warrant, whenever appropriate. This item will disappear with ASIP, following the elimination of the concept of project financing.

From the point of view of MOA financial officers and accountants, the streamlining of the classification, and the distribution of a short manual explaining how to use it, would make the record keeping easier and reduce the danger mis-allocating entries and of accounting errors. This is of great importance in view of the limited number of qualified administrative staff available, an even more acute problem at District level that at MOA HQ.

From the point of view of donors, the question is to which extent the above classification needs to be clarified and/or modified so that it would become an acceptable basis for disbursement of donor finance in support of ASIP. This matter needs to be reviewed by specialists. Prima facie, however, it would appear that the major classification problems will arise, as already mentioned, with two items, upkeep and running costs. However a closer scrutiny will probably reveal that other adjustments/clarifications/additions are required.

In the Districts, upkeep includes the operating costs of the FTC, but other minor items are also included. Furthermore, as the FTCs do run some production units, the operating costs of such units are also budgeted under this item. With ASIP, such activities of the FTC should cease, and catering services would be privatized, so that upkeep would only consist of the legitimate operating expenses of the FTC, incurred in accordance with the policy, including the contracts with catering contractors.

Running costs refer to the operating expenses of the production units, mostly animal production units in the mountain Districts, but other items, such the operating cost of the TOU, costs involved in relief operation, and other minor items are also included. This item will be sub-divided in such a way as to distinguish the different activities which it provides resources to. Privatization of some of the animal production units, and of the TOU would significantly reduce the importance of running costs in the Mountain warrants.

Pending clarification about the two most critical items, and a more specific review, it seems that the changes that need to be made to the District Warrant itemization to make them consistent with the disbursement categories of most Donors would not be very significant and could be accommodated by the MOF without major difficulties.

The Decentralization of the Recurrent Budget at District Level

As of next Fiscal Year, 1997/98, the decentralization policy of the MOA will make an important step forward. The District Agricultural Offices will become primary warrant holders, at par with Ministry Departments. As a result of this decision, steps have been taken to strengthen the District Accounting capacity as discussed in Section B.

The details of implementation of this decision are still to be fully elaborated. They would involve the participation of the DAOs in budget hearings, a practice which ought to be better defined not only with respect to internal arrangements within the MOA, but, in particular, in terms of the formal meeting with the Ministry of Finance and the Ministry of Planning. With the new procedure, the District will receive a warrant directly, and the (yet to be appointed) Senior Accountants, will act as controllers of the District financial accounts. The Senior Accountant will also be responsible for the monthly Status of Funds statements in the Districts.

Two issues in expenditure management not discussed before must be resolved by the MOF to make the decentralization decision effective. One is the reported late arrival of the Vote Book to the sub-Accountant Offices. The other is the late payment by sub-Accountant of claims of contractors and suppliers raised with payment vouchers issued by DAOs. With the higher status of the DAOs, and similar measures being contemplated by the Ministry of Local Government for the District Administration in general, it is envisaged that such delays would be eliminated.

Under ASIP, the District Agricultural Offices would receive the approved budget appropriation in their own right. This would make it possible for the Management Team to re-discuss the Work Plan in the light of the total resources received, and to allocate resources to each District Section in accordance with the requirements of the annual targets set (or to be re-set) so that they will be consistent with the resource allocation.

Consideration will also be given under ASIP to the opportunity of changing the way the FTCs are financed. At present, the departmental vote includes an allocation for the FTC under upkeep, as well as allocations for Departments paying the FTC of their staff using the facilities (some is accounted for under local training some under local subsistence). The Resident officer receives cash in payment of service rendered to non-government clients (which is turned over to the sub-Accountant) and credit notes of Government staff (of MOA and of other Ministries). The net financial position of the FTCs is not clear, nor the extent to which its operations are actually subsidized. This prevents the rationalization of the operations and the setting of the rates at a level commensurate to costs or at least to a transparent subsidy policy. Under ASIP, consideration will be given to (1) establishing a revolving fund for the FTCs, (2) setting adequate rates to cover costs, (3) adjusting the subsistence allowance budgeted by Government Agencies accordingly, and (4) taking a transparent decision as to the extent to which FTC services to private individuals should be subsided, if at all.

District Sub-Warrants for the District Sections

The decentralization of the budget at District level would permit the introduction of an innovation which would pave the way to easier Donor support for the Mountain ASIP.

The DAOs, having received their Quarterly Warrants for the whole District, would be in a position to issue sub-warrants to their Section Heads, specifying the resources they are allocated to perform their tasks. The sub-warrant would include all the items required by the District Section to perform its tasks, in accordance with the budget. However, DAOs would remain the sole MOA Certifying Officers authorized to sign payments vouchers in the Districts.

The District Senior Accountant and his staff would then keep separate ledgers for each Section, issue separate Status of Funds reports and consolidate the financial reporting for the District Agricultural office as a whole. Though this implies that about 10 books will be kept instead of one, it would not be too complex to manage. In fact, it may well ease bookkeeping rather than make it more difficult. The consolidation, however, may well increase the number of aggregation errors. Computerization, by a user-friendly software, would greatly simplify matters.

The introduction of the practice under ASIP would have considerable advantages:

For the District internal procedures:

• resource allocation would be known to all Section Heads;

• Section Heads would be able to understand the relationship between the Work Plan and the Budget allocation they have received;

• Section Heads would receive essential information about their use of financial resources against their respective allocation;

• physical performance under the Work Plan could be evaluated against financial resource use;

• a substantive rational basis would be provided to justify the request of virements.

For Donors’ contribution to ASIP financial requirements:

• Donors could identify the ASIP activity (Section) they would be willing to support;

• Donors could identify the warrant items they would be willing to finance for each Section the activities of which they would be willing to support.

The rationalization of the Budget classification would possibly make the suggestion acceptable to Donors. The MOA Financial Controller has confirmed that re-arrangements of the classification, and also splitting of the accounts for each warrant item to record separately GOL and Donor’ finance of each budget item can be negotiated with the MOF.

Donors financing of items included in the regular MOA budget and accounted for by MOA Departmental accountants has been successfully tested by the EU with the STABEX programme implemented by the Department of Livestock Services.

Summary of the Essential Measures to be Taken

Under ASIP, a general reform of the MOA is expected to take place and to go hand in hand with important changes expected to take place as well in the rules, regulations and procedures of the MOF, the MOP, the Ministry of Public Function and the Establishment Commission. All that would take some time to materialize.

The minimum essential measures that need to be taken for external Donors to be able to finance the Mountain ASIP as a Government Budget Programme rather than as a separate project include:

1. the Central Tender Board ceiling must be revised from the present level of M3,000 to a more reasonable level at least 10 times higher; relatively small building maintenance contracts (M10,000 to M20,000) must be allowed without clearance by the CTB, for example;

2. a decision must taken regarding the software to be used for the computerization of the budget proposals and of the financial accounts of the MOA (Districts included);

3. the Budget classification must be reviewed, the necessary modification introduced and detailed instructions/explanations issued;

4. the financial reporting procedures must be computerized, staff trained to that effect, the established internal controls on the Status of Funds enforced, and the circulation established of financial information to Department and Division Directors at Centre level, and to Section heads at Districts level;

5. the practice of sub-warranting the Sections of the District Agricultural Offices must be introduced, and the related expenditure reporting procedures by the District Senior Accountant established;

6. the utilization of external auditors until such a time as when the Internal Audit Unit of the MOF is fully staffed must be authorized.

Scope for a Unified Approach to Donors Contributions to the Mountain ASIP

Under ASIP, the Recurrent and Capital Budgets of the MOA would include only activities consistent with the ASIP objectives strategies and policies. Donors would discuss these activities in the context of Donor Meetings, and would select which activities they would like to support from the menu offered. It is expected that some activities would receive substantial support, while other activities may receive little or no donor support. GOL will complement Donor finance with its own resources in order to undertake all ASIP activities.

With regard to the ASIP for the Mountain Areas, all relevant activities to be undertaken by the Department of Livestock Services (in the Field Services, Veterinary, and Range Management Divisions), and all activities to be undertaken by all Sections of the three Mountain Districts Administration have been specifically identified.

On the recurrent budget side, the annual cost estimated for each activity retained under ASIP is identified by way of items corresponding as much as possible to the budget classification, and the cost of the activity correspond to the allocation required by the MOA unit responsible for the implementation of the activity. Salaries and allowances as well as all warrant items have been specifically identified.

Donors will be able to select which budget item of which activity (Department, Division, District Section) they wish to finance to assist the GOL in implementing the Mountain ASIP.

Donors’ contribution to the Capital Budget would be much easier to channel, since the itemization of the capital budget is not subject to classification in the same way. Donors would select which item of the Budget they would contribute to and to which extent.

Details of the financial procedures would have to be negotiated with each Donor, within the framework of the MOA improved budgeting and financial control practice implemented under the ASIP. The experience of the EU STABEX programme offers an interesting precedent to be carefully evaluated by Donors at the time of their ASIP appraisal.

D. Recurrent and Capital Budget: Estimate of the Model ASIP Budgets of the Mountain Districts and of the Range Management Division of the DLS

Approach Used in Estimating the Recurrent Budget under ASIP

The recurrent budget has been estimated at full ASIP development, when all activities envisaged are expected to be carried out, all the required staff is in position, and surplus posts have been eliminated. The estimate includes all items of expenditure which will be retained by MOA on a continuing basis, and therefore represent a claim on Government resources beyond the time horizon of the ASIP. This presentation is required for the Government to appreciate the commitment they will enter into in undertaking the ASIP, and is consistent with the basic concept of ASIP formulation and financing.

The budget items headings are those of the Budget classification and their content is explained in the following section.

A similar method has been used for the Range Management Division of the Department of Livestock Services, Maseru.

The estimated recurrent budget at full ASIP development has been compared with the 1996/97 budgets of the three Districts and of the RMD (table series BD 1 to 4) to obtain an insight as to the size of the incremental budget required to sustain the ASIP activities in the long run.

The phasing of the incremental recurrent costs of the 5 years of the ASIP is not possible at this stage, nor is actually required. Despite the great number of details used in the estimate of the annual recurrent expenditure under ASIP and beyond, these are more interesting as an indication of the relative cost of the single activities to be undertaken, than as a precise measure of what the cost of the budget main components would be. This will evolve in the course of time, depending of the circumstances of each section of the administration. The estimates have a methodological value in as much as they relate costs to functions, objectives and activities: they must be interpreted as indicative, but adequate to provide a rational basis for projecting the overall financial requirements of the ASIP for the mountain areas, and to ensure, that costs are determined, to the extent possible, on the basis of an efficient use of all staff and resources.

Many decisions not yet taken, and the degree of success in implementing those decisions, will influence the pace at which the annual budgets will eventually approximate the expected budget projected for full ASIP development. Success in staff recruitment is one critical factor, since the ASIP model involves more changes in terms of a different structure of personnel, with emphasis of better qualified people, than in terms of absolute numbers, which actually are expected to increase only marginally in terms of the number of personnel actually in post, and to decline in terms of established posts in each District.

The implementation of privatization schemes is another cause for uncertainty. Annual phasing of recurrent ASIP costs will be a difficult task anyway, involving very large margins of errors, particularly in the initial years, and must be left to the preparation of the annual Work Plans and Budgets.

In view of these uncertainties, a rough and ready way of estimating the total cost of a 5 year ASIP for the Mountain Areas has been applied: the incremental annual recurrent budgets estimated for the full development phase have been multiplied by the factor 4. This procedure, and the estimate of the total Capital Budget over the 5 years period, provides a basis for estimating the envelop of incremental finance required to support the ASIP: assuming this would be fully met by Donors, it would also provide the basis for estimating Donors commitment to finance the ASIP. It is suggested that ASIP appraisal should not pretend to go much further, and that Donors should switch the detailed appraisal of their contributions to the stage of approval of the annual Work Plan and Budgets. The GOL would accept that Donors approve the Budget as a condition of release of Funds from their Special Accounts, pari passu with Treasury’s release of resources originating in the revenue account.

Budget Classification

The annual recurrent Budget of the ASIP model is presented in accordance with the Budget classification, and includes the items (called sub-heads in the official classification) making up the quarterly warrants issued to the Districts in 1996/97. It has been estimated on the following basis:

Salaries and allowances:

• the Staff to be retained by each District has been estimated on the basis of the adjustments to the establishment list required to fit with the staffing pattern discussed in Section II;

• the salary bill for the retained staff has been calculated on the basis of the scales published in the Establishment List, 1996/97;

• monthly allowances have been estimated at a fixed rate of M250 per month for all staff posted in the Mountain Districts;

• other allowances are entered as lump sum in the cost of the District Administration.

Warranted items:

• vehicles repair and maintenance: the vehicles fleet has been estimated on the basis of the annual mileage required by each operating unit and of the need to reduce to the minimum the need for sharing vehicles among operating units having to travel at the same time during the same working season; the annual cost has been estimated at 10% of the replacement value of the vehicles;

• fuel and lubricants: the cost is based on an average rate of M0.42 per kilometre. Mileage has been estimated for each section of the DAOs and for each operating unit;

• general office expenses ( warrant items: power, communications, office overheads, printing and stationary): the estimate is based on the 1996/97 budget amounts, adjusted by a coefficient which reflect discussions with the DAOs and the Livestock Department in Maseru; however, the following costs have also been included under the sub-head printing include the maintenance cost of office equipment to be purchased under the capital budget;

• building maintenance: the estimate is based on the 1996/97 budget amounts, adjusted by a coefficient which reflect discussions with the DAOs;

• upkeep: includes the allocation to the FTCs. The budget for 1996/97 have been retained. Notice that: (i) the production units will cease to operate, which would bring about some savings, (ii) catering services will be privatized, which is also expected to result in lower costs, (iii) on the other hand, a higher utilization of the facilities is expected under ASIP. Double counting results from the allocation of a budget to the FTC per se, while at the same time payment of FTC rates are included in the local subsistence item wherever appropriate. This was done in the absence of a detailed analysis of the costs actually incurred currently by the FTCs. The double counting must be eliminated after a decision is taken about establishing a revolving fund for the FTC, and adjusting the FTC rate policy;

• running costs: these include (i) items related to the animal production units, which need close verification and are subject to a considerable margin of error, partly depending on decisions about privatization policy and the speed of its implementation; (ii) items related to costs of the operating units, inclusive of renewal of camping equipment and demonstration materials ;

• subsistence: the item includes local and international travel: subsistence costs are based on the Government rates approved under the 1966/67 budget; applied to the number of days of travel estimated for each activity retained by the typical District Budget model, and envisaged under the RMD programme.

• local training: costs have been re-estimated for each section and operating unit: some entries under this items might have been wrongly inputted (subsistence?) due to uncertainty about the definition of the content of the item;

• the sub-head demonstration material has been included under running costs.

Summary of the Annual Recurrent Budget under ASIP: the Incremental Recurrent Budget of the ASIP for the Mountain Areas

The ASIP for the Mountain Areas would increase the MOA recurrent budget by a total of M 3.1 million a year on a permanent basis, 2.2 million on account of the three Mountain Districts, 0.8 million on account of the RMD, and 0.16 million on account of the ARD. Over the 5 years of the ASIP period, the estimated total incremental recurrent budget would be about M 12.5 million.

The total public expenditure to be incurred by the GOL to implement the 5-year ASIP in the mountain areas would be about M 83.8 million. This figure is obtained by adding the total recurrent budget estimate to the total capital budget estimate, and includes all activities presently undertaken which will be retained under ASIP.

In terms of incremental costs, the recurrent budget would require additional appropriations totalling about M 12.5 million over the 5-year period, and the capital budget additional appropriations totalling M 33.5 million, since M 4.3 million are already committed under the STABEX projects. As a result, the incremental finance required is equivalent to about US$ 10.8 million, of which US$ 2.8 million would finance the incremental recurrent budget, and US$ 8 million would finance the capital budget.

Reconciliation with a Project Component Approach and Identification of Budget Items Eligible for Donors’ Financing

Lending to support ASIP would be quite different of lending for conventional structural adjustment or general sector support. Donors will insist that they ought to be able to recognize what their funds will pay for, on both recurrent and capital account, and will not be satisfied with providing domestic resources to finance the Government budget through balance of payment support, such support made conditional to introducing policy reforms. Under ASIP, Donors would wish to recognize the familiar shape of the projects they are used to finance: the danger actually exists that Donors may wish to projectise ASIPs, which would go contrary to a major objective of ASIP: namely; rationalizing the budgetary and expenditure process.

In order to present the concept of the ASIP for the Mountain Areas along the familiar lines of project components, ASIP activities were classified into four components: (i) Institutional Development, (ii) Crop Development and Diversification, (iii) Livestock Improvement, and (iv) Natural Resource Conservation. Recurrent and capital budget items were re-arranged in a different matrix (see following table), in which:

the columns show:

• the Administration sub-divisions to which warrants or sub-warrants are proposed to be issued, such sub-divisions representing the level of accounting that would be kept by the Government;
• the aggregation of the sub-divisions the activities of which make up one component;

and the rows show:

• the budget classification heads and sub-heads.

The major issue of the above presentation is that the level of aggregation may not correspond to the level of disaggregation desired by some Donor. This would not be problem with the capital budget, which is not subject to the same accounting constraints, but is a really difficult one with the recurrent side of the budget. The Livestock sub-sector offers a typical example of the difficulty: from a structure and management point of view, in the District Administration, Range Management ought to become an Operating Unit of the Livestock Section. From a conceptual component point of view, Range Management is part of the Natural Resource Conservation component. District accounting at a level lower than the section is impractical. As a result, the column headed "Livestock Improvement" is incomplete, and accounting (as distinct from budgeting) for District Range Management will not be possible.

Methodology

The public expenditure analysis elaborated for the programme was meant to provide answers to the following questions:

1. is the total projected expenditure within the ceiling that would be allowed by the monetary authority? if an increase is forecast on recurrent account, can possible savings of present recurrent expenditure in the sector be identified to offset part of the increase?

2. what is the projected expenditure going to pay for? are the activities to be financed in line with Government objectives, policies and strategies?

3. are the projected expenditure items adequate for the administration to implement their work plan? and consistent with least cost conditions?

4. is the structure of expenditure expected to change in such a way as to indicate a move towards more efficient use of resources?

5. whenever economic benefits to individuals, or groups of individuals, are expected as a result of incurring public expenditure, are such benefits commensurate to the cost taxpayers have to bear to finance that expenditure?

6. is expenditure in activities expected to promote economic benefits justified from an economic point of view?

To answer the above questions several criteria have been applied.

ceiling on available resources: since the ASIP will increase MOA recurrent budget in the Mountain Areas, mostly on account of activities in Livestock and Range Management, one source of possible savings to offset this has been identified through privatization of activities currently undertaken by the Special Services Division of the DLS;

compliance with policy objectives: this analysis has been undertaken in the context of the selection of the activities to be retained under ASIP, for each one of the current and new activities of the Departments, Divisions, District Sections, and operating units examined (Section B). The expected impact on the food security and poverty alleviation objective has been estimated in Working Paper 12 of the FAO Formulation Report;

least-cost solution: the recurrent and capital budget estimates have been worked out with a view to achieving least cost. Proposals contrary to the principle have not been included in the budget;

relationship with planned activities and expected productivity of Government staff: staff requirements under the District recurrent budget model have been estimated (Section B) with a view to ensuring adequate capacity to achieve the targets set in the technical Working Papers; redundant staff has been eliminated in the model wherever necessary; operating resources have been budgeted to ensure that staff has sufficient means to implement their tasks. The changes in the structure of the recurrent budget resulting from ASIP have been analyzed to obtain an indication as to whether the ASIP budget resource allocation would be conducive to improving performance of Government Staff. The conclusions of such analysis rests on the assumptions made about expected cultural and behavioural changes which are discussed in Section B, particularly with regard to the observed impact of precision in defining tasks, and of availability of resources for task implementation, on staff attitude toward work and performance.

Effectiveness of the income enhancing activities and balance of allocations between sub-sectors: this involves a judgement as to the relationship between the cost incurred by Government to undertake these activities and the outcome expected of them.

In appraising Government expenditure in activities aimed at bringing about specific benefits to specific groups of people, effectiveness indicators have been estimated to show what benefits can be expected by the tax payers in exchange for the resources withdrawn from them to pay for activities which bring about those benefits. From the point of view of the clientele of Government services (the taxpayers), the benefits are measured by the expected increase in the net income of those farmers who participate in Government sponsored programmes. By comparing the present value of the stream of incremental net farmer income to the present value of the Government expenditure incurred in the related activity, a measure of the effectiveness of the activity is obtained. The balances would show the taxpayers’ gain or loss. Naturally, benefits may occur to taxpayers who are not the same people as those who bear the costs, resulting in a transfer of resources, so that the progressiveness of the public expenditure programme can also be evaluated. The ratio of the two present values indicates to which extent the incremental resources generated by the tax-payers as a result a Government activity exceed the resources withdrawn from tax-payers by Government for that purpose. The ratios (effectiveness indicators) have been estimated for different activities to provide an insight as to how effectively public money is used in each activity for the benefit of the public, and to evaluate the balance of the ASIP structure by activity from the point of view of the clients of Government services. Ratios greater than one would suggest a public gain, activities showing effectiveness indicators less than one would need solid justifications on other grounds. Obviously, such indicators are not measures of economic returns on Government expenditure, although in some cases they can be used as proxies for them. These indicators are subject to the validity of the assumptions that benefits are entirely due to the related activity, an assumption generally made by all project analysts, and which is seldom entirely justified.

The following procedure has been adopted:

to estimate public benefits:

• the annual stream of the incremental net returns to labour expected from adoption of Machobane system, seed multiplication and related uptake, horticulture development, and livestock improvement have been estimated over a period of 15 years; the present value of the streams has been calculated using a rate of interest of 10%;

to estimate public costs:

• the net present value of the annual capital budget expenditure of the District Crops Operating Units (horticulture and seed multiplication) and of the stream of the respective recurrent expenditure budget and over a period of 10 years has been estimated; to each flow 1/2 of the present value of the overhead expenditure of the Crops Sections and 1/3 of present value of the recurrent and capital expenditure of the Extension Section has been added;

• the net present value of the capital budget expenditure for the extension contracts-Machobane system has been estimated; to this 1/3 of present value of the recurrent and capital expenditure of the Extension Section has been added;

• the net present value of the recurrent budget and of expenditure under the capital budget of the Department of Livestock Services imputable to the production of the estimated benefits, and of the Range Management Division have been separately estimated, combining expenditure of the Headquarters and of the Districts, and the respective present values calculated for LS and RMD separately;

to calculate the indicators:

• the balances of the net present values for horticulture, seed multiplication, extension contract-Machobane system, for the combined crops development activities, livestock improvement services, and RMD have been calculated and the ratios of tax-payers benefits to incremental costs to tax-payers worked out to obtain the effectiveness indicators.

Increased Allocation and Structural Analysis of the ASIP Model Recurrent Budget

The implementation of the ASIP would increase the recurrent budget as follows:

• Thaba T’seka District up 26% (personnel cost, 5%; warrant items, 20%)

• Mokhotlong District up 42% (personnel cost, 5%; warrant items, 37%)

• Qacha’s Nek District up 24% (personnel cost, 5%; warrant items, 18%)

• RM. Division up 75% (personnel cost, 16%; warrant items, 59%)

The expected increase in the Mountain Districts and RMD budgets will be more than offset by the elimination of staff currently employed in the Animal Production Units in the lowland, which will be privatized, and of the corresponding running costs.

The following table summarizes the structure of the recurrent budget, by major sub totals of sub-items of expenditure aggregated on a functional basis, and compares the expected structure under ASIP to the structure of the current budgets.

The recurrent budget structure of the RMD under ASIP also shows larger shares for maintenance of assets, local subsistence and staff training. Transport costs on recurrent account are expected to relatively decline despite higher mileage, since current costs on vehicle repairs are very high, due to the age of the vehicles. Vehicle replacement under ASIP would reduce those costs considerably. The increased share of the total cost of the Range Management Education Centre (RMEC) is due to bringing the allocation in line with that of other FTCs, and subject to the same reservations.

On balance, the structure of the projected recurrent budgets under ASIP, compared with the current structure, indicates, in the case of both District and RMD, a better allocation of resources towards items which would enhance the productivity of the staff and its impact on production. The importance of the crop development effort during the ASIP period is underestimated by the above figures, since they exclude the extension contracts with the NGOs related to farmer training in the Machobane system, which is budgeted on capital account.

Structural Analysis of the ASIP Capital Budget

The structure of the Districts and of the DLS & RMD Mountain ASIP Capital Budget is shown in the following table. In the Districts, the largest share (34%) of the Capital budget is absorbed by programme 4, reflecting the cost of building the Area Resource Centres, followed by programme 1 and 2 for similar reasons. Civil works (mostly dip tanks) and buildings will absorb 56% of the estimated budget, which reflects the lack of basic facilities in the Mountain Districts. The second largest (12%) expenditure item under the District capital budget is vehicles: this also reflects the inadequate facilities available at the moment, which seriously handicap the field work performance of the staff. The extension contracts with NGOs (Machobane system), very important for the food security objective of the ASIP target group, would account for 9% of the total Districts Capital Budget .

With respect to DLS, 37% of the total capital budget estimated for the Mountain ASIP is accounted for by Animal Production and Field Service Division, represented by the two EU STABEX projects, LIC rehabilitation and support to the NW&MGA, for which funds have already been committed; and 7% is represented by the replenishment of the national drug and vaccines revolving fund (Veterinary Services Division), which needs to be updated on account of a combination of storage losses, revenue losses, and to product price inflation not recovered by rate increases. A detailed analysis of the financial position of the Revolving Fund, and a study of the scope for increasing charges, and eventually to provide the rationale for possible subsidies will be undertaken by the DLS under ASIP.

Expected Effectiveness of the MOA Units undertaking ASIP Activities Aimed at Increasing Farmers’ Income

Expected impact on net farmers income. As a result of ASIP, at full development farmers’ net returns to labour in the mountain areas is expected to increase by a total of about M 11.2 million per annum, at 1996 prices . The livestock sector would account for M 2.5 million of the incremental farmer income. This estimate is based on an increase of the value of livestock production equivalent to 0.5 kg of wool per animal obtained from 50% of the sheep and goat herd of the mountain districts in 1996, largely to be obtained from a combination of more and better quality wool and mohair as a result of the enhanced effort towards sheep dosing and introduction of fodder cultivation and winter feeding. No specific benefits of the RMD activities, could be estimated for lack of basic information. The crops sector would account for a total of M 8.7 million per annum, of which about M 2.5 million as a result of the introduction of Machobane system, about M 2.7 million as a result of seed multiplication and uptake, and about M 3.4 million as a result of the establishment of intercropped orchards (fruit trees and vegetables).

Government expenditure (total recurrent and capital budget) has been provisionally phased to work out the effectiveness indicators. In the following table, the present values of the streams of incremental farmer income are compared with the present value of the expenditure incurred by MOA Departments, Divisions, District Sections and Operating Units, whose activities are expected to bring about such increase in income. The incremental farmer net income represents the benefits expected by taxpayers as a result of Government spending; Government expenditure represents the cost of such benefits to taxpayers. The balances show the public gain or loss. Since most of the activities under consideration are aimed at the poor sections of the population, and the tax burden on the rural people is less than the average, there will be a transfer of resources generally in favour of the less privileged people, in line with the policy objective of poverty alleviation.

Two important conclusions are emphasized:

• The benefits estimated for the crops sector are large compared with the resources Government will spend to encourage the sector, even when the impact of the slow development of fruit tree production is taken into account. At full development, negative benefits would occur only if net farmer income growth has been overestimated by an overall factor of about 1.6, considering seed multiplication and uptake, horticulture and Machobane system combined. Considering the prudent approach taken in estimating net farmers income, which includes the expectation of a substantial drop in the price of fruits as a result of the increased production, this is not very likely. Inter annual fluctuations of crop yields may seriously affect the results, but the negative impact would be partly offset by the introduction of the Machobane system, and by the investment on micro-irrigation horticulture. The main risk is the possibility of gluts in the potato market, which will occur, but will hopefully be a temporary phenomenon. Given the overall market situation in Lesotho (demand is sizeable and largely met by imports) private sector operators are expected to seize the opportunity and supply their clients from within the country as soon as they realize that the mountain areas have a surplus to sell of adequate size for commercial collection. Similar considerations apply to horticulture, with the added risk involved in the difficulty of controlling product quality in horticulture, pre and post harvest.
  

• Livestock services activities (aimed at improved animal health, better animal nutrition, and breed improvement) are an effective way of public spending, but negative net benefits would occur if benefits are overestimated by a factor of only 10%. On the other hand, the above analysis suggests that the RMD does not appear to have an obvious justification in terms of foreseeable taxpayer benefits. This is the result of not having quantified the benefits to be expected from the RMD programme. There is no quantitative basis, however approximate, to attempt such quantification. The inclusion of the RMD programme in the ASIP has other justifications.

In fact while the balance between the resources allocated to crops and extension and to livestock services (other than the RMD) seems appropriate, and the allocation to extension contracts seems low considering the level of the effectiveness indicator, the question could indeed be asked: why spend money on the RMD programme, if the activity turns the effectiveness indicator of Livestock Services to (considerably) less than 1?

There are several answers to this question: (i) ASIP benefits in the livestock sector are underestimated because the long term impact of the activities of Range Management on pasture productivity and on the environment have not been quantified; (ii) wool and mohair are Lesotho’s main source of agricultural export earnings: ensuring the long term sustainability of the sub-sector is important to adequately comply with Government objectives, even if the effectiveness of the activity is difficult to quantify at the moment; and (iii) the scope for substantial further improvement of the mountain livestock, as a result of success in range improvement and better animal husbandry practices, is considerable; it cannot be seriously predicted at the moment, but the ASIP RMD programme would establish the necessary foundation for a significant leap forward to become possible in the future.

Finally, it is important to recall the expected impact of the ASIP on the objectives of poverty alleviation and improved food security, which is the main justification for undertaking the programme. The expected increase in production from the Machobane model farm and from the intercropped orchard model farm would represent, respectively, 2.3 and 1.7 times the average expenditure of poor households recorded in 1993, re-valued at 1996 prices. The adoption of these models by poor farmers would bring about a real improvement in their living conditions, and thus achieve Government objectives with respect to the ASIP target group in the mountain areas.

Economic Evaluation of Crop Development and Diversification and of Livestock Improvement Activities

At the present stage of formulation of the ASIP a conventional economic analysis of the programme would involve making a very great number of assumptions, and therefore an ever larger gap with reality the more the assumptions built into the model. Furthermore. inter-annual production fluctuations, characteristic of Lesotho agriculture, make the application of conventional analysis tools rather unsuited to the country. These uncertainties, combined with the uncertainty about farmers’ response which is implicit in the demand-driven ASIP approach, further complicate the estimation of economic benefits. The annual phasing of Government expenditure is still to be discussed upon and worked out, the specific causes of uncertainty about the phasing of total incremental recurrent expenditure, and the difficulty of estimating the incremental recurrent expenditure by units of the administration, having been already emphasized.

Nevertheless, some Donors may still be interested in the application of conventional economic analysis yardsticks to the productive components of the ASIP they support. For this purpose, the programme re-aggregation into four major components has been used, namely:

• institutional development

• crop development and diversification

• livestock improvement, and

• natural resource conservation

and costs have been re-arranged, making several rather arbitrary assumptions with respect to the part of the recurrent costs of the components which should be considered incremental. Phasing has followed a similar pattern as that used for the effectiveness indicators.

The economic evaluation of the productive components has been carried out using the market prices prevailing in the Mountain Areas at the time of the Socio-Economic and Farming System Study, and largely confirmed at the time of the FAO ASIP formulation mission which took place after the partial market liberalization of 1996. Benefits are derived from the estimated flow of net returns to farmers, used to calculate the effectiveness indicators, adjusted for the opportunity cost of labour. The opportunity cost of labour used also reflects the market value paid in the area for unskilled farm labour. No adjustment has been made for salaries of Government employees.

The estimated streams of incremental cost and benefits have been discounted, using an interest rate of 10% to calculate Net Present Values, and to calculate the internal rates of return. The combined Crops Development and Diversification Component returns an IRR of 18%, while the Livestock Improvement Component an IRR of 10%. The IRR of the two production components combined is 16%.