In an earlier article, “Public Administration for What?” (Finance and Development, September 1967), Albert Waterston argued in favor of limited administrative reforms in developing countries where—as frequently happens—attempts at global reform would be unrealistic and ineffective. Here he applies his arguments more specifically to development planning.
EXPERIENCE shows that a great deal can be accomplished when partial improvements in administrative procedures and organization are purposively fitted to requirements of planned development. The considerable potentialities of this approach became clearer to me during a series of visits which, in my capacity as Advisor for Planning Organization in the World Bank, I recently made to 16 countries in Africa, Asia, and Latin America on behalf of the Bank. Almost all the countries visited have been or are committed to planned development through the use of over-all development plans with periods ranging from four to ten years. With few exceptions, the countries concerned have failed to achieve the targets in their development plans. Indeed, in several, development plans have been abandoned before their time was up because they were deemed to be beyond hope of fulfillment.
There was nothing new in finding such disappointed hopes; studies completed in the World Bank have revealed that many, if not most, low-income countries in early stages of development are unable to plan their over-all development with reasonable accuracy for periods of much more than a year. This does not necessarily imply that medium-term plans are useless or that they should be discarded by low income countries. Rather, the World Bank studies provided evidence that growth and development are not so much dependent on elegantly integrated, over-all, medium-term plans as they are on the ability of governments to execute soundly conceived and prepared projects in the public sector, control the allocations and expenditures of financial resources for these projects with reasonable efficiency through their budgets, and allow reasonable economic incentives to operate so as to stimulate investment in the private sector.
This article and its predecessor, “Public Administration for What?” are adapted from a paper prepared by the author for the Meeting of Experts on the United Nations Programme in Public Administration, New York, January 1967.
Fitting well-conceived, carefully thought-out, properly controlled projects of public investment into a medium-term national plan remains the ideal, even if for many developing countries it appears to be difficult to attain at present; very often, these countries cannot manage even the individual projects satisfactorily. Because of this, there is almost always a wide gap between what is called for, on the one hand, in medium-term, over-all plans prepared by a central planning agency and, on the other hand, what goes on in technical ministries, departments, and agencies, and in the budget office.
A Pragmatic Planning Approach
To bridge the gap which frequently exists between medium-term, aggregative plans, on the one hand, and projects, budgeting, and economic policy, on the other hand, I advised governments in the countries visited to adopt a pragmatic approach to planning which comprised two main elements:
Annual, over-all planning tied to budgets, which focused the attention of decision makers on what needs to be done immediately for development of the public and private sectors; and
Multiannual programing of strategic sectors of the economy, which focused attention on basic long-term policies and the systematic accumulation of well-prepared projects for each sector.
Such an approach required, first, a reorientation of planning procedures and organization of central planning agencies; second, the institution of a systematic and close relationship between plans and budgets; and third, the establishment of machinery for sector programing and project preparation in technical ministries, departments, and agencies.
Since such a planning system represented a considerable departure from the conventional one, the ideal approach would have been to overhaul all major administrative procedures and the organizational structure to conform to the new system. As I indicated in my last article, this approach was very tempting; it seemed so logical, and appeared to offer great rewards. It seemed to me, however, largely illusory. Given the circumstances in the countries visited, the chances of achieving major administrative changes appeared to be small, despite the willingness of most governments concerned to adopt the new system.
A more modest approach was therefore adopted, which was more in line with the realities of political power within the governments concerned, as well as within what I believed to be the limits of administrative capacity. In fact, taking a leaf from medicine, where conservative practice dictates the administration of only the minimum dose of a drug required to produce a cure, I eschewed the temptation to propose more than minimal changes in administration required to ensure reasonably effective operation of the new planning system. Agreement among those who would be most involved in carrying out the reforms was considered of much greater importance than maintenance of the purity of a principle, except for the very few principles with which compromise was impossible without jeopardizing the entire planning system.
Reorganizing the Central Planning Agency
The setting up of an appropriate organization within central planning agencies to prepare the annual plans, which was the first of the two important elements in the new planning system, proved to be a relatively simple task. Without exception, each central planning agency was short of well-trained and experienced planning technicians. The shortage was often exacerbated by the way in which the technicians were distributed in the various units. Generally, organization charts followed a familiar arrangement of units along functional and sectoral lines. Some units were well-staffed or even overstaffed; most were not. In some units, only low-level jobs had been filled and existing personnel were inadequately supervised; in others, there was a director with no staff at all, and because of the lack of assistants, he had to do clerical as well as analytical work. Thus, over-staffing and understaffing of units existed side by side although, taken as a whole, central planning agencies almost always had fewer planners than they needed.
To make the most effective use of the available staff for annual planning, a two-pronged approach was adopted. First, wherever possible, central planning agencies were advised to divest themselves of nonplanning tasks and to farm out planning tasks, especially long-term studies, which could be done elsewhere on contract or by any other arrangement. For example, in one African country, it was found possible to have the economics department of the university conduct some studies of manpower, economic criteria, and other matters for the central planning agency. In other countries, possibilities existed for farming out work on national income to the central bank or to research organizations.
Second, central planning agencies were advised to employ a technique for deploying their manpower which has been used with excellent results in other countries that were short of planning technicians. In these countries, task forces were set up, each with a work program arranged according to priorities determined after consultation with officials and political leaders. While each country has its own requirements, it was generally found useful to recommend the establishment of three task forces, one concerned with human, natural, and financial resources; another with problems of plan coordination; and a third with plan execution.
It was recommended that available staff be assigned to each task force in accordance with functions and work load. From time to time, as required, the work loads would be reviewed and revised and staff reassigned. By shifting planners from one task force to another, planning technicians would receive training in a range of subjects which would add to their knowledge and make them valuable in more than one planning specialty. Each task force would be headed by the best person in the central planning agency available for the job, and each such head would have a foreign advisor. The heads of the task forces would constitute the membership of an internal coordinating committee having as chairman the head of the central planning agency. The task force method was suggested not only because it mobilizes available personnel for more concentrated effort on vital tasks than is possible when the staff is dispersed in small groups over a large number of organizational units, but also because it is flexible and capable of responding to the changing environment characteristic of developing countries.
The Relation Between Plans and Budgets
Unless the budget reflects the items in an annual plan, the public sector portion of that plan will probably not be implemented. The close relationship between plan and budget, which is essential for carrying out plans effectively, requires a system of budgetary classification and budgetary procedures which permits easy translation of the plan into budgetary terms, as well as expenditure controls which allow timely preparation of quarterly, semiannual, and annual reports of plan progress.
But in many less developed countries the classification systems actually adopted in budgets make it difficult to relate projects and programs in annual plans to budgetary allocations. Good planning also requires that recurrent expenditures, e.g., for training personnel, be related to capital outlays so that all parts of a development project may be made to move forward at appropriate speeds. In addition, it is useful in planning to be able to extract from the recurrent side of a budget development items which, when added to capital items for development, constitute a comprehensive statement of budgetary outlays for development. But in many countries it is impossible to do these things.
Another impediment to effective planning is to be found in the way in which development and recurrent budgetary estimates are prepared in many countries. The development estimates are generally prepared in the central planning agency, while the recurrent estimates are prepared in the ministry of finance. There is often an inadequate system of communication between these two bodies while each is preparing its estimates, with the result that neither may be aware of the details of the estimates that the other is preparing until after they have been completed. In one Asian country, recurrent estimates are prepared before capital estimates. These procedures not only do not allow proper phasing of capital estimates and recurrent estimates for development, they also make it difficult to measure the impact of current investment outlays on future recurrent expenditures. This is a serious shortcoming, since capital expenditures almost always increase future recurrent expenditures. Unless this is borne in mind, the rise in the recurrent budget can greatly reduce the surplus on current account available for development. A final shortcoming of the system employed in many countries is that operating ministries are required to make two budgetary presentations, one for capital estimates to the central planning agency, the other for recurrent estimates to the ministry of finance.
Although serious, the problems outlined can be resolved with relatively simple administrative changes. To obtain a classification system which permits development expenditures (whether on recurrent or capital account) to be identified in budgets by projects and programs requires only a functional-economic classification system which identifies development items. The preparation of such a classification system involves some problems, but they are surmountable if a committee composed of representatives of a country’s central planning agency, the ministry of finance, and other interested bodies (e.g., representatives of autonomous public agencies) is set up with appropriate technical assistance to iron out problems.
As for unifying the budget, a delicate problem confronts anyone who advises that the capital and current parts of the budget be prepared by one agency when two are engaged in the task. The advisor who recommends this may have principle on his side, but he is unlikely to convince one agency to relinquish its prerogatives to the other. Given the political and administrative realities in many countries, another course generally has to be found for accomplishing the same result.
Fortunately, another proven method already exists, and the countries visited were advised to adopt it, where there were dual budgets. Nigeria and some other countries have an effective Budgetary Coordinating Committee. Top officials representing the budget office and the central planning agency act as permanent members of the Committee, and a high official of each technical ministry, department, or agency presenting budgetary proposals acts as a temporary member when his organization presents proposals for inclusion in the budget. Each technical organization presents its capital and recurrent estimates simultaneously. The estimates are considered and discussed by the Committee, which reconciles investment and recurrent estimates and brings total budget estimates into line with available financial resources. This procedure not only permits realistic estimates to be made of the impact of capital outlays on recurrent expenditures but also reduces from two to one the number of budgetary presentations which technical organizations have to make.
Programing Units in Operating Organizations
There is general agreement that a central planning agency should prepare over-all plans and that technical ministries, departments, and agencies should prepare sector (or subsector) programs and projects in their fields of interest. Sometimes, central planning agencies take over project and sector program preparation as a temporary expedient when technical ministries, departments, and agencies are unable to carry out these tasks as efficiently or as quickly as planners wish. Experience shows that when a planning agency takes over work which properly belongs to technical organizations, friction almost invariably develops. Moreover, what starts as an expedient often ends up in a long-term arrangement which is hard to change. When this happens, technical organizations are often virtually outside the sphere of planning, with pernicious results for planned development.
Under the proposed pragmatic system of planning advocated here, central planning agencies were expected to prepare annual development plans, in most cases in addition to a medium-term plan, a task which involves a considerable amount of work in a variety of fields, and to follow these through. It was recommended that the central planning agencies should not attempt to prepare sector programs and projects, which belong to—and must be effectively carried out by—technical organizations.
But experience has shown that most ministries, departments, and agencies do not know how to prepare soundly conceived projects and programs, carry these out according to a well-devised cost and time schedule, and operate (or maintain) them efficiently after completion. Experience in India, Pakistan, and elsewhere has also demonstrated that the most effective way of getting personnel in technical ministries, departments, and agencies to perform these tasks properly is to establish programing units in these organizations. The governments visited were accordingly advised to create such units, at least in the most important technical organizations or, where such units had been established, to make them more effective than they were.
Since programing units would constitute a vital link in tying projects and sector programs to annual plans and budgets, and, in addition, would constitute a relatively new administrative concept for many of the countries visited, much time was spent explaining their nature, form, functions, and procedures. The programing unit was described as the virtual equivalent for its organization of a central planning agency for a national government. Its functions were defined to include (under the control and leadership of the head of the organization): (1) the setting of standards and criteria for technical departments or other units to follow in preparing and carrying out projects; (2) the formulation of the over-all development program and the recurrent budget for its organization; (3) the preparation of alternative development policies for the area under the jurisdiction of its organization, after consultation with the various heads of technical departments or other units; (4) the setting of standards for technical departments or other units to follow in reporting on the progress of projects and, on the basis of reports from technical departments or other units, the preparation of reports and evaluations of its organization’s over-all program; (5) the coordination of the technical assistance program for its organization; and (6) liaison for its organization with the central planning agency.
In order to institutionalize liaison between programing units and the central planning agency, it was recommended that an inter-ministry planning committee be created with the heads of all programing units and a representative of the budget office as members and a high official of the central planning agency as chairman.
We know that programing units, to be successful, must be located administratively at a high level. In the British type of government administration, this means that the programing unit ought to be located immediately under the permanent secretary—i.e., the chief nonpolitical official in a government ministry or department. In the French type of government administration, it also means that the programing unit ought to be placed immediately below the highest permanent official in an organization. Indeed, if these officials were not as busy as they are with day-to-day tasks, they would have been the logical heads of the programing units in their organizations. But given the prevailing circumstances in the governments of the countries visited, it seemed more practical to suggest that the head of each programing unit be the second in rank in each technical organization.
It was considered desirable that the head of each programing unit outrank the heads of technical units within each organization because experience has shown that officials do not easily yield their prerogatives to other officials of the same or lower rank. In the countries visited which had established programing units, almost all were ineffectual because, in part, they had been placed at too low a level in their organizations; in part, because they had been assigned duties extraneous to programing; and, in part, because there was little understanding of the special staffing requirements of programing units.
Thus, no special efforts had been made to find suitably trained or experienced persons for the programing units. It was not understood that programing in a technical organization, like planning in a central planning agency, is a highly specialized field which generally requires technicians like engineers, agronomists, etc., who have mastered enough economics and accounting to make cost-benefit analyses for projects and to evaluate projects economically for the purpose of combining them into sound sector and subsector programs on the basis of general economic and other criteria.
Since experience throughout the world has demonstrated that the establishment of viable programing units requires specialized outside technical assistance, it was, finally, recommended that steps be taken to obtain suitable technical assistance for this purpose.
Implementation of the Recommendations
Most of the countries concerned were visited in 1966 and 1967. It would therefore be premature at the time of writing to expect definitive results. However, the steps taken to give effect to recommendations have been encouraging. In some countries, only a few recommendations are being implemented, but in more than half of the countries visited implementation of recommendations has meant that important changes in administrative organization and procedures have been introduced.
Many are adopting the system of annual comprehensive planning cum multiannual sectoral programing which was recommended (although some combine this with medium-term over-all plans). In Thailand and Colombia, task forces were set up in central planning agencies to deploy personnel efficiently for planning; in Ceylon, Malaysia, Morocco, the Sudan, and Ghana, budgetary coordinating committees will consider recurrent and capital estimates simultaneously; and in Ceylon, Ethiopia, Ghana, Ivory Coast, Malaysia, Morocco, and Thailand, steps are being taken to establish programing units in operating organizations. One major difficulty has been encountered: an unusual scarcity of foreign technical assistance for sectoral programing and project preparation. Until we solve this problem, viable programing units cannot be widely established soon in low-income countries. But taken as a whole, it seems fair to say that the outlook for carefully considered partial administrative betterment, when tied to planned development innovations, is good.