Journal Issue

Reviewing Public Investment Programs: How The Bank Works with Countries to Assess Priorities

International Monetary Fund. External Relations Dept.
Published Date:
March 1986
  • ShareShare
Show Summary Details

Basil Kavalsky

Assisting countries in preparing and revising their public investment programs is not a new activity for the Bank. Throughout its existence borrowers have asked the Bank’s advice in formulating five year plans and selecting priorities from among competing project proposals. The Bank’s combination of work at the micro, sectoral, and macro levels makes it well placed to assess particular projects in relation to the government’s broad economic objectives.

The past five years have seen a dramatic increase in the number of reviews of public investment programs undertaken by the Bank. During the 1970s many developing countries raised the share of investment in GDP by expanding their public investment programs. With world recession in the early 1980s and increased debt-service burdens, investment levels in these countries have dropped. In sub-Saharan Africa, for example, investment fell from 18.7 percent of GDP in 1980 to 14.7 percent in 1983. Reductions were particularly severe in the high-debt Latin American countries. The ratio of investment to GDP in Brazil averaged 28 percent from 1973 to 1978 and fell to 22.5 percent between 1979 and 1983. For Argentina the comparable figures were 24.6 and 20.5 percent.

These lower investment levels were the outcome of the need to reduce the public sector deficit in the face of shortages of resources, including those from external borrowing. In response to this pressure, ministries of finance often lowered expenditure ceilings for other ministries by a uniform percentage across the board. Ministries such as agriculture or education were then left to determine the allocations to individual programs. Since cuts in public employment and salaries are generally the most difficult to make politically, the cuts tended to fall disproportionately on nonsalary components of recurrent expenditures—for example, teaching materials or hospital supplies—and on investments. New investments were postponed and those in progress delayed. With the increasing recognition of the high costs of neglected maintenance and delayed construction schedules, governments have turned to the Bank to assist them in identifying the critical activities on which to spend scarce budgetary resources.

Another stimulus to the Bank’s review of public investment programs has come from the Fund. The Fund’s concern with the viability of country adjustment programs over the medium term has demanded that it go beyond assessing aggregate levels of public expenditure and the resulting budget deficits. The implications of unproductive investments for future fiscal and external payments balances suggest a need to examine carefully the allocation of public investment and the process of project selection. For the past few years, therefore, the Fund has been asking for the Bank’s view of the country’s public investment program before making its own decision as to whether to go ahead with a stand-by or extended Fund arrangement. The Bank has also been in a position to advise on whether the ceiling on the government deficit may result in the exclusion of high priority development expenditures. These reviews have therefore provided a focus for the exchange of ideas between the two institutions.

These developments have coincided with the evolution of thinking and practice within the Bank. Since the mid-1970s the Bank has become increasingly concerned about the environment in which its investment projects must operate. In part this is a matter of the policy framework for the economy and the sector in question. Irrigation projects, for example, will not yield their full potential returns if distorted agricultural prices reduce farmer incentives. But neither will they yield their full potential returns if associated investments are not made, say in upgrading farm-to-market roads, or if the institutional support from the research, extension, and credit agencies is inadequate. In planning its lending operations the Bank thus evaluates the policy framework, the public investment program, and institutional support.

The commonality of interest between borrowing countries, the Fund, and the Bank has resulted in the Bank conducting about ten missions a year specifically to review public investment programs. In addition perhaps ten general economic missions a year review public investment programs. These latter focus on the program’s broad composition and macroeconomic impact, and are less detailed in their review of sectors and projects. Most of the Bank’s sector missions (for example, to appraise conditions and prospects in agriculture or transport) now include in their terms of reference a systematic review of the sectoral investment program. In this way public investment programs in virtually all the Bank’s major borrowers are reviewed about once every five years.

Objectives and coverage

What are the objectives of the Bank in reviewing public investment programs? To date these reviews have had two major purposes. The first is to make recommendations on how the program can be tailored to fit available resources at the lowest cost to long-run growth. The second is to evaluate the system of investment programming and to identify ways in which it can be improved to provide a basis for better decision making in the country. In future, as the constraints ease on countries’ resources, the focus will shift to a more forward-looking view of the potential for new investments and the selection by the Bank of new projects for financing.

The first questions that arise when the Bank is asked to take on a public investment review concern the coverage. Should the entire public expenditure program be reviewed or should the scope be limited to investment? Should public enterprises be included or only the government sector? The answer to the first of these questions depends on the circumstances. For example, in low-income countries, particularly those in sub-Saharan Africa where recurrent expenditures have been too low to maintain the capital stock in recent years, it makes little sense to exclude these expenditures from the review’s agenda; the analysis must start by considering the adequacy of current outlays, considering new investments only after providing for the maintenance of existing assets. In the larger middle-income countries, on the other hand, tax resources are usually available for those recurrent expenditures needed to maintain existing facilities and services, and ministries of finance and planning are more concerned that the Bank contribute to the process of project selection.

The issue of whether public enterprises should be covered is an especially difficult one. The Bank’s reviews of public enterprise performance have consistently stressed the need for greater autonomy in the management of these enterprises—autonomy that extends to their investment decisions. Where enterprises are using public funds, however, or relying on government guarantees for their borrowing, there is a legitimate argument for comparing the returns to their use of funds with the returns to other uses.

In principle public enterprises should stand or fall on their financial viability and the government as the major shareholder should assess that viability at appropriate times, but leave to management the decisions involved in the attainment of financial viability. This system works if management and workers in an enterprise are subject to financial discipline, but in most cases public sector agencies face only a “soft” budget constraint and incentives that encourage solving problems through additional investment rather than through improved efficiency. In Brazil huge investments by public enterprises, financed by borrowing abroad, were a major contributory factor in the subsequent debt crisis. For these reasons, Bank reviews of public investment are likely to continue to cover proposals of the public enterprises, but many of the reviews will be combined with studies of how public enterprises are being managed, so as to identify the root causes of problems.


In general the Bank carries out its analysis of public investment programs by sending a group of five to eight general and sector economists to the country for about three weeks. Drawing on the Bank’s prior analytical work on the country, in particular its sector reports, and project experience, extensive briefing discussions usually take place before the visit. In many cases (and as the norm for larger countries) the review is divided into two field phases, each of about three weeks’ duration, separated by some months. The first phase emphasizes data gathering and initiates work within the government while covering some, but not all, the sectors. The second phase analyzes the data, discusses preliminary conclusions, and completes the sectoral review. Much of the value of these exercises derives from the interaction between macro and sector economists and between Bank economists and government economists.

Analytical methods. A fairly standard analytical approach to these programs has evolved over time. This consists of the preparation of a macroeconomic consistency framework agreed upon by the government and the Bank, combined with a detailed review of the individual components of the investment program.

The macro framework is used to assess the consistency of the various sectoral proposals with one another and with the government’s broad economic goals. The framework used is normally the Bank’s Revised Minimum Standard Model (RMSM), which has the advantages of incorporating Bank projections of commodity prices and also allowing the debt implications of different-sized investment programs to be evaluated.

The basic approach used iterates backwards and forwards between the macro framework, the formulation of sector strategies, and the appraisal of individual projects. Formal planning techniques are of limited value in this task. Input-output matrices and material balances may have some role in large, predominantly self-sufficient economies, but for the majority of developing countries they rarely yield useful information at the economy-wide level. They may be a useful adjunct to planning heavy industrial development in countries where this forms a significant part of the public investment program and is mainly oriented to the domestic market. Macro models, for their part, rarely incorporate the amount of detail that is needed to evaluate the impact of individual projects on the economy. Cost-benefit analysis is an indispensable tool for project preparation and appraisal, but it does not form a sufficient basis for assembling a public investment program given the difficulties inherent in quantifying benefits and in dealing with noneconomic objectives (such as the provision of services to disadvantaged groups or regions).

At the micro level, different sectors demand different analytical techniques—for example least-cost studies for power supply or intermodal analyses for transport. Sectoral studies generally require more time to undertake than the mission has at its disposal, so that if they are to be incorporated in the analysis they need to be carried out through prior sector or project work.

Assessing priorities

Macro level The Bank’s reviews therefore proceed on both the macro and sector levels. At the macro level, the country economist must analyze the macroeconomic situation and derive estimates of total resource availabilities, taking into account the government’s own targets and the views of the Fund. This analysis raises fundamental questions about the level of public expenditure relative to private, and the level of aggregate consumption relative to investment and savings.

Differences of view about the government’s plans tend to arise most often over the prospects for external financing, and for commodity prices and the revenues associated with them. Differences can also arise where government projections are targets, based on the assumption that project implementation will improve and making inadequate provision for exogenous events such as the effect of poor weather on harvests. This approach causes a built-in overprogramming of public investment, which must be hastily cut when reality supervenes. Instead, the Bank has tended to argue for a core investment program supplemented by a reserve shelf of projects that can be taken up if enough resources become available.

The first priority in allocating total resources is to ensure that adequate provision is made for recurrent expenditures. Unfortunately, developing countries’ plans are rarely organized so as to make this transparent; most budget documents list recurrent expenditures in a form that relates to means or instruments, not ends or objectives. For example, an allocation for salaries does not show whether the salaries are used to carry out the department’s own planning and administrative functions or to maintain irrigation canals.

Projections of investment expenditures typically give little attention to the implications of new investments for recurrent expenditure. Many governments have taken on obligations they might have chosen to avoid, had they realized what recurrent expenditures were involved. The approach taken by Peter Heller of the IMF in his article in Finance & Development (“Underfinancing of recurrent development costs,” March 1979) is a possible path to a suitable proxy for deriving adequate recurrent expenditure levels. In the long run, however, there is no substitute for governments undertaking such analysis directly. Botswana’s recently formulated economic Plan is a model in this regard. Aggregate recurrent expenditure needs were derived from historical relationships and compared with detailed projections of the recurrent expenditures entailed by new projects, to make sure they were consistent. This procedure allows the full costs of an expenditure program to be evaluated. The Bank is supporting efforts to move toward a similar approach in Zambia.

Sector vs. sector. When the aggregate resource availabilities are defined, the sector specialists are given guidelines on the resources available for investment in each sector. This consideration of sectoral priorities is the most arbitrary part of the review process. In theory it is easy to invent models which will enable one to say that at the margin a certain amount should be shifted from agriculture to transport. In practice, however, the relationships and data requirements are much too complex. The Bank’s analysts therefore use information on the historical shares of sectors, and their experience of the capacity to implement sectoral programs, adjusted in accordance with the government’s stated priorities and by reference to the macro framework. The Bank’s RMSM model calculates historical capital/output ratios by sector, which can be applied to the new investment mix to see what it implies for output. This is valuable in raising questions and making the underlying assumptions explicit.

Sectoral programs. It is at the level of individual sector programs that the Bank is on its firmest ground, particularly where the review has been preceded by in-depth sector work. During a public investment review mission much of the emphasis is on clarifying sectoral objectives and strategies in consultation with the responsible government officials, and then on assessing whether the project and program proposals are consistent with these objectives. Even when information is readily available, a program will rarely be organized in the form of neatly evaluated project proposals with rates of return calculated for each. In the time available, therefore, the Bank staff must use relatively simple criteria to evaluate the projects and programs.

Some features of the evaluation are common to all sectors. For example, the extent to which sectoral objectives are likely to be achieved is always assessed in relation to the total package of expenditure programs, sectoral policies, and institutional capacity. In every sector the largest investment projects are scrutinized to assess their economic viability. In addition, specific techniques are applied to evaluation in some sectors. In power development a level of demand is derived and then a least-cost program is designed to meet that demand. A transport evaluation generally starts by ensuring that the system as it stands can be maintained. New investments are then looked at in coordination with the review of the agricultural and industrial proposals, to judge their priority. In the social sectors where capital expenditure is usually a minor part of the expansion program, the coverage and quality objectives (for example, health care at what level of sophistication, for what parts of the population?) are looked at and the needs for increased expenditure evaluated against them. Consideration is also given to whether a more economical use of existing resources could allow a reduction in proposed expenditures.

The key to reviewing public investment programs is the iteration among the sector specialists and between the sectoral and macroeconomists. This permits different options to be identified and the program to be modified accordingly. For example, the industry specialist may calculate that a proposed industrial project will not yield adequate returns, and its exclusion from the program may in turn permit a reduction in the power investment program. In most cases the linkages are of course far more judgmental and take longer to have effect. For example, the rates at which different sectors are expected to develop may suggest that priority be given to various levels of education over the longer term, but the available information will almost never permit a precise quantification of this relationship.

The policy framework

As indicated earlier, a coherent view of public investment programs must encompass the policy and institutional environment. In carrying out public investment reviews it is particularly important to evaluate pricing policies. In the energy sector, for example, appropriate pricing policies give an impetus to conservation with consequent reductions in the levels of investment needed in energy-using sectors such as power and transport. The recent review of Mexico’s public investment program found that raising petroleum prices toward world levels resulted in substantially less growth in electric power use and that investment plans in the power sector could therefore be scaled down.

Pricing policies are important not only in sectors producing physical outputs. When services are offered free or at highly subsidized prices, the excess demand does not necessarily mean that additional public services should be provided. In the social sectors, governments generally provide substantial subsidies for users. The rationale for subsidies is that these services are directed at the poor, but in the Bank’s experience this is often not the case. Thus university studies, which yield very high private returns in most developing countries, are often far more highly subsidized than primary education. Similarly it is not uncommon to find medical care systems in which the highest subsidies go to patients in private or semi-private hospital rooms. Even where subsidies reach the poor, there is evidence that the poor would be willing to pay more (whether for public or private services) if that meant better quality, or easier access to services. For example, studies carried out by the Bank in some of the Sahelian countries suggest considerable willingness and ability to pay for better education, and the extent to which children from quite low-income families in Africa are sent to private schools supports this. Introducing charges that allow public services to recover their operating costs is often a critical step toward using these services more efficiently and providing additional resources for their expansion.

Stronger planning, coordination

The end-product of the Bank’s public investment reviews is not a new public investment program, but a process within the government concerned. The reviews are intended to highlight crucial questions, relating to the program’s design and financing, for resolution within the government. Even more than most Bank activities, the reviews are conceived as collaborative exercises, being carried out jointly by Bank staff and the ministries or departments concerned. The reviews focus explicitly on institutional issues, evaluating the country’s process for identifying and selecting projects and for integrating them into a macro framework. These missions often result in technical assistance to the government in those phases of the process that need strengthening.

Public investment reviews are becoming an increasingly important mechanism for aid coordination. For donors, a project’s inclusion in a program that has been reviewed provides some assurance that there is agreement on its economic priority. For the country, the program provides a line of defense against donors’ tying of aid to projects that seem mainly designed to benefit equipment suppliers in the aid-giving country. These documents are increasingly being asked for by aid group meetings and are being discussed at them.

Issues for the future

With all their positive features, there are a number of areas where steps could be taken to make the Bank’s reviews more effective. The reviews are strongest where they rest on the framework of in-depth sector work carried out by the Bank and the country, but often judgments must be made only on the basis of what the specialist can learn during a short stay in the field. The Bank will need to expand and deepen its analysis of sectoral policies and investment programs, and plan the timing of these analyses so as to provide inputs into its reviews of the overall investment program.

A second issue is the quality of information. Though the government is usually requested to prepare information on the public investment program in advance of the mission, it is surprising how difficult this often proves to be. Many governments do not have information on individual projects. And where state and local governments have considerable autonomy, as in Brazil for example, it may be difficult for the central government to get precise information on their activities. Information on public enterprises is almost always difficult to obtain. One of the important benefits of the Bank’s review in many countries is precisely that it provides an impetus to the gathering of information in one place, so that officials in the central ministries can consider the likely impact of the investment proposals. The Bank needs to take the initiative through its continuing economic and sector work and through technical assistance in upgrading the data base well in advance of mounting public investment reviews. Adequate data will be particularly important in following up on the implementation of the agreed program.

A third issue is that the analytical framework does not permit exploration of the longer-term effects of a package of investment projects on the availability of government revenues and on the country’s foreign exchange earnings and burden. In many of the large Latin American debtors, the projects financed by foreign inflows were viable, ex ante, but crises were caused by difficulties in converting their returns into government revenues and foreign exchange. Bank reviews need to pay close attention to this problem in future, by explicitly reviewing the operating and maintenance costs associated with the investment program, the potential for recovering costs, and the foreign exchange implications.

Public investment reviews will remain an important part of the Bank’s efforts. They are of substantial use to the Bank in deciding on its own lending strategy. More important, the authorities in most of the countries for which they have been undertaken have expressed their satisfaction with the outcome. The reviews provide an opportunity to take a fresh and objective look at investment programs that over time have become mere collections of projects and have lost their coherence in relation to the government’s own goals.

One clear lesson from the experience so far is that these reviews are most effective when they are carried out in collaboration with the government concerned. Both the Bank and the government learn in the process and the resulting recommendations have a greater credibility than if they originated with the Bank or the government alone. Future reviews are likely to rely increasingly on such collaboration.

Finance & Development is available on microfilm from University Microfilms, P.O. Box 1346, Ann Arbor, MI 48106, USA, and on microfiche (English only) from Microphoto Division, Bell and Howell Company, Old Mansfield Road, Wooster, OH 44691, USA

Other Resources Citing This Publication