Country Commitment to Development Projects Richard Heaver Arturo Israel ~~ World Bank Discussion Papers Country Commitment to Development Projects Richard Heaver Arturo Israel The World Bank Washington, D.C., U.S.A. The World Bank 1818H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United Statesof America First printing November 1986 DiscussionPapers are not formal publications of the World Bank. They present preliminary and unpolished results of country analysis or research that is circulated to encourage discussion and comment; citation and the use of such a paper should take account of its provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Any maps that accompany the text have been prepared solely for the convenience of readers; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area or of the authorities thereof or concerning the delimitation of its boundaries or its national affiliation. Because of the informality and to present the resultsof research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank acceptsno responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The most recent World Bank publications are described in the catalog New Publications, a new edition of which is issued in the spring and fall of each year. The completebacklist of publications is shown in the annual Index of Publications, which contains an alphabetical title list and indexes of subjects,authors, and countries and regions; it is of value principally to libraries and institutional purchasers. The continuing research program is described in The WorldBank Research Program: Abstracts ofCurrent Studies, which is issued annually. The latest edition of each of these is available free of charge from Publications SalesUnit, Department F, The World Bank, 1818H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'ICna, 75116 Paris, France. Richard Heaver is an economist in the Population, Health, and Nutrition Department of the World Bank; Arturo Israel is head of the Public Sector Management Unit in the Bank's Projects Policy Department. Library ofCongress Cataloging-in-Publication Data --- -- Heaver, Richard 1952- , Country commitment t o development projects. (World Bank discussion papers, ISSN 0259-210X ; 4) 1. Economic development projecte--Developing countrfes. 11. Israel, Arturo, 11. Title. TIT. Serfea. HC59.72.E44H43 1986 338.91 86-28114 ISBN 0-8213-0860-2 ABSTRACT Although the commitment of the Borrower to implementation is widely recognized as one of the key factors affecting project performance, the Bank has developed no systematic approach to analyzing the question of commitment. The paper is a first attempt to review the concept of country commitment and to discuss ways of assessing, building and sustaining commitment, in the context of Bank project design and implementation. The model of human behavior implicit in the paper is one of interest groups, bargaining and conflict, and it stresses the need for analyzing the political factors affecting Borrower behavior as well as the technical and economic factors on which the Bank has traditionally concentrated. With regard to the concept of commitment, the paper notes that commitment is not monolithic; that it varies both over time and between different Borrower actors depending on the degree to which project proposals or activities affect their interests. Commitment is also likely to vary according to the type of program activity; in general, there is greater commitment to "hard" technology activities such as construction, and less to "so£ter" technologies such as institutional development efforts. The paper suggests a practical approach to assessing commitment, centering on the identification of key Borrower actors who have the power to influence program outcomes, and analysis of the way in which informal incentives and pressures may motivate them in directions other than the official project or program goals. With regard to influencing commitment, the paper notes that there is a spectrum of possibilities, ranging from adopting project designs which fit the existing pattern of commitment; to building commitment through education and persuasion; to seeking changes in organizational structures and processes which will give those who are committed greater influence over those who are not; to the use of financial leverage. It is argued that Bank staff should develop explicit strategies for influencing commitment, often combining elements of these four approaches, but always stressing the importance of genuine dialogue where both Borrower and Bank are prepared to listen, learn and modify their approaches. A number of case studies illustrate practical ways of assessing and influencing commitment. TABLE OF CONTENTS Page A. INTRODUCTION 1 B. CHARACTERISTICS OF COUNTRY COMMITMENT 2 - Commitment varies among actors 2 - Commitment varies over time 3 - Commitment varies according to the type of program activity 3 C. ASSESSING COUNTRY COMMITMENT 4 - Identifying key actors 4 - Assessing the objectives of key actors 5 - Assessing the commitment of external interest groups 6 - Assessing commitment in the context of Bank operations 7 D. STRATEGIES FOR INFLUENCING COUNTRY COMMITMENT 9 - Minimizing conflict and building commitment in the choice 9 of program activities - Building commitment through dialogue with the Borrower 10 - Influencing commitment through changes in Borrower 11 organizational processes and structures - Using financial leverage. 12 E. FOLLOW-UP 13 ANNEXES I. Assessing Informal Versus Formal Commitment: A South Asian 15 Resettlement Scheme. 11. Focus Group Interviewing: Assessing the Commitment of Health 18 Service Providers in a West African Country. 111. Using Workshops to Build Commitment: Policy Reform in the 22 Electric Power Sector of a Latin American Country. IV. Using Workshops to Build Commitment: Agricultural Investment 25 Planning in a Central African Country. V. Using Workshops to Build Commitment: Project Launch Workshops 27 VI. Designing a Management System to Increase a Bureaucracy's 29 Commitment to Serve its Clients: The Training and Visit System of Extension. VII. Organizing Communities to Increase the Commitment of 31 Bureaucracies to Serve Them: Communal Irrigation in the Philippines. A. Introduction 1. The positive effects of a strong Borrower or country commitment to project implementation are almost universally recognized as one of the main factors explaining the success of a lending operation. Conversely, lack of commitment is one of the most commonly quoted causes of unsatisfactory project or program implementation and of lower than expected impact. Yet it is seldom made clear what adequate commitment or lack of it means, or what can be done to increase and sustain commitment. The adverse results of weak commitment may range from minor benefit losses or wst increases, as when a minor project component is inadequately implemented, to total project failure, as when there is a fundamental lack of commitment to an entire program--a land reform program, for example. 2. Weak country or Borrower commitment can occur when one or more key Borrower actors--whether institutional or individual--fail to support the agreed goals of the project or program; the agreed priority attached to meeting the various goals; or the agreed means of achieving them. Commitment may be formal or informal. The key case of interest to the Bank is when the Borrower formally declares commitment, but when significant actors are informally uncommitted. Such a situation is illustrated in Annex l. 3. Several factors tend to discourage attention to the issue of commitment in Bank work. First, analysing commitment is conceptually difficult and necessarily subjective: there is no accepted means of measuring or building it, and Bank staff with technical training may feel uncomfortable in an area which essentially requires a political understanding of Borrower actors and their motives. Second, the pressure to lend can discourage the analysis of intangibles which may threaten a project's clean bill of health at appraisal. Third, the procedures which have been developed for most other aspects of the preparation process may in themselves crowd out correspondingly full consideration of factors--such as commitment--for which there are no accepted procedures in Bank work. 4 . This paper is a first attempt to review the concept of country commitment and to discuss ways of assessing, building and sustaining commitment, in the context of Bank project design and implementation. It is not, of course, a first systematic approach to the question of commitment, which has been treated, directly or indirectly, by political economists and writers on development, perhaps most notably Albert Hirschman; and by economists and organization theorists, perhaps most notably Herbert ~irnon.'/ - This paper, as a first operational paper on this - l/ Thus, for example, the model of human behavior implicit in this paper is one of interest groups, bargaining and conflict (political theory), in an environment where uncertainty is high and rationality bounded, so that organizations maintain considerable slack (Leibenstein) and individuals satisfice rather than maximize (Sirnon). Similarly, the strategies for influencing commitment discussed in this paper could be reinterpreted in Hirschman's terms, as ways of increasing the voice of the Bank or of Borrower interest groups; building loyalty; or threatening exit. subject, is limited i n scope. It does not fully explore the commitment issues raised by major policy-based lending operations such as s t r u c t u r a l adjustment loans, which often have a p o l i t i c a l dimension closer to the IMF's stabilization loans than t o traditional Bank operations (see a l s o paragraph 26). Given the growing importance of t h i s category of lending, it deserves separate treatment. The paper looks f i r s t a t some basic characteristics of commitment which, while often obvious, are often not taken i n t o account. Then the paper discusses what may i n practical terms be done to assess, build and sustain commitment. B. Characteristics of Country Commitment 5. The concept of country commitment can mean different things to different people. From the viewpoint of Bank operations it means two things. F i r s t , that there is a coincidence of program or project objectives and of methods f o r achieving those objectives, between the country and the Bank. Second, that there is in the country a positive coalition of i n t e r e s t s among the key actors involved to support those objectives. A "program" could refer to traditional project or sector operations, but also t o policy-based lending. Thus, commitment can apply t o investments, i n s t i t u t i o n a l changes and policies. The concept of commitment has several facets, as discussed below. 6. F i r s t , commitment varies among actors. Country or Borrower commitment is often discussed as i f the country were monolithic. In practice, t h i s is never the case. Different ~brroweractors w i l l have various degrees of commitment to different program objectives according to whether these further or conflict with t h e i r particular interests and p r i o r i t i e s . It is useful to distinguish between institutional, group and individual commitment. Thus, there may be formal commitment from the executing agency of the program, but covert opposition or lack of interest on the part of certain individuals or groups within the executing agency o r agencies. Most commonly, there are differences in commitment between levels of the hierarchy, a s when the Bank and policy-makers agree on a reform, but implementing staff are not committed to change; o r the reverse situation, when the Bank can agree on a course of action with o f f i c i a l s a t the technical level, but where change is unacceptable a t the p o l i t i c a l level. 7. It is also useful to distinguish between the commitment of actors internal and external to the project's implementing agency or agencies. Such external actors may also be: a ) individuals f o r example, a politician; b) groups for example, landlord and tenant groups among beneficiaries; i n social or people-oriented a c t i v i t i e s , the commitment of beneficiaries is as important as that of program officials; o r c ) i n s t i t u t i o n s f o r example, i n ' u s t r i a l companies affected by a t a r i f f reform, o r central government agencies able to contribute or deny resources to the program. Internal actors w i l l by definition have a high degree of influence over project outcomes; but so may the Bank, through its resource contributions and conditionality, have considerable influence over internal actors. I n general, the Bank w i l l have l e s s influence over some external actors--project beneficiaries, f o r example--but external actors may have c r i t i c a l influence on project outcomes. However, the Bank may have influence on central or core agencies through other lending operations or the s i z e of the lending program. 8. Second, commitment varies over time. Commitment may increase o r decrease during program preparation and implementation as the incentives and pressures affecting actors change, or as the actors themselves a r e replaced. Common examples of the l a t t e r include changes of government, ministerial reshuffles, creation or reorganization of agencies, p o l i t i c a l appointments of state-owned enterprise managers, transfers of key c i v i l servants. Examples of the former include changes i n government policy o r p r i o r i t i e s , i n the a v a i l a b i l i t y and cost of financial or other resources, i n personnel regulations affecting pay, promotions or postings, or i n market prices and demand. The likelihood of some of these changes taking place is very high, considering that a typical program or project cycle lasts seven to ten years (although less for most of the policy-based lending.) Some of the above events--such as a change i n government-may lead to a change of formal commitment, and program or project cancellation or renegotiation. Most are more l i k e l y to lead t o a loss of informal rather than formal commitment, which may be evidenced by vacancies in or inappropriate staffing of key posts, by f a i l u r e to release counterpart funds f o r certain a c t i v i t i e s , or by a slow-down i n project implementation. The variable nature of commitment over time means that the task of building up and then sustaining commitment is an essential one throughout the whole cycle of a program; it must be seen as a continuing process. 9. Third, commitment varies according t o the type of program activity. The degree of commitment is a t l e a s t p a r t i a l l y predictable according t o the type of program or a c t i v i t y involved. In general, there w i l l be greater commitment to 'hard' technology a c t i v i t i e s such as construction, where the payoffs are v i s i b l e and where inadequate implementation is obvious, measurable, and often dangerous--the half-built o r structurally weak dam is an example. There may be less commitment t o 'softer' technologies, for example to quality primary education, family planning, environmental and i n s t i t u t i o n a l development efforts, where outcomes are hard to measure or poor performance is not immediately c r i t i c a l . This distinction is particularly important f o r technical assistance components: actors within countries are more likely to accept technical assistance i n high technology a c t i v i t i e s , such as complex i n d u s t r i a l processes, than i n education or i n s t i t u t i o n a l development, for which solutions are (often rightly) considered t o be more local and more l i k e l y to be found through local expertise. Also country commitment may be low i n projects or project components aimed at benefitting the very poor; not only are technologies s o f t , but the very powerlessness of the poor gives them l i t t l e voice i n the a c t i v i t i e s of the bureaucracies supposed t o serve them. 10. Borrower commitment may also vary with the different a c t i v i t i e s undertaken a t different stages of the program cycle. For example, program o r project preparation a c t i v i t i e s often unite a variety of actors with a common i n t e r e s t i n securing external resources, but who may or may not share an i n t e r e s t i n the efficient use of resources once secured. This is often true of projects which consist of f i r s t a construction and then an instance. Commitment to the first phase, where performance is visible and easily measurable, will often be greater than to the second phase, where inattention to preventive maintenance is not immediately critical, and where minimal bureaucratic reward is coupled with a drain on scarce recurrent funds. Education and health projects, where an infrastructure development phase is followed by a service delivery phase, are another example. In these cases, it is not that commitment is lost at a particular point in time; effective commitment to the neglected activities was never present; but only at particular points of the program cycle does the divergence between formal and informal commitment manifest itself. 11. Commitment to complex multi-agency programs--or multi-donor, for that matter--is likely to be weaker or less uniform due to the large number of actors involved; integrated rural development projects, with multiple implementing agencies and beneficiaries, are an example. Probably the most complex network of commitments occurs with some types of policy-based operations, such as SALs. A broad structural adjustment program usually involves a large number of central and sectoral agencies, not all of which, especially those which do not benefit directly from the financial resources channeled through the SAL, will have equal commitment to the program. In addition, policy changes introduced under SALs will often be to the disadvantage of particular social or corporate groups, who can be expected to be opposed to particular elements of the program. 12. Thus, the notion of commitment, to be operationally relevant, has to be disaggregated by type of activity; by the actors within the implementing agency or agencies; by the actors external to the implementing agency; and by time periods over the program cycle. C. Assessing Commitment 13. A reasonably systematic approach to assessing commitment is both possible and applicable by Bank staff in the context of routine operations. Only in certain cases involving policy-based lending may specialized skills in political analysis be required (see para. 26). The sections which follow focus first on assessing the commitment of internal actors, then on assessing the commitment of external actors, including other government agencies, project beneficiaries, and interest groups in society who will be affected by the implementation of a lending operation. The assumption on which the approach is based is that Borrower actors will be more or less committed to a project according to the degree to which it furthers or conflicts with their own interests and objectives. 14. In assessing commitment, it is important to distinguish between lack of progress due to weak commitment from that due to operational inefficiencies. Given the widespread institutional weakness in many countries, strong commitment is a necessary but not sufficient condition for achieving progress in program and project implementation. 15. Identifying key actors. Not everyone's commitment is equally intense, or equally important to program success. Therefore, an assessment of commitment should concentrate first on identifying individuals and groups who share common interests and who have the power to influence program ormance */ A t the upper levels of implementing organizations, key individuals may Fe important; a t lower levels, wider groups are l i k e l y t o share the same influence and interests--for example, the e n t i r e group of f i r s t - l i n e supervisors i n an extension service. Interest groups may be defined around levels i n the hierarchy (managers versus foremen), departmental loyalties (marketing versus production), professional l o y a l t i e s ( i r r i g a t i o n engineers versus r u r a l development specialists), cultural loyalties (religious, ethnic or t r i b a l loyalties), or age (which may determine attitudes to or p o s s i b i l i t i e s f o r career advancement). Intersecting with such i n t e r e s t groups w i l l be patron-client i n t e r e s t groups, which may or may not coincide with formal departmental organizations. 16. Most organizations w i l l be made up of large numbers of intersecting, and often shifting coalitions of interest. Once identified i n outline, these groups can be reduced t o a few key actors (groups and individuals), whose s t r a t e g i c placing gives them significant influence. Given the generally hierarchical nature of bureaucracy and public enterprise, the i n t e r e s t s of more senior individuals and groups tend t o determine the incentives and actions of those below. An assessment of commitment can therefore start with key actors a t or near the head of the organization. If they effectively share the aims of the program design, attention can be turned to those next i n the hierarchy, with the a i m of determining where conflicts i n motivation occur which break the chain of common interest; f o r it w i l l be a t these points that commitment w i l l be lost. 17. Assessing the objectives of key actors. The commitment of an individual or i n t e r e s t group is no more monolithic than the commitment of a Borrower. In assessing the commitment of key actors, it is useful to divide the incentives and pressures which together determine objectives and hence commitment, i n t o different groups whose strength may vary, and which may complement or conflict with each other: ( i ) Official objectives and pressures. The o f f i c i a l goals and pressures influencing key actors are government policies and procedures, and the agreed goals and methods of the program. While these may be clear t o Bank s t a f f and senior Borrower managers, the same may not be true f o r more junior Borrower officials. Key middle-level s t a f f who are unclear about the goals of a program and t h e i r role i n it, cannot be committed i n any r e a l sense of the word t o a program. ( i i ) Inf ormal objectives and pressures. Inf ormal pressures on s t a f f r e f l e c t the informal goals of key actors, where these are not i n agreement with o f f i c i a l goals. For example, where a state-owned enterprise manager's informal goal is personal profit, subordinates are unlikely to be committed t o maximizing corporate profit; or, where there is p o l i t i c a l pressure to undertake v i s i b l e development a c t i v i t y i n r u r a l areas, roads and health centers may be b u i l t , but there may be l i t t l e commitment to road 21 A process more formally known i n organization theory as environmental scanning and stakeholder analysis. See f o r example EDI Course Note CN848, 'Analyzing the Project Environment'; July 1980, and Staff Working Paper No. 375, the Design of Organizations f o r Rural maintenance or the quality of health services once in operation. In practice, informal goals will often be more important than formal goals. In this context, though this can be hard to remember from a Bank-centered view of the world, the Bank and its goals will often be perceived as peripheral by Borrowers. (iii) Personal objectives. Commitment may be influenced not only by incentives and pressures stemming from outside the individual, but by personal motivation. Personal values and ambitions for self, family or special group will often determine attitudes towards work and career, and hence influence the pattern of informal pressures affecting subordinates. 18. A key actor's commitment to a project will therefore be determined by the relative strength of his personal and informal objectives, as compared to agreed formal objectives, and by the degree to which the project will support whichever objectives are dominant. Creating a situation where key actors cannot ignore formal objectives during implementation (i.e., where formal goals become the dominant objective), or where there is a minimum of conflict between formal and other objectives, is a major task in program design (see para. 27 ff). Identifying key interest groups and individuals and assessing the incentives and pressures affecting them, is a precondition of successfully creating such a situation during implementation. Thus, the key questions to be answered by Bank staff is who are the winners and who are the losers from the project as proposed? 19. Assessing the commitment of external interest groups. The previous analysis has referred to the different factors determining commitment within an implementing agency. A similar analysis can be made with regard to the commitment of individuals and interest groups outside the implementing agencies who can influence project or program outcomes. Where the key actors are public agencies, their identity will often be obvious; for example, the ministry of finance, the planning office, or the parent ministry, in the case of an independent agency or public enterprise. But the pattern of influences will depend on each circumstance; for instance, the actual power of planning offices varies widely among countries, and the in£luence of the ministry of finance will depend crucially on whether the implementing agency requires budgetary resources or not. In interagency relations, commitment can often be more easily assessed to the degree that it is determined by the top management or a particular group within the agency and does not require the full participation of actors at all levels, unlike the case with the implementing institutions. For example, the support of the planning agency for a sectoral program will be determined by the head of the agency and the person in charge of that particular sector. It should also be noted that interagency power relationships work both ways. The implementing agency or agencies will be able to in£luence other relevant agencies to a certain degree, thus affecting their commitment. 20. Commitment may be less easily assessed in the case of lending operations where the role of external actors other than government agencies is important, and the Bank has little direct access. These are in particular policy-based lending, where for example it may be difficult to assess the influence of industrialists who will lose from implementation of a t a r i f f reform program, or the reaction of consumers to an energy price increase; and people-oriented projects, where f o r example it may be d i f f i c u l t t o predict slum-dwellers' willingness to pay f o r maintenance of a water supply system, or farmers' readiness t o adopt a new technology. Finding out about commitment can be easier to the degree that organized publics are involved. This is more often the case with policy-based lending operations, where affected i n t e r e s t groups such as corporations and trades unions may be organized, a r t i c u l a t e and p o l i t i c a l l y powerful. Where organized publics are absent, as is the case i n people-oriented projects whose beneficiaries are distant, scattered and inarticulate, assessing commitment--i.e. i n t h i s case, the likelihood of participation-may require a deeper analysis than what Bank s t a f f can undertake on routine missions (see para 25). 21. Assessing commitment i n the context of Bank operations. Because lack of commitment may make it prudent not t o process further a project or program which i n i t s e l f is developmentally desirable, commitment should i n i t i a l l y be assessed during sector work or a t the project identification stage, i.e., by the Bank project officer or task manager, before consultants and technical assistance are normally involved. An early assessment of commitment is also important because t h i s should affect basic project design, whether i n terms of the choice of goals, methods, components, implementing and coordinating agencies, or conditionality. During preparation and appraisal, a more thorough assessment of commitment should then trace through the network of a c t i v i t i e s involved i n implementation, stop at each c r i t i c a l a c t i v i t y or assumption (e.g. about coordination), and assess whether the incentives e x i s t f o r implementing o f f i c i a l s and external actors t o behave as planned. Finally, given the v o l a t i l e nature of commitment, and the turbulent context of organizations and environments i n most developing countries, a continuing watch must be kept during project implementation. 22. Bank staff probably has the largest comparative advantage f o r assessing commitment t o a particular policy or project. The nature of the work of programs and projects staff and still more that of resident mission s t a f f allows them t o acquire the type of country knowledge that is required t o perform t h i s task. Consultants can be useful t o the degree that they simultaneously have in-depth knowledge of the country and of the Bank 'S policies. However, assessing commitment is a substantial and important task, requiring more Bank staff time than is generally devoted t o it now. 23. Assessing the commitment of implementing organizations is complicated t o the degree that Borrower o f f i c i a l s have incentives to disguise personal and informal goals, or may only be p a r t i a l l y conscious of the e f f e c t s on t h e i r behavior of others' informal goals. There are two main practical methods of assessing commitment. The f i r s t , assessing motives, incentives and pressures through in£ormal discussion with Borrower s t a f f , raises the dangers of subjective judgement, and the projection of the evaluator's own beliefs and assumptions onto Borrower s t a f f . Nevertheless, such discussion, i n particular with more junior Borrower s t a f f , often reveals a very different view of organizational resources, goals and p r i o r i t i e s from those gained from senior managers, and may be a good clue t o informal p r i o r i t i e s . This s o r t of assessment can usually be carried out i n the routine course of one-on-one economic and sector work (ESW), policy and project discussions. But sometimes, where it is necessary to assess the commitment of groups rather than individuals in an organization, a more structured approach may be useful. Guided interviewing of groups of peers in the absence of their supervisors can be particularly valuable, and can be non-threateningly presented as discussion of operational problems and solutions (Annex I1 contatns a case study of the focus group method of interviewing). 24. The second, more objective test of the commitment of an implementing agency is performance checking to test the consistency of officials' professions about priorities with their actions in practice. If, for example, non-action in certain areas declared to be a priority cannot be explained by insurmountable resource or skill gaps, then actions and non-actions and the amount of time spent on different activities will reflect real priorities in a form of revealed preference. The evidence of performance provides an objective check on whether there is genuine commitment to program activities, and will complement the findings of interviews. Assessing performance poses obvious problems at the program identification stage; yet, it is often possible to find projects or programs similar to that proposed, and to draw inferences from these about potential commitment to the proposed program activities. 25. Assessing the commitment of actors outside government agencies presents significantly greater practical problems for Bank staff, especially in the case of people-oriented projects, given the physical and cultural distance of beneficiaries, and the language barrier. More use could be made of locally recruited sociologists and anthropologists to conduct structured interviewing andlor participant observation of groups affected by the project. Details of these techniques are available from the Public Sector Management Unit, in case study form and in the draft of a book entitled "Listen to the People; Operational Evaluation of Development Projects" by Lawrence Salmen. These approaches may often be more productive than the commonly used questionnaire survey, as a tool for discovering social relationships and attitudes. For Bank staff, there is no substitute for more time in the field talking with project beneficiaries--something which has been systematized in the supervision methods developed for the Bank's agricultural extension projects using the training & visit approach. Regional managers may need to increase the coefficients for selected people-oriented projects in order to allow for this. 26. Assessing the commitment of external actors affected by major policy-based lending poses problems of a different dimension, since these operations often have broadly political and not just bureaucratic-political implications (this is not to minimize their bureaucratic-political effects; SALs, for example, often involve strengthening the control of finance or planning ministries over line ministries or public enterprises, and so can give rise to intense bureaucratic conflict). This raises the issue of whether the Bank should develop, within programs divisions, a formal capacity for political assessment, as a key input in formulating the Bank's position on the scope, timing and pace of reforms. Even without a change in the mix of staff skills, programs divisions could be encouraged to do more systematic political assessments. In particular, they could use the country team approach for a more systematic collation of headquarters g- staff's assessment of the motives and influence of key actors and interest groups. D. Strategies for Influencing Country Commitment 27 At least four conceptually different approaches are available to Bank staff in reacting to the results of an assessment of commitment in a given project situation. First, the Bank may accept the pattern of Borrower commitment as it stands, and react by advocating projects, project components or policy changes in areas where commitment is high, and avoiding those where it is weak. Second, it may use its dialogue with the Borrower to build commitment through education and persuasion. Third, it may attempt to influence commitment through negotiating changes in the structures and processes of Borrower organizations so that Borrower actors who are committed have greater influence on those who are not. Fourth, the Bank may make use of its financial leverage. These approaches are on a continuum, from most reactive, to most proactive. In practice, they can most profitably be used in combination, with the Bank both adapting to and attempting to change the political environment. For example, in designing a first state-owned enterprise (SOE) reform project in a country, the Bank may avoid a privatization component which would be highly controversial; initiate a dialogue on greater SOE operational autonomy in the context of a mutually agreed strategic plan; help set up a holding company with financial powers, and improve accounting and reporting procedures; and make reduction in SOE deficits a condition of release of SAL tranches. What is important is that the Bank should have both a conscious strategic plan for assessing and then influencing Borrower commitment, and that the plan should be adapted during implementation in the light of its effects and in the light of changes in the environment. The different possible elements of a strategy for building commitment are discussed in more detail below. 28 Minimizing conflict and building commitment in the choice of program or project activities. While for a given project or program type, certain components may be a sine qua non of Bank involvement, there may be considerable discretion over the inclusion of others--a discretion which may be widening in sectors and regions (e.g. rural development and Sub-saharan Africa), where the Bank has recognized some previous designs as over-complex. In such circumstances, components may be avoided where the Borrower is likely to be uncommitted, and where this may lead to poor performance or conflict between Borrower and Bank. The application of this principle may be less obvious than the theory, which is something of a truism, since program components which are apparently technical and therefore uncontroversial in nature, may have important bureaucratic-political consequences for key actors, and thus raise issues of commitment; an example might be a technical assistance component for improving accounting and financial reporting. 29. The corollary of this principle is that, at least in the early stages of the Bank's involvement with an institution, programs should focus on areas where consensus can be found between implementing agency (and key interest groups in the country more broadly) and Bank. For example, where an urban development authority's main problems are in resource mobilization and efficient resource allocation, the thrust of an initial project might be in improving the efficiency of revenue collection, an activity which w i l l often be of i n t e r e s t to key implementing o f f i c i a l s (i.e., the attractions of expansion w i l l outweigh the complaints of the newly taxed). A subsequent operation might successfully tackle the more sensitive policy issue of investment programming only a f t e r the f i r s t project had developed the Borrower's commitment to working i n partnership with the Bank. Having successful projects is probably the most effective way of strengthening commitment. 30. For the Bank, there are d i f f i c u l t trade-offs to be made. The obverse of the point that commitment may decline over time, is t h a t commitment can be b u i l t up over time. The Bank must decide how f a r it is prepared t o s a c r i f i c e immediate objectives which are developmentally desirable but p o l i t i c a l l y sensitive, i n the hope of building the commitment needed t o tackle the sensitive issues over the longer term. A t the same time, the Bank must show f l e x i b i l i t y during implementation t o modify or sometimes even eliminate components i f it becomes clear that commitment has disappeared. Policy-based lending often requires an exception t o t h i s approach of minimizing conflict i n the short term and building broad consensus over the long term. Policy reforms can both be urgent and unavoidably involve winners and losers; often, the l a t t e r cannot be expected to become reconciled to the changes proposed. In such cases, i f key actors i n the Government and the Bank are convinced that the developmental benefits of reform outweigh the short term social and p o l i t i c a l costs, they may decide t o proceed i n the face of opposition. 31. Building commitment through dialogue with the Borrower. Commitment r i s e s i n direct proportion to the degree to which Borrowers f e e l projects t o be developed by themselves rather than imposed by the Bank. A s a means of building commitment, there is no substitute for dialogue i n a process of joint BankIBorrower identification of projects, when t h i s is seen as an exercise i n which both sides l i s t e n , learn and modify t h e i r approaches. Given the existence of multiple Borrower i n t e r e s t groups, t o be effective the dialogue cannot be confined to a small number of senior policy-makers; many examples of a wider process of consultation and education are available from the Bank's experience. Annex 111 contains a case study of a process of joint policy formulation i n the e l e c t r i c power sector i n a Latin American country. Annex I V i l l u s t r a t e s the use of structured workshops, involving agriculture o f f i c i a l s with varying i n t e r e s t s and from different levels and agencies, t o surface differences and develop consensus on an investment program i n a central African country. Annex V discusses the use of project launch workshops to disseminate a project concept and strategy to a l l those involved i n implementation, and t o e n l i s t their public commitment to project execution. Each of these techniques can also be used during implementation t o review progress and strategies, and to help sustain commitment. 32. A special investment of time spent i n reaching joint understanding with key actors is warranted i n projects where weak commitment may be anticipated a priori, or where changes proposed under the project are l i k e l y t o be particularly controversial. With regard t o the former, as noted above, weak commitment can be expected i n areas where achievement is hard t o measure and hence poor performance not e a s i l y detected; and those where achievement is i n the long term, and hence offers l i t t l e immediate p o l i t i c a l or bureaucratic gain. Particular areas where commitment is frequently weak in the Bank's experience, are a) technical assistance in the area of institutional development, b) monitoring and evaluation components, in cases where improved performance reporting might be seen as threatening by Borrower management, c) poverty-oriented lending, d) service delivery and maintenance activities and e) family planning and environmental projects. Other areas where weak commitment is predictable are usually identifiable from the Bank's experience or other projects' past performance in the particular country, sector, or institution. 33. Special attention should also be given to planning the policy dialogue in the case of projects with controversial policy reform elements, where commitment-building is particularly difficult. This will involve a) ensuring that the ESW program provides a relevant and timely technical input; b) planning the dissemination of ESW to all key actors, as feasible in each case; c) using group discussion to build commitment (Economic Development Institute seminars which bring together discussants from different countries can be a useful complement to direct Bank-Borrower dialogue); d) planning systematically which Bank staff at what level will raise which issue with which key Borrower actors; e) ensuring that there is staff continuity; £1 ensuring that staff of appropriate status are involved in discussion with the most senior civil servants and politicians, and g) planning the proportion of dialogue to be held at the political as opposed to the technical level. This last poses the dilemma for the Bank of how far to develop relations with politicians whose commitment to reform is critical, but whose tenure in office may be short-lived. 34. For managers, the broad lesson from Bank experience in building commitment through dialogue is that in the past attention has focussed on project content and conditionality, while what is needed in the future is greater attention to the process of learning and persuasion during joint ESW, project design and supervision with the Borrower. Use of the Bank's financial leverage, and of conditionality in loan agreements, should only be seen as a complement to rather than a substitute for this process. 35. Influencinp commitment through changes in borrower organizational processes and structures. Most Bank projects include institutional development activities which can be designed so as to influence Borrower commitment by increasing or decreasing the motivation and/or power of different actors. In particular, four complementary management development activities can powerfully affect commitment at the middle and lower levels of an implementing organization, assuming the existence of a committed top management. First, setting clear and attainable official goals increases control, and hence commitment, by reducing subordinates' scope for pursuing informal goals and giving superiors norms by which to judge subordinate performance. Second, developing monitoring systems increases accountability, and hence also commitment. The use of planning and feedback systems to influence motivation is discussed in more detail in Staff Working Paper 537, Bureaucratic Politics and Incentives in the Management of Rural Development, pp. 39-49. Third, developing supervision systems simultaneously increases both control and accountability. Fourth, commitment can also be increased by linking staff rewards to performance-- whether rewards are material, as in linking managers' bonuses with SOE performance, or non-material, for example the increase in prestige which the award of badges of merit has brought to successful field staff in the Indonesian population program. ~*nexV1 contains a case study of the training and visit system of agricultural extension, illustrating how all of the above approaches can work to increase commitment and effectiveness in a type of activity--increasing the responsiveness of bureaucracy to small farmers--where low commitment is the norm. 36. In projects where the commitment of external actors is important, at least three additional approaches can be used to increase commitment across organizational boundaries. First, influence over other organizations can be built through the use of a variety of coordinating mechanisms, varying from weak (ad hoc meetings) through regular planning meetings and the development of liaison positions, to strong (written working agreements, full-time committees, participation in matrix structures). These and other options are explored in greater detail in Staff Working Paper No. 375, The Design of Organizations for Rural Development Projects. Second, budgeting systems can be developed so as to increase leverage over other organizations, as in Colombia's rural development program, where the annual budget given to implementing agencies depends on their performance in the previous year (for details see World Development Report 1983, Box 9.9). Third, new organizations can be developed to influence existing institutional actors; for example, a state enterprise board with the power to set prices, or a tariff reform department in a prime minister's office which will provide a focus for policy reform. Organizing communities to have effective influence over service delivery agencies is also an important way of increasing bureaucratic commitment in people-oriented programs; a case study is given in Annex VII. 37. Each of the above strategies is a means of increasing the influence of a committed actor or actors (individual or institutional) over others less committed. The success of such approaches obviously therefore depends on the existence of a strategically placed, committed core group or individual. So the reform of organizational processes and structures, like the use of leverage and conditionality, is once again only a complement to and not a substitute for the process of education and persuasion needed to build this initial commitment among key, senior Borrower managers. Where this process is ineffective, new organizational systems and structures may be installed, but are likely to be no more than symbolic responses to Bank influence; an example is an East African country's National Commission for Population and Development, established under a SAL condition but for years staffed by the Borrower Government with personnel too junior to be an effective constituency for population policy development. 38 Using financial leverage. At the country level, the Bank may make the size of a lending program conditional, among other factors, on economic performance and hence, implicitly, on commitment to pursuing key economic policies. This use of the Bank's financial leverage is outside the scope of this paper. At the project level, the Bank has two main forms of financial leverage: Releasing funds in tranches according to performance, and cancelling or suspending disbursements. (a) Phasing implementation and tranching financial assistance. Large programs or projects with predictable rates of disbursement may be least well suited to maintaining Borrower commitment. The alternative--a series of smaller operations with future assistance dependent on results--can strongly influence Borrower performance for the better in many circumstances. Similarly, where individual programs are divisible over time, they might be designed so that release of a second tranche of funds will take place only after a mid-term review reveals satisfactory progress or leads to agreement between Borrower and Bank for redesign of the subsequent phase. These are already current practices with many traditional project operations. Tranching is of course an integral feature of SALs and other policy-based lending operations. However, tranched, fast-disbursing loans may not on their own be the most suitable vehicle for building commitment to institutional reforms (e.g., in budgeting processes), which have long time horizons. In order to permit a sustained focus on institutional development efforts, it may be effective to pursue institutional goals through long term technical assistance projects, progress with which could be a factor in deciding the release of tranches from a separate series of policy reEorm operations. (b) Cancelling projects or suspending disbursements. The decision to cancel a program or project during implementation is always difficult, when costs have been sunk, and the only means of maintaining some control over Bank funds is continued dialogue with the Borrower institution. It may be even more difficult to stop work on a project during preparation on the grounds of anticipated lack of commitment, since this may be harder to demonstrate than, for example, technical inadequacies on the part of the Borrower. Nevertheless, where there is a serious threat to developmental objectives fundamental to the program, the Bank should not proceed. Adoption of a more systematic approach to the assessment of commitment may be beneficial in such cases, both for determining when such a threat exists, and in giving Bank staff a more soundly based justification for their recommendation not to go ahead. The decision not to proceed with a project on the grounds of inadequate Borrower commitment should not automatically be interpreted as a failure on the part of the Bank. In certain circumstances, withdrawing support from a Borrower institution may be the most positive means of securing needed reforms, as well as freeing Bank resources for more profitable investment elsewhere in the short term. E. Follow-Up 39. In spite of its importance and pervasiveness, the subject of commitment has seen little systematic treatment or research; and practical guidelines for Bank staff in assessing and influencing commitment are unavailable. Bank management could encourage more systematic treatment of commitment issues by requiring that project officers summarize in project briefs, ESW proposals, staff appraisal reports, supervision reports, back-to-office reports, and project completion reports, their assessments of Borrower commitment and the steps they will take or have taken to build and maintain it. With regard to research, more could be done to review Bank experience to evaluate cases of high and low commitment and their causes; the relationship between Bank staff continuity and quality of assessment of commitment; the relative success of different approaches followed to assess and enhance commitment, etc. With regard to operational guidelines, Project Policy Department could produce a brief "how-to" manual, summarizing the steps and strategies proposed in this paper, but focussing more on the role of different programs and project staff, at different stages in the ESW/project cycle, and for different types of lending operation. Annex I - 15 - ASSESSING INFORMAL VERSUS FORMAL COMMITMENT: A SOUTH ASIAN RESETTLEMENT SCHEME Evidence from Performance The Regional Development Board was responsible for managing the integrated development of a major foreign donor-assisted irrigation/resettlement scheme in a South Asian country. The program formed a major plank of the newly-elected Government's economic strategy. The high level of commitment to the program at the political level could be explained by its potential benefits in terms of employment creation; independence from food imports; and a symbolic return to the power and prosperity of an ancient kingdom, the ruins of whose irrigation systems lay beneath the jungles which were to be felled, settled and cultivated once more under the program. Yet field visits to the area under development uncovered a range of implementation problems which seemed inconsistent with the priority attached to the scheme, and the capacity of the country's relatively strong development institutions. Agricultural credit default rates were extremely high; farm equipment had not arrived; farmers were late in beginning cultivation operations at the start of the wet season; essential inputs such as fertilizer arrived late or in insufficient quantity; farmers were growing flooded paddy rice on porous soil instead of the subsidiary food crops recommended in the project plan; some farmers harvesting late had no time to thresh their crop before the land preparation period of the subsequent season; irrigation water use was double the per acre allocation in the project plan, and hence water supply for the tail end of the Scheme still to be irrigated and settled was threatened; finally construction work at the tail end was haphazardly planned, with priority often going to activities off the critical path. Further analysis suggested that most of these problems were linked in vicious circles. Farmers took longer than planned in land preparation because tractors to be provided under the project had not been procured and distributed. This meant that farmers kept paddy fields flooded for tillage after the peak period of rains, and hence used more irrigation water. Non-arrival of threshers under the aid loan meant that crops were often stacked after harvest until buffalo or tractor power became available for threshing. This increased grain wastage, and delayed the receipt of cash from the marketed surplus. The result was often default on agricultural loans, with the prospect that the use of modern inputs in the following season would be reduced. Delay in harvesting and threshing also meant a late start on land preparation in the next season, missing the peak period rains and thus continuing the vicious circle. Thus, the agricultural problems could be traced to the absence of much of the government support designed into the early period of implementation. The Development Board was responsible for procuring tractors and threshers under the aid loan, and had not done so. Annual inputs were arriving late or not at all, again the Board's responsibility. Planned marketing services for the subsidiary food crops had not been organized, so that farmers had every incentive to grow their traditional paddy crop on the porous soils, and hence vastly increase water consumption Annex I in general. The evidence from performance in the field suggested that the Board was making little progress in agricultural development compared to irrigation system construction and jungle clearance. The question was whether this was due to technical and managerial deficiencies, or to low commitment to the neglected activities. The Key Actors and their Motives The Chairman and Chief Executive of the Regional Development Board (RDB) had been appointed by the Cabinet on his promise that he would rapidly speed up implementation of the resettlement scheme. He was therefore the chief formal actor in the organization. The Chairman was a mid-career engineer, young for the responsibility of his new post, and ambitious. His management style was personalist rather than bureaucratic, relying on verbal instructions rather than the written word. No significant decisions with regard to plans and priorities were delegated. The Chairman surrounded himself with a loyal and effective team of engineers in the head office, which clearly formed the key professional interest group. The Chairman controlled the flow of information to the RDB's governing board, and also to the Cabinet, since briefings on project progress were made personally by the Chairman to senior politicians. A monitoring and evaluation unit in the RDB made regular reports of project progress, but only to the Chairman and not outside the organization. Given the Chairman's power, his motives were likely to be the key determinant of incentives and performance in the RDB. The Chairman's political future depended on making visible progress towards targets which outside observers agreed were unattainable. His personal incentive was therefore to concentrate on jungle clearance and canal construction activity, since the most visible measure of progress was the number of acres delivered with irrigation infrastructure and ready for settlement; progress in agricultural development was neither so visible nor so attractive a priority in terms of the Chairman's professional background. This was therefore the key divergence in incentive terms. between the formal and informal goals of the project. The Chairman's actions in terms of staff appointments in head office provided corroborative evidence of this divergence; all engineering posts were filled, as was that of the manager in charge of settlement (the arrival of settlers being a politically sensitive measure of progress), but the chief managerial post for agriculture was left vacant for over a year. Given the Chairman's motivation, what appeared haphazard planning in the field was in fact rational pursuit of informal goals. Thus, RDB concentrated on building irrigation infrastructure rather than coop stores essential for community development purposes and for the provision of agricultural inputs. Within the activity of irrigation development, the concentration was on jungle clearance and major canal construction rather than minor canals and field channels. RDB's reports of progress in the field consistently highlighted construction activity, and downplayed agriculture and community development. The Chairman's commitment to informal goals influenced the behavior of staff down the line. Acceleration of the project meant that implementing officials were overworked in all sectors and therefore that Annex I they could not attend to all the elements of their formal job descriptions. Given the pressure to obtain construction equipment, it was rational for overworked procurement officers in head office to concentrate on procuring bulldozers rather than agricultural equipment. Similarly, it was in the overworked field project manager's interest to take the line of least resistance in concentrating on construction rather than agriculture and community development work. Agriculture staff in the field, who were highly motivated professionally, had little alternative but to acquiesce to informal goals since they had no support from head office and hence neither incentive nor ability to carry out their work according to the project plan. Coop officials responsible for procuring agricultural inputs could not plan requirements in advance, since they could not predict next season's cultivated acreage on the basis of the inflated targets for expansion of the irrigated area given them by the engineers. Water bailiffs attempting to reduce water issues faced political hostility (including telegrams to the local MP) from top-end farmers who stood to lose, and had no incentive to enforce restrictions, given the lack of support from top management. I Conclusions While there was commitment at the political level to the official goals of the project--generation of employment and of an agricultural surplus--the key actor in the implementing agency was in fact committed to significantly different informal goals. These could be pursued successfully--at least in the short term--because the key actor's control over organizational agendas and over information going to overseeing agencies made him imperfectly accountable. The key actor's commitment powerfully influenced the behavior of those below him. While subordinate l staff were not consciously aware of the key actor's informal goals, in a situation of overload they rationally responded to the pressures from above l to accelerate work on certain activities, and had little time to pursue other activities for which they would anyway receive little reward or recognition. Foreign sector specialists evaluating performance in the field in agriculture and community development attributed the problems to managerial and technical deficiencies. In fact, poor field performance could be traced not to lack of capacity at the local level, but to lack of commitment at the head office level to the project goals agreed between Borrower and donors. Annex I1 FOCUS GROUP INTERVIEWING: ASSESSING THE COMMITMENT OF HEALTH SERVICE PROVIDERS IN A WEST AFRICAN COUNTRY A. The Focus Group Technique The design of the focus group technique is to bring together a small group of participants (usually six to twelve) for an informal group discussion, led by a moderator working from a prepared discussion guide. Participants should feel comfortable and secure and at ease to express their views frankly, even when they differ from prevailing norms. The session works best when participants respond one to another in a dynamic that stimulates the discussion to deeper probing of issues than would be possible in a formalized interview. Experience shows this to be most feasible when the participants are more or less of equal status and have similar perspectives on the subject under discussion. It requires skill to lead the discussions properly--to dispel inhibitions and draw out the participants without letting anyone dominate, and to be flexible and let the discussion flow freely while regulating it so that important topics are covered adequately. Someone is assigned to take notes, and the session can be recorded on tape as well. The basic data to analyze consist of the statements recorded in the session. Unlike the sample survey, this approach yields qualitative information that is inappropriate for statistical generalization. Rather, the purpose is to develop insights into the structure of social relations, of norms and values and of individual attitudes and behavior. Unlike the participant-observer approach of ethnographic research which yields similar insights but requires long immersion in a specific community, the focus group approach produces results quickly and each study can encompass a multiplicity of settings. In market research, this technique has been applied to test consumer reactions to new products. It has only recently been introduced into social science, with particular application to demographic research on fertility and family planning. It should be noted that the conclusions which are drawn from the group discussions depend entirely on the analyst's judgment and interpretation. Therefore, to control bias, the use of multiple analysts is recommended. B. Case Study: Explaining the Under-utilization of Health Services in a West African Country Ten focus group sessions were held--seven with rural groups (six of health care consumers and one of traditional medical practitioners) and three with urban groups (two of consumers and one of orthodox health care providers). Sites were selected to provide some representation of the major sub-cultural zones of the state. The groups consisted of: village women (3 groups); village men (2 groups); elementary school teachers (male and female); traditional medical practitioners (male and female); male civil servants of junior grades; female civil servants of junior grades; and nurses and midwives at a General Hospital. Discussion Topics With the groups of health care consumers, the discussion ranged over the following topics: General knowledge, attitudes and practices regarding childbearing and its impact on the health of mothers and children; basic health values; prenatal, postnatal and preventive care; perception of infant and childhood morbidity and mortality and of their impact on fertility; steps taken when illness occurred; cost of health care; access to government health facilities and perception of utilization of those services; and problems encountered with government health facilities, and recommendations for government action. With providers of health care, the discussions centered on: Perceptions of recent trends in the utilization of government health services; explanations for these trends, for example, problems that providers knew patients had encountered, or that providers had themselves observed in the course of working inside the public sector health care system; and providers' sense of the public perception of government health services. For health care providers in private practice (traditional medical practitioners), the balance of the session dealt with six questions relating to what diseases clients came to see them about; whether clients previously gone to government health facilities, or came directly to them; how far clients travelled; what fees were charged, and how flexible they were concerning payment; recent trends in their client flow, for example, related to general economic conditions, or to fee increases at government facilities; and the problems their clients reported having with the government health services. Findings: Reasons for Under-utilization of Government Facilities a) Consumer'S Views Members of the public across the state were acutely aware of several problems inhibiting utilization of existing government health facilities. All the different respondents in this survey identified a remarkably similar set of problems, which may be grouped under seven headings. i. Limited Access Rural residents especially had been disadvantaged by the concentration of public sector health facilities in and around urban centers. ii. Prohibitive Costs The cost of health care at public sector facilities had become prohibitive for many, especially in view of the unsatisfactory service they report. In addition to travel expenses, all respondents reported that going to government health institutions always entailed considerable costs in time--waiting in line to register, waiting to be seen by a doctor, waiting for laboratory tests, to fill prescriptions at the pharmacy, etc. Annex I1 iii. Lack of Supplies and Equipment Chronic shortages of drugs at government health institutions are so notorious that the expression "O/S" (out of stock) has become standard in everyday speech. Many participants reported having had to go out and buy prescribed drugs from private pharmacists and then bringing them back to be administered to patients admitted into general hospitals. iv. Attitudes of Health Care Providers There was a widespread perception of uncaring attitudes on the part of health service personnel. A remarkable amount of time in each focus group session with health care consumers was taken up by complaints about the attitude and demeanor of nurses in particular. They seemed to be universally perceived as harsh, rude and uncaring, even to maternity patients. Doctors also came in for similar criticism, and for making hasty prescriptions without listening appropriately to the patients. v. Nepotism Nepotism was a widespread problem. There were frequent complaints about relatives and friends being allowed to skip registration lines at government health facilities, and of previously "O/SW drugs suddenly becoming available at the hospital pharmacy when such personal connections arrived. vi. Diversion of Supplies and Services Many respondents reported that they had purchased drugs from the open stalls in the market place, with the government stamp on the packages. Such drugs had obviously been diverted from the government health facilities. It also came out in every group session that doctors frequently referred patients to their private clinics where they would give them the solicitous care not available at the public facility. vii. Unofficial "Surcharges" It came out in every focus group session that members of the public perceived an atmosphere of corruption at government health facilities, whereby the way to assure adequate care was to arrive at a personal arrangement with the attending physician. For example, a pregnant woman would pay a "surcharge" during her antenatal period as "insurance" for adequate care during labor. Surgeons were especially notorious for corrupt practices. b) Provider's Views Private Sector Providers In sharp contrast to the problems of the state health institutions, a private dispenser pointed out that his thriving practice owed much to the fact that he had low fees with installment payment options, drugs were always in stock, patients did not have to wait long before he saw them and his attitude and manner of approach re-assured them Annex I1 - 21 - so that they would recommend his dispensary to other people. His patients frequently passed by several other facilities, both in the private and public sectors, in order to come to his dispensary. Similarly, the rural midwife in private practice reported charging low fees with installment . payment options, and offering solicitous service that gratified her patients ii) Public Sector Providers From their perspective as insiders, the government health workers interviewed substantially confirmed public perceptions of the problems of the state health service. First, all agreed that fee increases had hurt. But on the other hand, nurses and midwives reported that even before recent fee increases, patients had been leaving the general hospitals. All government health workers also recognized the shortage of drugs and hospital supplies ("01s syndrome") and the lack of equipment as serious problems. Concerning the attitudes of health care providers, the nurses and midwives conceded that nurses, attendants and ward orderlies used to be seen as rude but maintained that they had begun to change since the Commissioner for Health assembled them to discuss these problems. Evidence of professional conflict in the hospital between doctors and nurses also emerged, with each side stating a case for controls on the autonomy or authority of the other. On the question of private practice and diversion of patients from the general hospitals, the resident doctor confirmed that several government doctors also had private practices. As for informal surcharges, the nurses and midwives reported that when they asked patients for the formally stipulated fees, often the patients would protest that they had already paid the doctors. Conclusions Questionnaire surveys might have elicited information about the problems of access to care, high cost of services and lack of supplies and equipment in this case. But they would not have elicited information about health provider attitudes, nepotism, diversion of services and supplies and unofficial surcharges for care. A project design which was targeted at resolving the problems uncovered by a questionnaire survey might have focused on construction of additional health facilities and supply of drugs and equipment--and would almost certainly have had little impact on community health. The value of the focus group method of interviewing was that it uncovered critical information, particularly from service users, about the commitment of health care providers, which in this case was a more important determinant of program performance than technical or financial constraints. Annex I11 USING WORKSHOPS TO BUILD COMMITMENT: POLICY REFORM IN THE ELECTRIC POWER SECTOR OF A LATIN AMERICAN COUNTRY Background For more than 20 years, the country's electric power sector has been dominated by a number of regional generating and distribution companies, which have resisted integration into a national grid. While the national interest would be served by drawing electricity from the least-cost distributor at any given time, the regional companies have been interested in expanding generating capacity in their own areas whether at high or low cost. A second problem is the existence of many smaller distributors which primarily serve remote areas at high cost, and who would need to be merged into larger companies to become financially viable. The sector was also facing major financial problems, due to a mixture of overestimation of demand, underestimation of project costs, low average tariffs, and a heavy debt burden exacerbated by a falling exchange rate. The Bank, for its part, has committed more than $1.5 billion in over 20 operations to the power sector in the country, and felt by the early 1980s that further investment in individual utilities could not be justified until the sector as a whole had been rationalized. From the start, project staff realized that the key issue was how to generate commitment to reform from the powerful regional companies with their differing priorities and interests. Planning the Dialogue The process of building commitment to reform began in Washington with a one-day workshop in January 1985 which brought together all Bank staff with experience in the power sector in the country. The workshop was highly structured (see attachment l), focusing first on diagnosing and prioritizing the problems; then on assessing the influence of the key actors in the country; deciding on the policy changes required; and designing the process to be used to influence Borrower officials to do what needed to be done. The workshop concluded that the country's best institutional response was to create an energy board with planning, regulatory, pricing and efficiency monitoring powers. The grcup identified two alternative strategies for establishing this board: (a) Enactment of legislation to create the new Board, with a sector policy loan conditional on the outcome (this approach would involve the Bank convincing the national Finance, Energy and Planning ministries, which would then be responsible for convincing the regional companies); and (b) directly seeking consensus among the different regions in the country as to what should be done. The group chose the second strategy, which they felt was more likely to build real as opposed to merely formal commitment to rationalization. The Dialogue in the Country The dialogue in the country centered on a workshop, designed by the Bank, whose purpose was to bring together the key actors in the sector to diagnose their own problems and then reach agreement on their own Annex 111 - 23- objectives for the sector and strategies for solving them. The three-day workshop involved about 60 participants, including six current or former ministers, two congressmen, many current and former heads of utilities and oil and coal companies, key Government officials, three IDB representatives, and six Bank representatives, of whom two were consultants hired as facilitators of the workshop. The workshop was carefully structured to raise the level of discussion away from the day-to-day conflicts in the sector. It began with a session at which participants were asked to put aside the existing structure of the sector and to think through what it should ideally look like. The next phase of the workshop focused on the current situation and its problems. More than 70 problems in the sector were identified, and then prioritized into six key problems, each of which in turn were seen to have at their root the strong power and differing interests of the regional companies. It was recognized that no individual distributor felt responsible for achieving the sector financial self-sufficiency which was the goal idealized by participants at the national level. The final phase of the workshop focused on developing strategies which could move the sector closer to the agreed objectives. The workshop ended with a number of options which participants were willing to explore further, including the creation of a sector-level "Rector", composed of the Ministry of Finance, the National Planning Department, the Office of the President and other organizations, which would be empowered to set tariffs and take other key decisions on the sector. Assessment No final agreement on a particular course of action was reached at the workshop, which was not a decision-making body. But while the ultimate success of the policy reform process remains to be seen, soon after the workshop a decree for the establishment of an energy board was submitted to the President for signature, and various detailed proposals for reform were being pursued. The climate in the sector seems to have changed. The key actors have moved from a situation of inaction in the face of the issues to one where there is a commitment to formulate detailed proposals for change. It is the process adopted by the Bank, rather than the Bank's technical advice, which facilitated this change. The Bank had provided a methodology for problem-solving, leaving country officials to do their own diagnostic work and thereby learning more fully about their own problems. Annex I11 WORKSHOP ON POWER SECTOR Introduction What is Expected i n Workshop Image Exercise - What is Your Image of the Power Sector i n the Country Problem Identification - Identify a l l the problems i n the sector (20 minutes) - Group the problems, p r i o r i t i z e them and identify t h e i r causes (40 minutes) Environment Analysis - Identify a l l the key actors and t h e i r relative power t o impact the key problems. Discuss each problem separately. Policies i n the Power Sector ( a ) What should be the country's policy? (b) What should be the Bank's policy? (c) How are the problems identified i n the morning session affecting the achievement of the objectives? Strategies f o r Achieving Agreed Policies ( a ) Which strategy can best achieve the policies agreed upon? (b) What assumptions are being used? (c) Which role should the key actors play, what authority do they have? Design of the Implementation Program ( a ) How do we get the Borrower o f f i c i a l s t o do what needs t o be done? (b) What in£luence strategies can we use? (c) Summary - Next Steps Annex IV USING WORKSHOPS TO BUILD COMMITMENT: AGRICULTURAL INVESTMENT PLANNING IN A CENTRAL AFRICAN COUNTRY Sectoral investment policies and plans are often decided with little consultation by a few senior staff in the sectoral ministry and the national planning agency. As a result, investment projects often fail to respond to priority problems and needs at lower levels, and, in aggregate, overstretch line ministries' implementation capacity. Under these circumstances, lower level staff are unlikely to be committed to plan implementation. In early 1985, Eastern Africa Region set up two workshops for middle and senior level executives involved in agricultural development in a central African country as a way to diagnose problems and develop realistic proposals for investment studies through participative methods which would maximise commitment to later implementation. The first workshop brought together about 40 middle-level executives from the Ministries of Agriculture and Forestry, Animal Industry and Fisheries, and Cooperatives and Marketing; from the Central Bank; from agricultural parastatals, including financial institutions; some farmers; two Bank-financed workshop designers/facilitators; and three Bank staff as resource persons. The participants represented the different regions of the country, and were also drawn from different levels of the hierarchy; district, regional, and central. Discussion during the 5 112-day workshop was organized around three questions. First, participants were asked to identify those functions which their ministries were (a) currently performing, (b) currently not performing but ought to perform, I (c) currently performing but ought not to perform and (d) organizations which ought to be responsible for the functions identified under (c). Second, they were asked to prioritize the functions that were or should be performed by their respective organizations, according to whether they fell among the top five priorities, the second top five priorities, or below. Third, they were asked to prepare problem statements for the top ten priority functions. The workshop produced interesting results. First, the problem statements focussed almost universally on problems of process, rather than on the need for additional capital investment, reflecting the perspective of implementors as compared to that of many senior head office managers and foreign donors. Problems identified centered on financial management (delayed disbursement from headquarters and inadequate expenditure auditing); personnel management (inadequate logistical support, low pay and allowances); and policy formulation (inadequate consultation, lack of information, and inadequate farmer incentives). Second, after reviewing the range of implementation problems, participants revised their views on the distribution of functions away from their initial tendency to support expansion of the government sector; a more realistic picture of ministries' implementation capacity had emerged. Third, all participants in the workshops volunteered to take part in follow-up studies to identify solutions to the problems identified. The second workshop brought together about 45 senior executives from the same organizations, plus the same expatriate participants, again for 5-112 days. The workshop began with a presentation of the middle Annex IV executive workshop findings by the Chairman of that workshop. The participants of the second workshop then spontaneously and unanimously elected this spokesman as Chairman of their own workshop. Discussion in the senior workshop was structured around a 207-item questionnaire, prepared by the workshop designers, and divided into sections dealing, for the key sub-sectors, with (i) priority functions, (ii) current operational problems, (iii) priority study needs, (iv) transfer of functions and (v) other issues. The questionnaire was compiled from a number of sources, including problems and issues raised at the first workshop, and issues presented in a recent Bank agricultural sector memorandum. Problems and issues were posed in the form of statements with which participants could indicate levels of agreement or disagreement. Participants were divided into six working groups and first discussed the entire questionnaire in their groups with emphasis on the differences of opinion among group members. They were explicitly asked not to achieve consensus at this initial stage of the discussions. After three hours within their small groups, each group designated a rapporteur to present to the plenary session a summary of their group's discussions about the tradeoffs concerning the two or three most controversial statements in the questionnaire from their own point of view. All participants then completed the questionnaire on an individual, anonymous basis (only ministry affiliations were indicated on the questionnaires). The results of this process were then rapidly tabulated, and used as an input into further small group discussions, during which the participants were asked to identify the top ten priority needs for pre-investment or policy studies. The final three days of the workshop were devoted to detailed planning of the actions required to follow up on the priorities identified. The workshop discussions led to a number of substantive conclusions. The senior executives reached strong agreement on most of the process-oriented priorities identified at the first workshop; in fact, of the priorities identified, no less than half were in the area of managerial change--structural, procedural or financial--of interest to implementors. This emphasis contrasts with typical investment programming discussions focussing on capital investment projects and technical ventures. From the Bank's point of view, the workshop process had proved valuable as a means of learning about the priorities of different Borrower interest groups; of confirming that many of the issues raised in the agricultural sector report were priority concerns; and of building a shared commitment between policy-makers and implementors to acting on many of the report's process-oriented recommendations. With regard to the latter the workshop discussions had both increased policy-makers' commitment to solving implementors' problems, and increased implementors' commitment to carrying through projects which they had agreed to be relevant and feasible. The workshops also provided the Bank with valuable warning of potentially variable commitment to some courses of action, through surfacing strong divisions of opinion on some issues--merging the Ministry of Agriculture and Forestry with the Ministry of Animal Industry and Fisheries, for example. Annex V - 27- USING WORKSHOPS TO BUILD COMMITMENT: PROJECT LAUNCH WORKSHOPS Project Launch Workshops, pioneered by the Bank's Urban Projects staff, bring together participants from all major and minor agencies and groups concerned with project implementation to build understanding of and commitment to the goals of the project. The workshops serve more than an information and planning function. A key feature is the public commitment made by each agency to fulfilling particular roles and responsibilities, for which agencies can later be held accountable. Similar workshops can usefully be repeated, to review progress and to renew commitment, regularly throughout the implementation period. The Workshop Process The workshop proper can usefully be preceded by a "senior management seminar" to familiarize top managers of the key implementing agencies with the aims of the project, since it is unlikely that all key actors (including Bank staff) were in post when the project cycle started. A senior seminar of about half a day can be followed by an opening plenary workshop session of representatives from all participating agencies. The opening session should, if possible, be chaired by a key political figure and should be the occasion to formally announce approval of the project and the start of implementation. Press and TV coverage is desirable, as this gives politicians a stake in the success of the project, as well as publicizing the project itself. Public association with the project is a critical step in solidifying commitment--the elected officials are on record. Working group sessions following the plenary opening sessions can be held for the various subproject component managers and their staff. Topics such as procurement, auditing and accounting, monitoring and evaluation and project control techniques can be presented and discussed. These joint sessions build commitment to cooperation, as well as understanding of the project concept. The drawing up of critical path diagrams and bar charts is a useful tool in structuring these working group sessions and can later serve as a focus for coordination and for solving implementation problems. The Project Launch Workshop should end with a final plenary session and a structured discussion of the conclusions and recommendations of the working group sessions. This wrap-up session would be the end of official project launch activities and would serve as "time zero" for implementation monitoring purposes. The best time to hold a Project Launch Workshop depends on the requirements of the project. If many conditions must be met before the project loan becomes effective then the workshop might be delayed until these conditions are met. A premature Project Launch Workshop is useless if consultants are not in place, if key management posts are unfilled, or if vital legislation is not passed. Annex V1 which often offer opportunities for personal profit. Thus, T&V offers no material rewards beyond the extension worker's regular government salary. In many projects, strong high level commitment from senior officials and politicians has been needed to overcome vested interests in traditional multipurpose extension systems and convert them to the single task approach. Once the system is in place, however, extension workers find substantial psychological rewards in the recognition they receive from farmers, and the prestige they gain in local society, from their role in introducing innovation and raising incomes. Conclusion The managerial structure and processes of the T&V system are designed to increase extension worker's commitment to providing a service to farmers. First, they ensure that extension workers, supervisors and researchers all spend substantial time with farmers. Second, they ensure that farmers' reactions as to the usefulness of extension advice are the main criterion in judging field worker performance. Adoption of a service orientation thus becomes in the rational self-interest of extension agents. But for the T&V system to work, understanding of and commitment to the system's principles is essential on the part of the key senior extension managers. The Bank's experience is that building this understanding and commitment involves more than a one-time dialogue. The rapid turnover of senior extension managers in many Borrower countries means that continuing orientation in the T&V philosophy by Bank staff has to be carried out throughout the implementation period. - l Annex V11 ORGANIZING COMMUNITIES TO INCREASE THE COMMITMENT OF BUREAUCRACIES TO SERVE THEM: COMMUNAL IRRIGATION IN THE PHILIPPINES In 1976, the National Irrigation Administration (NIA) in the Philippines began a reorientation of its small-scale irrigation development program, with the aim of giving farmers more control over planning and implementation, and of instilling into the bureaucracy a greater commitment to service. The following extract is from David Korten's report on the process in the Public Administration Review:l/ - Government assistance to small gravity irrigation systems in the Philippines...during the 1950s and 60s was dominated by "pork barrel" politics which spread available funds over so many different projects that planning and construction were often inadequate. In the early 1970s efforts were made to correct the deficiencies of the past, but even with more rational allocation of funds many completed systems fell ... [In rapidly into disuse or served substantially fewer farmers than intended. response, the NIA] initiateId1 a pilot project at Laur in Central Luzon to experiment with a more integrated approach in which the capacity of the water users association would be developed through active involvement in the planning and construction activities: planning system layout, obtaining water rights and rights of way, organizing volunteer labor inputs to system construction, and exerting control over project expenditures. Integrating social and technical development proved extremely difficult. In one community it was learned how difficult dealing with local power struggles can be--leading to the abandonment of construction plans until the local association reorganized itself some two years later. In a second community it was learned that a high level of commitment from a cohesive farmer group does not necessarily make things easier for the engineers: scheduling and system design issues resulted in numerous delays and changes; organization of volunteer labor presented unfamiliar problems worked out only through lengthy meetings; and farmer insistence on monitoring purchases and limiting personal use of vehicles using gasoline charged to the farmers' loan accounts was not always welcomed by project engineers. The farmers even questioned the engineers on basic technical judgements, such as the type of material chosen for dam construction, -l/Community Organization and Rural Development: A Learning Process Approach, PAR September/October 1980. - Annex V11 32 - insisting that the proposed structure would not withstand the force of local floods. Finally, however, the new dam was completed using the design favored by NIA's design engineers--only to be to be washed out a few months later. The experience was sobering in the difficulties which it suggested the NIA must face if it were to work effectively in support of community managed irrigation; its capabilities on both the technical and the institutional side would need to be upgraded and integrated. Numerous changes in operating procedures were implied. Yet, it established in the mlnds of NIAfs leadership that there were major benefits to be gained in return. Not only could farmer participation in system planning and construction result in a stronger water user association better equipped to operate and maintain the finished system, but it could also result in a better designed and constructed irrigation system more likely to meet farmer needs. The result was a strengthened commitment by one of the largest public agencies in the Philippines (43,000 employees) to build a new capacity for community level action. NIAfs approach to development resulted in a bureaucracy committed to a service orientation, and in project designs suited to local needs. Many also argue that the "empowerment" of communities involved in this approach is also a development benefit in itself. But NIAfs experience also indicates two cautions. First, organizing local communities is difficult, and staff-intensive, both for the Borrower and the Bank. Bank projects adopting this approach are likely to require a heavy social science input (as did the NIA program) during both design and implementation, to collect social data, test working methods, and provide feedback. Second, the process of reorienting a bureaucracy to a service ethic takes time and as a precondition requires commitment to change at a high level in the bureaucracy itself. Bank involvement in such approaches will require careful choice of implementing institutions based on assessment of senior managersf commitment to change, and in addition persistent commitment on the part of the Bank to sustain its support for a development process which may require longer than a single project cycle to take root. The World Bank Headquarters European Off~ce Tokyo Office 1818 H Street, N.W. 60. ,~vcnucd'li-113 Koku\ai I J t ~ i l d i n ~ Washit~gton.l>.(:. 20433, U.S.A. 751 10 I'aris, F:r:~ncc 1-1 Marunouchi 3-chomc Tclephot~c:(202) 477-1234 Tclepl~ot~c:(1) 47.23.54.21 <:l~iyocl~~-ku,'l'okyo 100,J ~ p a n Tclcx: WUI 04135 WOI