Archive

Agency for International Development

Recommendations and Actions

AID06: Reengineer Management of AID Projects and Programs

Background

The President's fiscal year 1994 international affairs budget totals about $21.9 billion, of which $6.8 billion is requested for the Agency for International Development (AID).(1) Much of AID's budget is spent on programs and projects that range in complexity from a single training workshop to developing financial systems in a foreign country's government. AID funds approximately 2,000 projects annually in more than 100 countries.

Through the years, AID has shifted its focus from the funding of discrete projects to an emphasis on comprehensive programs and increased use of what AID calls non-project assistance, which promotes broad-based economic and policy reforms in developing countries. This portion of the report will focus on what AID traditionally has called the project management system. However, it also has wide application to AID's programs and non-project activities.

Although AID headquarters in Washington, D.C., plays an active role in project management, AID's field offices or missions have primary responsibility. A mission is based in the foreign country that receives AID assistance. AID refers to these foreign countries as host countries. Generally, an AID mission funds a development project to respond to a specific problem within the host country, although funds sometimes are earmarked by Congress for a particular country, sector, or activity.

Since a major portion of the U.S. foreign assistance budget is specifically targeted to the goal of alleviating poverty in developing countries, the participation of the poor in these countries in AID projects is critical for project success. A successful project is defined as a sustainable one that not only achieves its objectives, but also is continued by the intended beneficiary after AID funding terminates. Therefore, project sustainability is largely contingent on beneficiary support.

Congress also recognized the important role that project beneficiaries play in the development process. In 1966, Congress mandated through Title IX of the Foreign Assistance Act that priority "be placed on assuring maximum participation in the task of economic development by the people of the developing countries."(2) The Foreign Assistance Act of 1973 advocated "self-help efforts . . . and the involvement of the people in the development process through democratic participation."(3) In 1975, subsequent legislation provided that "greatest emphasis . . . be placed on countries and activities that effectively involve the poor in development."(4)

Although AID understands the linkage between customer involvement and project sustainability, the current project management system can prevent AID from fulfilling its congressional mandate in this area.(5) As a result, many AID projects are developed that either do not adequately serve AID's customers or lack the necessary customer commitment. Therefore, such projects may not continue once AID funding terminates. Based on an extensive review of the literature and interviews with AID staff and other development specialists, several problems in the way that AID's project management process is structured have been identified. These problems seriously constrain the ability of AID to respond to the needs of the poor in developing countries.

Recent reports have highlighted numerous problems with AID structures and procedures that hinder the agency's ability to achieve results. The project management cycle generally involves seven phases: problem identification, project design and negotiation, obligation of funds, implementation, evaluation, project redesign, and closure. Each phase requires an array of internal reviews, streams of paperwork, and complicated procedures, resulting in a rigidly crafted blueprint design of the development project.

The detailed blueprint of the project and the requirements of the system do not allow for the necessary flexibility and may inhibit the beneficiary input required for project sustainability. A 1988 internal evaluation of 62 AID health projects indicated that more than half either failed before the project ended or were unlikely to be continued after U.S. funds ceased.(6)

This high failure rate has been attributed to AID's top-down, blueprint project design and management approach, which impedes the ability of project field staff to make the necessary mid-course corrections during project implementation.(7) It also has fostered an environment that discourages risk-taking and the use of creative problem-solving strategies that may deviate from the project design.

Although changes may be warranted to respond to the uncertainties inherent in all phases of the project cycle, the standardized, rigid structure of the project cycle specified in AID's operational project handbooks does not allow sufficient flexibility. AID has streamlined some of its project requirements to allow for greater flexibility in certain regions of the world and plans to adapt these simplified procedures agencywide in early fiscal 1994.

The number of distinct steps necessary for project approval involving field, headquarters, and Congress results in a burdensome, duplicative, and time-consuming system. Frequently, a project can take up to three years before it begins. By the time a project gets off the ground, the assumptions built into the project design and the initial project participants may have changed. Continuity is lost. The implementation team attempts to carry out a project that was sometimes planned by another team three years earlier. Adding to the problem is AID's overseas contracting and procurement system, which often impedes project results by contributing to logistical delays, higher costs, and customer dissatisfaction.(8)

The incentives at AID tend to reward work in one part of the project management cycle (design and obligation of funds) at the expense of others (e.g., start-up, implementation, evaluation, results). People are rewarded for designing new projects, not necessarily for implementing them.

The General Accounting Office (GAO) recently suggested in a draft report that the push to obligate earmarked funds drives AID to fund million-dollar projects to which the host country is indifferent, as in the case of judicial reform initiatives in Central America.(9) In the area of implementation, more focus has been placed on audit trails, processes, and resource accounting than on how or if the project is achieving results. Evaluations focus greater attention on inputs (number of activities) than outputs (project impact).(10)

The mission director in each host country has ultimate responsibility and accountability for a specific AID project in his or her jurisdiction. Ensuring that the project works on a day-to-day basis, however, involves many people working on separate pieces. The AID project officer often is assigned to a project that was designed, budgeted, and authorized by others and is rarely present throughout the entire life cycle.

One reason for this is that AID projects are generally long-term, while Foreign Service personnel have shorter rotation assignments. It should be noted that this is addressed under a separate AID recommendation which allows for longer stays in host countries. The current structure fragments accountability. Project development involves the whole range of mission staff, from legal advisors to contracting officers, to ensure compliance with AID's rules and regulations. It appears that everyone is involved, yet nobody is really in charge to ensure overall project results. Noted management specialist Michael Hammer summarized the problem:

Classical business structures that specialize work and fragment processes are self-perpetuating because they stifle innovation and creativity in an organization . . . are unresponsive to large changes. . . . [I]nflexibility, unresponsiveness, the absence of customer focus, an obsession with activity rather than result, bureaucratic paralysis, lack of innovation, high overhead--these are the legacies of one hundred years of American industrial leadership.(11)

Compounding the problem, AID relies heavily on different outside contractors to handle separate phases of the same project. A project could be designed by one contractor, implemented by another, and redesigned by still another contractor. Contractors are evidently relied upon to provide skills not available within AID and to compensate for reduced staffing levels. Many staff do not have the training required for AID's new program directions of environment and democracy, including management of judicial reform projects in Latin America.(12)

As a result, AID sometimes pays contractors at U.S. rates for work that indigenous organizations might be able to do at the local rate for a fraction of the cost. Indigenous organizations are those that have been formed by the local population within the host country. Although U.S. contractors are needed in countries where technical expertise is lacking, there is a definite advantage in using the growing reservoir of host country professionals and indigenous organizations to assist with development projects. AID's existing policies, which create incentives that encourage U.S. contractors to form partnerships with indigenous organizations, are among the steps that AID has taken to help indigenous organizations strengthen their institutional capabilities.

Many procedures and incentives, however, preclude the full involvement of intended beneficiaries such as indigenous organizations in developing countries. Even though many AID staff expressed interest in working directly with indigenous organizations, they cited a number of requirements in AID's operational project handbooks that are confusing and onerous and act as barriers to beneficiary participation. Some of the requirements include audit, financial statements, personnel systems, and an existing track record. Many indigenous organizations may not have access to such basics as typewriters and accountants that are standard for U.S. organizations but are luxuries in developing countries.

Experience has shown that projects that support and do not supplant community efforts within developing countries are more likely to achieve results. Many innovative and successful models have been structured to build upon the efforts of people within developing countries. The following four projects have been included in this report to illustrate how their processes, incentives, and structures have been designed to foster an environment that focuses the organization on results and the customer. AID continues to rethink traditional approaches and processes to further its reinvention endeavors and has incorporated some of these innovations into existing projects. It should be noted that the following four projects represent only a few of the many innovations occurring within the development field.(13)

The Grameen Bank of Bangladesh. The Grameen Bank is an example of a development model in which credit has been used as an entry point for the poor and now serves as a catalyst for change.(14) It is not directly funded by AID, although AID has recently provided $2 million to the Grameen Trust, which provides services and technical assistance to other microenterprise programs around the world. Basically, the Grameen Bank provides small loans to poor people who own less than half an acre of cultivable land or assets not exceeding the value of an acre of land. Loan repayment has been excellent. Grameen has helped its borrowers accumulate capital and generate employment opportunities.

The Grameen Bank was developed using careful experimentation to determine what did and did not work. It continues to evolve using this method. Grameen recognized early that there was no prescribed formula for development. Had the Grameen been designed under a rigid, blueprint approach, it no doubt would have failed in the project design phase. It needed the experimentation of a social learning process in the implementation phase for innovation, risk-taking, and flexibility to adjust to local conditions. It continues to learn from its mistakes and to respond to feedback drawn from all levels of the organization. People are encouraged to bring up problems and are rewarded for developing creative solutions for solving them. It has been very effective at balancing the administrative demands of the organization with the participatory demands of the local community.

Grameen's success is also attributed to its incentive structure. Borrowers organize themselves into small groups. The members support each other and apply pressure when needed to repay loans. This pressure substitutes for collateral. Incentives are organized in a way that creates an interdependency between bank workers and bank beneficiaries. Within the group, mutual accountability ensures repayment and project sustainability.

Grameen has received worldwide attention for being a development project that works very well. It has been successfully replicated in other countries when it has been adapted to fit their unique cultural and economic conditions. It is important to note that there is no Grameen blueprint. The Grameen Bank has worked in Bangladesh only after many years of experimentation.

Canada's International Development Research Centre (IDRC). The IDRC's mission revolves around the belief that the problems of developing countries cannot be solved by importing outside solutions. The IDRC promotes development by providing funds to developing country researchers who are working within their own country on solutions to their nation's problems. It also helps in the dissemination of research results so that the research findings can be widely applied. This approach offers research opportunities that build up scientific capacity and thereby reduces the scientific brain drain in developing countries. It relies on a board of directors who are all from developing countries. The Canadian government has given the IDRC complete latitude to accomplish its mission by keeping it free from micromanagement restrictions and structuring it in a way that insulates it from Canadian politics. This project offers a variety of lessons for development organizations on how people within developing countries can be supported to solve their problems.

The Philippine National Irrigation Administration (NIA). Local participatory approaches to development are often viewed as not particularly relevant to organizations such as AID, which mainly work with host country governments at the ministry level and rarely with rural villagers. NIA, however, is an example of a large government agency in the developing world that implemented programs in ways that empowered local communities to take an active role in development activities. In fact, AID has promoted water users associations throughout Asia, based on NIA experience. AID has also worked with other donors to establish an international irrigation institute based on this model. It also provided funds in 1987 to NIA to support participatory approaches to irrigation systems.(15)

The NIA sponsored small-scale communal irrigation systems throughout the Philippines that involved local farmers from project initiation through implementation and evaluation phases. The NIA's program is one innovative approach to development that transformed itself from the conventional non-participatory approach where agency personnel viewed themselves as the implementors while the local farmers were considered passive beneficiaries, to a highly participatory one where farmers eventually took over the operations of the project and continued to sustain it through local fees.

Restructured incentives--such as partial recovery of collection fees and investment costs, amortization payments, clear points of responsibility for the results, and new performance measures based on financial viability where income exceeds expenses--encouraged greater participation and helped to curtail unnecessary spending, while promoting a sense of collective responsibility and ownership of the irrigation facilities.(16)

Since much of the AID funding is channeled through government agencies, this project offers a variety of relevant lessons about field approaches, organizational design, and management processes needed to transform the agency's role in developing countries from rowing to steering.(17) Although AID has incorporated many of these approaches in Asia, where AID sponsors irrigation and natural resource management, these methods could be expanded to other sectors and countries.

Inter-American Foundation (IAF). The IAF exists to support the efforts of indigenous organizations working on behalf of the poor in Latin America. Unlike AID, the IAF has no field presence. It accomplishes its mission through grants awarded directly to indigenous organizations. The IAF provides grants for projects that are initiated, designed, and implemented by indigenous people within host countries, and not by IAF staff. In-country support teams made up of indigenous people are responsible for monitoring and evaluating project performance. Although the IAF has been criticized for becoming too bogged down with regulations perceived as hampering innovation, its strength has traditionally been attributed to its unbureaucratic internal structure and decentralized decision making processes, which have helped the organization foster risk-taking and beneficiary participation and the freedom to experiment and learn from mistakes.

The importance of timely, accurate, and relevant information moving vertically and horizontally throughout an entire organization in all phases of the project process cannot be overestimated. One way that AID obtains feedback on its projects is through the evaluation process. It appears, however, that evaluation findings are not always acted upon or taken into account in the planning and budget allocation processes. Sometimes funding continues even after evaluations conclude that projects are producing no benefits and the host country has not demonstrated a commitment to the project.

Several external barriers prevent AID from withdrawing funds from unsuccessful projects. These include:

--- congressional earmarking of funds in statutes and related reports,

--- contract commitments and concerns about possible litigation,

--- the political difficulty of reallocating committed funds,

--- rigid and complicated planning systems, and

--- consideration of the views and commitments of contractors and grantees such as U.S. non-governmental organizations (NGOs) and U.S. private voluntary organizations (PVOs).

AID's 1992 In-Country Presence Report concludes that because the programs are broad and staff is increasingly stretched thin, AID missions may be working in isolation from other donors, with the result that missions can be unaware of donor activities within the same sector or country.(18) This could result in a very skewed approach to development, with some regions inundated with too much funding, while others are totally ignored.

Actions

1. AID should reengineer the project and program management processes to emphasize innovation, flexibility, and beneficiary participation.

Efforts should focus on:

--- organizing around project outcomes;

--- identifying and removing steps, procedures, and requirements, both internally and externally imposed, that do not add value to the process and create impediments to accomplishing project results;

--- redesigning jobs, functions, structures, and other processes (e.g., contract and procurement systems), that support project results;

--- putting the decision point where the work is performed and building control into the process;

--- improving continuity and accountability for projects by timing rotation assignments so they encompass to the extent possible the project life from design through closure; and

--- identifying and removing barriers that impede the full participation of indigenous populations and creating incentives for their inclusion in all phases of the project management cycle.

2. The AID administrator should designate selected AID missions as pilots to experiment with innovative approaches to programming and delivery of AID assistance.

These missions should be given waivers on rules and regulations that are deemed impediments to achieving project results. The four models presented in this report provide a number of innovative approaches that could be adapted for AID projects.

3. AID should structure the reward and incentive system in project and program management to ensure that performance and accountability are linked to accomplishing project results and that innovation is encouraged.

Project success should be the primary factor in evaluation. Rewards and incentives should be designed to create project success based on interdependency between key players. AID staff and project beneficiaries should be rewarded on project success based on clear performance indicators including mutual accountability, cooperation, and results. For example, performance appraisal systems should include beneficiary satisfaction and quality of beneficiary participation as performance indicators to measure AID's staff and contractor performance at the field level.

Likewise, host countries should be required to demonstrate sufficient commitment toward a project before it begins. Future project funding should take into account host country performance on existing projects. Where appropriate, future contract and grant awards should take into account a contractor's past performance and its ability to establish collaborative linkages to indigenous organizations.

Incentives should also be included that reward innovative approaches to development at the field level. The projects listed in this report provide a number of ways that organizations can encourage innovation. These models should be analyzed and adapted for AID use in the field and headquarters where appropriate.

4. AID should establish systems for continuing critical review of all existing projects to ensure that they are achieving desired outcomes.

As part of this review, each mission should address the following:

--- How does the project relate to new priorities described by the AID administrator?

--- Are local conditions in place to ensure project success?

--- Are results on track?

--- If not, can the project be salvaged in an appropriate period of time (e.g., 6 to 12 months)?

Reviews should determine whether projects need to be terminated or provided additional funding. AID should identify and remove existing internal and external barriers that have prevented the agency from terminating unsuccessful projects. Mechanisms are needed to ensure that resources continue to flow to those activities that achieve results.

5. AID should strengthen the project and program evaluation process and integrate it into the planning, budget, and project allocation processes.

AID should develop mechanisms that incorporate evaluation findings into all stages of the decision making process. Evaluations should be used to help guide managers with planning, budget, and project allocation decisions. Evaluations should focus on results. Funding for projects that are not producing any benefits should be discontinued.

Impact evaluations should be more than lessons learned. They need to be taken seriously and their recommendations should be acted upon. It needs to be recognized that this may require strengthened analytical capabilities at AID. Increased focus on results in other assessment and accountability mechanisms, such as audits, inspections, and other reviews, would assist managers in planning, budgeting, and project allocations.(19)

6. AID should improve donor coordination, both at headquarters and in the field.

AID mission directors should designate an officer responsible for coordinating the efforts of the AID mission and other donors. Where appropriate, all AID mission staff should be encouraged to coordinate policies and leverage resources through such means as extensive consultations, joint ventures, and partnerships with other donors.

Implications

These recommendations support the reinvention efforts outlined by the AID administrator. Overall, they are intended to create a high- performance, results-driven agency. At AID, the internal structures, external barriers, and burdensome procedures that have accumulated over the years make it more difficult to achieve program results.

Reengineering the project management system under a Grameen-type social learning model of development could provide a more effective process, supporting creativity and flexibility while focusing on the customer to achieve project results. Incentives that tie performance to results will increase productivity and ensure accountability to beneficiaries.

The system and processes will be adapted to bring in the customer at each stage of the development process. By focusing on the customers (in this case the project beneficiaries), projects will be designed to adapt to local conditions. Bringing indigenous groups more directly into the project cycle will ensure that projects better reflect their needs. As a result, greater beneficiary ownership will be achieved to ensure sustainability and accountability. AID will leverage resources in collaboration with other donors and NGOs to support host country activities.

Projects deemed unsuccessful in the evaluation process should be phased out. Reallocated resources will flow to projects that are tied to new priorities and are actually achieving results. Although these recommendations target the project cycle, they can also be applied to AID programs and non-project activities that may encounter the same types of problems experienced in the project cycle.

Fiscal Impact

The specific savings associated with reengineering the project and program management system cannot be determined. Reduction of overhead administrative costs, increased reliance on indigenous populations, stringent reviews, and termination of unsuccessful projects could result in substantial savings. The process of reengineering is long and arduous, and an increased focus on strategic planning and evaluation may result in up-front costs. This is not expected, however, to create new spending demands. On the contrary, the removal of existing internal and external barriers should reduce costly administrative overhead. This should offset any added costs incurred as the agency reengineers its management of projects and programs.

Endnotes

1. The President's budget request as shown in the Congressional Presentation Fiscal Year 1994 (Washington D.C., April 1993), pp. 21- 23, and as amended by the President's request for an additional $200 million for the former Soviet Union.

2. Hellinger, Stephen, Douglas Hellinger, and Fred M. O'Regan, Aid For Just Development: Report on the Future of Foreign Assistance (Boulder and London: Lynne Rienner Publishers, Inc., 1988), p. 22.

3. Ibid.

4. Ibid.

5. Gang, Ira N., and James A. Lehman, "New Directions or Not: USAID in Latin America," World Development, vol. 18, no. 5 (1990), pp. 723- 732.

6. Rondinelli, Dennis, "Reforming U.S. Foreign AID Policy: Constraints on Development Assistance," Policy Studies Journal, vol. 18, no. 1 (Fall 1989), p. 74.

7. Ibid., pp. 67-85.

8. U.S. General Accounting Office, Foreign Assistance: AID Can Improve Its Management of Overseas Contracting, NSIAD-91-31 (Washington, D.C.: General Accounting Office (GAO), October 1990), pp. 25-35.

9. U.S. General Accounting Office, Foreign Assistance: Promoting Judicial Reform to Strengthen Democracies, Draft Report, NSIAD-93-149 (draft), pp. 1-25.

10. Ibid., p. 5.

11. Hammer, Michael, and James Champy, Reengineering the Corporation: A Manifesto for Business Revolution (New York: HarperCollins Publishers, Inc., 1993), pp. 28-30.

12. See, for example, U.S. General Accounting Office, Foreign Assistance: Promoting Judicial Reform to Strengthen Democracies (draft), p. 5.

13. Further innovative approaches can be found in Browne, Stephen, Foreign Aid in Practice (New York: New York University Press, 1990), pp. 147-153. They include the Saemaul Undong in South Korea, Sarvodaya Shramadana in Sri Lanka, Aga Khan Rural Support Programme in Pakistan, Harambee in Kenya, Amual Dairy Cooperative in India, environmental movements such as Chipko in India and the "Six S" in West Africa, urban slum groups in Santa Marta in Rio de Janeiro, Tondo in Manila, and Long Toey in Bangkok.

14. See Hossain, Mahabub, Credit for Alleviation of Rural Poverty: The Grameen Bank in Bangladesh (Washington, D.C.: International Food Policy Research Institute, February 1988), pp. 1-89, for more detailed analysis on quantifiable results achieved through the Grameen Bank.

15. See Korten, Frances F., and Robert Y. Siy, Jr., Transforming a Bureaucracy: The Experience of the Philippine National Irrigation Administration (West Hartford: Kumarian Press, Inc., 1988), for further analysis.

16. Ibid.

17. Ibid. See also Osborne, David, and Ted Gaebler, Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector (Reading, MA: Addison-Wesley Publishing Company, Inc., 1992), for a fuller discussion of innovation in government.

18. Agency for International Development, "AID's In-Country Presence: An Assessment," Washington, D.C., October 1992, p. 32.

19. See also the NPR Accompanying Report Mission-Driven, Results- Oriented Budgeting.

Return To Report Index