to concentrate our attention on these approaches and to attempt, on a broad basis, to engage the poor in the development process.

It is also important to stress, however, that while we are proud of our accomplishments and eager to move ahead rapidly in support of this strategy, we must also be modest as we discuss the future. Determining the precise application of general development approaches in specific cases remains, despite all our efforts and those of thousands of practitioners and scholars alike, a very murky, difficult, uncertain, complex, and intractable business. The rapidly changing circumstances and conditions, the special characteristics of individual societies, the vagaries of the international economic system, and so forth, all suggest that modesty, especially as we confront other nation's problems, should be an important governor on our actions. Our difficulties in predicting our own economy emphasize the need for restraint with others.

Moreover, while AID and the U.S. Government must, of course, be responsible for our programs, it is essential that we always keep in mind that the decisions regarding development policies and practices in the LDC's lie with the sovereign governments of those nations. Our influence is profound in some, slight in others, but in all cases must be exercised in a collaborative and sensitive style if we are to continue to be a welcome force for development. As is clear from recent overt expressions of the tension between developed and developing nations (for example, the breakdown of the energy conference in Paris, the group of 77 positions in various United Nations forums, and so forth), the changing pattern of relationships between countries underlines the need to respect the integrity of other nations if cooperative solutions to world problems are to be found.

In making their individual decisions on development strategies, recipient governments are now well aware of the emphasis of AID's own legislation. We have made a concentrated effort to inform key officials at all levels of the meaning of the new legislation and in general have found them sympathetic to the same goals. It is evident, however, that a great gap remains between sentiment, rhetoric, and policy in many nations, and in some cases, a fundamental concern that we may be construing the necessities of national development too narrowly from their vantage point. This is especially true in Africa where basic infrastructure is most needed. Governments are increasingly aware of what kinds of projects we are prepared to support, and in general these are the types of projects we are asked to consider. The shift in program concentration in the past 3 years, as shown in the table on the following page, is indicative of the degree of our change. There are projects turned away as inconsistent with the new directions-a sampling is contained at appendix 2, p. 41. Governments now turn to the international financial institutions, United Nations development program, and other bilateral donors for projects essential to a balanced development program which fall outside our priority areas; as is clear to all, the fact that we rarely fund major infrastructure projects or many other activities does not obviate the need for them, especially in Africa.

DEVELOPMENT ASSISTANCE FUNCTIONAL PROGRAMS, FISCAL YEARS 1973 THROUGH 1976

1 For purpose of comparability fiscal year 1973-75 data excludes operating expenses. In fiscal year 1976 operating expenses are proposed as a separate account.

As the committee has pointed out, the task of achieving substantial progress for the billion or so very poor people in countries receiving AID assistance is not an easy one; it requires an extremely large investment of resources, long periods of time, and very difficult policy choices for governments. We have not had universal success in our own country in eliminating poverty-and the situations are not totally unrelated-and so we have great sympathy for nations trying to make the right choices for their own societies. Because of the large investment of resources required, it is clear that a healthy economy, one which is multiplying job opportunities, government revenues, and foreign exchange earnings, will enhance the prospects of a successful attack on poverty. The pace at which governments are able to move toward concentration of their resources in projects that not only help but involve the real participation of the poor is one of the most difficult choices for LDC's. In part this is because of the political impact of such decisions, in part because the appropriate answer in each case is not always crystal clear. AID can help, but the political movement-which we believe is essential in the long-run interest of these societies— will not come overnight. We have embarked on a venture which requires that the United States make unequivocally clear its intention to stay the course.

B. DEFINITIONS

In addressing the issues raised by the new legislation we have considered it important to define certain terms and concepts so there would be a greater degree of consensus inside and outside the Agency. While these have been described at length in an Agency paper entitled "The Congressional Mandate: Aiding the Poor Majority" (see appendix 5, p. 63) issued on April 30, 1975, several key elements are summarized here.

WHO ARE THE POOR MAJORITY"

Few officials in developed or developing countries have spent much time on that question, perhaps feeling that "you know the poor when you see them" and that attention could more usefully go to designing and implementing programs for people who are obviously poor by any reasonable standard. We are sympathetic to this view and do not intend to belabor the question, but the need to be sure of our focus at a time when AID appropriations are particularly tight requires that we

always have in mind what we want most to accomplish and for whom. A closer look at the characteristics of the poor may suggest ways of improving the effectiveness of AID programs.

As an aid to characterizing the poor majority, we use several rough "benchmarks" or standards of poverty. Falling short of any one benchmark is enough to place an individual in this vast group which totals over 800 million people by our standards, or around three-fourths of the total population of AID-assisted countries; in some countries more than 90 percent of the population is in this group, while in other better off countries the proportion is far lower.

In interpreting these benchmarks we stress the need to consider the spirit of the legislative reform as precision will be difficult to achieve for some time given the frequent absence of reliable statistical data in LDC's. This is especially so as these indicators are intended to apply to varying proportions of country populations, not to countries as a whole.

The following benchmarks are used:

(a) Per capita income below $150 per year;

(b) Daily diet of less than 2,160 to 2,670 calories (depending on the country); and

(c) Several health indicators: life expectancy at birth of below 55 years, infant mortality over 33 per 1,000 children aged 0 to 1, birthrates over 25 per 1,000 population, or access to broadly defined health services for under 40 percent of the population.

HOW ARE AID'S TARGET GROUPS TO BE CHOSEN?

Moving the poor majority beyond the poverty benchmarks noted above would be an extremely expensive and lengthy process even in optimal policy settings. Although we cannot say with precision, the price tag for each year would most likely be a multiple of the LDC's present gross national product and the aid donors' share well beyond the realm of possibility. To suggest how large the job is, with 5 percent annual growth in real GNP and 2 percent annual growth in population, per capita income would double only after 25 years-assuming steady application of needed resources-and perhaps still fall short of $150.

As AID's resources-like those of other donors and of the LDC'sare limited, it is normally impractical to think of AID-financed programs affecting directly the entire poor majority in any country, much less moving it beyond the benchmarks in the near term. But wherever possible, AID support must be part of a development approach conducive to broad-based systemic change. While AID-financed programs must attempt to reach large numbers of poor people, AID's primary target group will often be a limited portion of the poor majority in each country depending on its economic and social conditions, its capabilities and desires, and other considerations which determine the programs yielding the most impressive benefits at least cost. AID's programs will also be designed to yield secondary benefits to as many as possible among the poor, and certainly to avoid worsening the plight of the poorest. Once again we recognize the difficulty of tracing out exactly who is affected by an activity and what the long range consequences are; AID's Philippines mission is making a valiant attempt to do so but finding it tough going.

In pressing ahead with the new legislation, AID assistance focuses

on:

-concentrating on countries whose development policies we can support and that can utilize our assistance effectively;

-concentrating on key sectors (food and nutrition, population and health, and education) affecting the basic well-being of the poor; -providing key components (frequently in concert with other bilateral and international donors) of development packages designed to involve and affect broad segments of the poor majority, thus multiplying the impact of our assistance;

-supporting selected pilot programs testing new approaches with potential for affecting many people, thus encouraging the experimentation needed to advance the art of development. Above all, we must be prepared to assume risks.

HOW DIRECT MUST AID'S ASSISTANCE TO THE POOR MAJORITY BE?

While we do not believe that there is any serious misunderstanding on this question, the frequent reference in committee reports and elsewhere to "direct" assistance to the poor majority suggest that a definitive statement on this topic would be useful.

AID supports and assists LDC agencies in planning, financing, implementing, monitoring, and evaluating programs and projects which promote development activities which primarily and directly deal with the problems of and are intended to benefit the poor majority.

AID therefore supports activities which directly benefit the poor majority or support, through assistance in planning and institution building, LDC agencies that deal directly with the problems of the poor majority. Almost invariably AID assistance would reach the poor majority not "directly" from U.S. advisors working with villagers, but through:

(a) public or private intermediary institutions, and

(b) advice leading to change in LDC policies which, in several ways, might improve benefits to the poor (for example, policies which influence the availability of opportunities—including employment-and the supply and cost of basic goods and services).

WHAT IS THE MEANING OF "PARTICIPATION"?

Development progress for the poor will require time-consuming systemic change. Programs most likely to succeed, and which receive highest priority emphasis under the congressional mandate and AID policy, are those involving the active and effective participation of the poor in all facets of the development process. This is clearly not a simple proposition to implement, especially as it has important political implications. We shall refer to participation frequently in the following section of the report, but to briefly describe the participation approach to our staff we have used the following:

-Economic benefits are widely and significantly shared by the poor with the objective of narrowing the relative income gap between rich and poor, for example, the co-op which benefits small farmers. -Decisions concerning the activities to be carried out are made,

55-310 O-75-3

preferably, by those benefited (for example, the poor), or if not, at least with effective consultation and substantial acceptance by those benefited.

-The activity in which they participate is, ideally, a learning experience for benefited persons, which increases their technical skills and/or their capacity to organize for common purposes and for greater access to the benefits of development.

-The poor make a significant contribution in effort and resources to the activities from which they benefit, for example, through personal savings, or serving as members of local planning or project implementation committees, or through actual project implementation.

-The participation and contribution of women should be explicitly taken into account under the above-mentioned considerations, for example, any of the above or other examples when the participants

are women.

C. POLICY ISSUES

As anyone knowledgeable about economic development problems will testify, there are a large number of very difficult policy choices that must be made. While these are the responsibility of the country receiving assistance, we must also decide on the most sensible course for the U.S. Government to pursue. Our support of a participation strategy concentrating on a few key sectors is an illustration of a choice made. But there are many other serious areas of economic, social and political policy that inevitably drive toward the core of national development as we try harder to change traditional approaches. This section will attempt to describe some of AID's work and thoughts in key areas, recognizing that much is still new terrain to all of us.

1. SECTION 103-FOOD AND NUTRITION 1

While entitled Food and Nutrition, in AID's view section 103 of the FAA is broadly interpreted in legislative history to consist of activities in support of rural production rather than simply agriculture, with the emphasis on linkages between agriculture, industry and marketing.

AID provides only a fraction-albeit occasionally a large one-of the foreign resources any LDC uses, and it is the LDC's own resources and its own development policies that are the primary determinants of development progress for its own people. Whenever possible, therefore, AID support must be part of a development approach conducive to the broad-based systemic change needed to affect the lives of the poor. Barriers to development in the economy, society, and politics of the LDC must be identified and a package of coordinated policies established to remove or weaken those impediments. (AID-assisted sector analysis work is a major step in helping LDC's move in this direction. Many Latin American sector loans have been based on such analyses.) Not all governments have fully committed themselves to this task, but AID's assistance seeks to support programs that contribute toward expansion of such a broad-based view. (A fiscal year 1975 loan to Nicaragua illustrates just such an effort.)

1 See appendix 3, p. 43, for a summary statement of AID's program in this account.