FOOD PRODUCTION AND EQUITY
This is a preliminary paper drafted by A.I.D. staff
and does not represent official Agency policy. It
was prepared principally by Martha Horsley and has
benefited from comments by Regional Bureau, TAB, and
PPC staff. The paper is intended to lead to a better
understanding of the many factors and complex relation-
ships involved in the analysis of production and equity
effects in agriculture and to serve as a point of departure
for discussion leading to better articulated AID policy.
U.S. Agency for INternational Development
Bureau for Program and Policy Coordination
Office of Policy Development and Anaysis
The purpose of this paper is to investigate, at a very general level,
the compatibility of two agricultural sector objectives, food production
and equity (i.e. increasing the real incomes of the poor), under existing less
developed country (LDC) conditions. Both of these objectives should be viewed
as subordinate to the ultimate goal of establishing a minimum standard of living
for the bulk of the LDC population. The purpose of the restriction "under
existing LDC conditions," is to define the strategy choice in terms of options
which are obvious and may be feasible in most LDCs. In crude terms, the two
options considered are "small farmer development" and generally-unfocused
agricultural development, which in fact turns out to be large farmer biased.
The latter is therefore referred to as "large farmer development".
An optimum strategy in most LDCs would include major changes in the
distribution of productive resources, in pricing policies, and in organizational
forms; these are discussed briefly at the outset. The paper does not focus
on such changes in detail, however. Rather, it concentrates on the effects
of the two producer strategies under given organizational structures, forms
of land and other asset ownership, and overall pricing policies.
Alternative means of achieving production increases -- investment in
education, health, extension, credit or group-oriented institutions -- are
discussed, implicitly, on an ad hoc basis only. Any assessment of their
relative merits would require a more specific context.
1. The Two Objectives: Food Production and Equity
AID has two major objectives in its program of assistance for agricultural
development. One is a "production" objective to help the less developed
countries (LDCs) increase their domestic food production in order to alleviate
hunger and malnutrition. The other is an "equity" objective to help increase
the real incomes of the lower income groups, commonly referred to as the
The objectives of food production and equity stem directly from the Agency's
Congressional Mandate as described in the Foreign Assistance Acts of 1973, 74,
and 75 plus relevant Committee reports, as well as U.S. commitments stated in the
World Food Conference.* Regarding the equity objective, the 1975 House of
Representatives bill states that assistance to the food and nutrition sector --
which claims the majority of AID's development assistance -- "shall be used
primarily for activities which are specifically designed to increase the
productivity and income of the rural poor." Regarding the production objective,
Resolution I of the Food Conference highlighted the seriousness of the world
food problem in the following terms:
An increase in agricultural productivity and sustained
expansion of food production in (LDCs) at a rate much
faster than in the past is essential in order to meet
the rapidly growing demand for food, due to rising
population and incomes (plus) the requirements for
security stocks,,and to raise the consumption of under-
nourished people to universally accepted standards...
The 1975 House Report of the Committee on International Relations states
its belief that concentrating development assistance on the smaller, and
hence poorer, farmers (which would have an equity impact in the sense that
*See refs 1, 19, 20, and 37.
Cheir real incomes would be increased) and increasing food production are
complementary activities .
the committee reiterates its belief that small-farm, labor-intensive
agriculture is the key to greater food production, as well
as to more equitable distribution of income in the rural areas
of the developing countries, where most of the world's poor live. (p.46)
The 1974 House Report, however, clearly recognizes the possibility that there
may be a conflict in attempting to satisfy both objectives simultaneously.
Thus the Report states, while a small farmer strategy is more consistent with
the Congressional injunction to support activities which "directly improve the
lives of the poorest people and their capacity to participate in the
development of their countries", there may be circumstances imposed by the
world food crisis which justify temporarily relinquishing a small farmer focus.
The Report goes on to say:
AID should, however, take care that its programs in support
of short-term production increases in certain countries, as
opposed to those in direct support of longer term structural
changes needed to increase the productivity of the small farmer,
are kept to the minimum required to deal with the current emergency
situation and are not used to the detriment of the small farz -.ana
the rural poor.
Perhaps the best way of looking at the issue is to say that the production
and equity objectives are both intermediary to the ultimate goal of achieving
a minimum standard of living for the bulk of the LDC population. Conceptually,
the food production objective can be interpreted as concern with the supply
side of the minimum standard of living objective, since food may account for
half or more of consumption requirements among lower income groups. The
equity objective can be interpreted as concern with the demand side. Unless
the real incomes of the poor are increased, they will not be able to purchase
the increased food supplies, or other goods necessary for a decent living
standard, in the market place. The question to be addressed here is: Given
the possibility of conflict, or trade-off, between the production and equity
objectives, what is the best means of achieving the combined minimum standard
of living objective? The balance of this first section of
will seek to clarify the meanings given to the objectives of food production and
equity before turning to this basic question.
Improved equity is defined here as improved ability on the part of those
who need it, the poor, to obtain a higher standard of living. It would thus
entail, at a minimum, an absolute increase in the real income of the poor,
and, ideally, a concomitant improvement in their real income relative to that
of higher income groups. The poor, or the "poor majority" in the terms of
the Congressional Mandate, have been defined in income, nutrition and health
status terms and probably include most LDC small farmers as defined below
in Section III A. An increase in real income could of course result from a
fall in food (or other basic consumer-good) prices or increased production
and consumption of home-grown food as well as an increase in cash income
(not offset by a rise in the general price level). A refinement of the
equity concept can be made by distinguishing between a "direct" equity
impact, where resources or benefits are channeled directly to the "poor
majority" target group, and an "indirect" equity impact, where real incomes
of the target group are raised through secondary effects, say, lowering
relative food prices or increasing demand for landless labor. This distinction
will be useful in subsequent sections which attempt to compare the equity
effects of two producer strategies.
Regarding the food production objective, it might be argued that
increasing aggregate supply rather than domestic food production alone,is
the appropriate objective since imports from food surplus countries could
provide an alternative means of increasing food supply. While not excluding
the possibility of trading non-food exports for food, this paper will focus
on increased domestic production as having the major potential for
increased LDC food supply. For some LDCs, such circumstances as relatively
small size, unfavorable physical environment for food crops, and favorable
foreign exchange position and prospects could justify significant reliance
on imports to meet food deficits. But reliance on imports to meet the bulk
of LDC food needs does not appear feasible for/ Projections to 1985 based
on current trends of income, population and agricultural production yield
up to a five-fold increase (from 16 to 85 million tons) from 1970 to 1985
in the cereal deficit alone for LDCs.* Developed countries could conceivaUty
provide the imports required to meet such a massive deficit but this assumes
that the balance between demand and supply growth in these countries will
yield the requisite surplus and that developed countries will be willing to
provide concessional financing on a much larger scale than heretofore (see
USDA source cited in footnote for further discussion).
There are, on the other hand, several real political and economic
reasons why many LDCs desire self-sufficiency in food production. These
include continuing restrictions on world trade in agricultural commodities,
and the role of political considerations in influencing allocation of food
aid. Finally, the fact that agricultural yields are quite low in most LDCs
(including many with favorable natural endowments) relative to those prevailing
*This is an FAO estimate used by the World Food Conference (the base period
is an annual average for 1969-71). The figures exclude Asian centrally
planned economics. While alternative projections, based on different
definitions and assumptions, prepared by USDA result in somewhat lower
deficit the relative increase is still massive (from 20 million tons
in 1970 to 50 to 70 million tons in 1985, depending on assumptions).
For a comparison and discussion of various projections, see USDA, Economic
Research Service, Foreign Agricultural Economic Report No. 98, The World
Food Situation and Prospects to 1985, pV. 32-39. (ref.36 .)
in developed countries, tuKKgets both to LDCm and donor agencies the potential
for meeting the bulk of food needs within LDCs themselves through improved
mobilization and allocation of local resources -- supplemented with aid from
developed countries in the form of agricultural inputs, research and technical
Section II identifies some of the major reforms required in many LDCa to
create an environment conducive to achieving the minimum living standard goal.
Although the discussion is brief, the policy implications for AID are clear.
Section III, which constitutes the bulk of the paper, attempts to analyze the
food production and equity effects of alternative producer strategies (small
farmer and large farmer) under existing conditions, i.e. assuming no major
reforms take place. Section IV summarizes the conclusions of the previous
section and discusses briefly the program/policy implications for AID.
II. Major Reforms
There is considerable agreement among development experts that major
reforms are required in most LDCs to create an environment conducive to
achieving the minimum standard of living objective for the rural areas. There
is less agreement, of course, on the particular form and extent of the reforms
required although they can be generally characterized as falling into one
of three categories:
i. reforms affecting the distribution of productive resources,
ii. reforms related to pricing policies, and
iii. reforms related to organizational forms and development approaches.
Certainly one of the factors inhibiting greater participation of the
poor in the development process is their limited ownership of (or access to)
*This is not to argue that food aid cannot play a complementary role beyond
meeting emergency situations. It will be suggested later in the paper that
food aid can help meet short run food deficits in LDCs where broadly-based
food production efforts have longer gestation periods.
. productit,/v rehourcets: lian, phyHical capital, and "human capital". LanA
redistribution, land-to-the-tiller prograHm, and resettlement on new land are
alternative means of providing lower income farmers or landless agricultural
laborers with their complementary primary resource.* This is perhaps the most
powerful redistribution mechanism in the rural areas since it has broad
implications for political power and the distribution of government services.
The wide-spread availability of agricultural inputs, and financial capital
needed to acquire these inputs, can facilitate the accumulation of physical
capital; and improving the quality and quantity of rural education can help up-
grade the human capital factor by giving the individual the means of acquiring
additional knowledge and skills. Similarly, improved health and nutrition
will help him attain his potential. Finally, success in population control
will increase the per capital level of all these resources and hence the
likelihood that the minimum living standard objective will be achieved. The
problem is, of course, that all these reforms,with the possible exception
of the land redistribution and land-to-the-tiller program (and even these
require substantial resource inputs if production levels are to be sustained
require massive human and financial resource inputs.
The second category of reforms, i.e. those relating to pricing policies,
involve agricultural terms of trade with other sectors, the pricing -of inputs
and outputs so that proper and adequate production incentives are transmitted,
and monetary and fiscal pricing policies which encourage saving, investment,
adoption of appropriate technologies, and desired income redistribution. An
agricultural land tax, based on potential rather than actual productivity, is a
frequently recommended example of fiscal reform. Pricing policies are also
difficult to change, however. Poor input and product prices are often due to
*See ref. 13 for a recent summary of the relationship between land reform and
'misconceptions and/or political design. Cheap credit for small farmers, for
example, sounds good but results in extremely limited total credit availability,
partly because of decapitalization. Political constraints on other policies,
e.g. the agricultural land tax, arise from the fact that political and economic
The third category of major reform, organizational forms and development
approaches, would entail increased attention to spatial organization, regional
planning, and integrated rural development. Depending on existing constraints,
reaching the poor might involve rural infrastructure development, promotion
of market towns and rural industrial development, and/or agricultural re-
organization, including land consolidation, collectivization, or decommunalizatior
Proponents of more effective regional and rural development** argue that
significant long-run benefits are associated with establishing rural growth
poles even if there are short-run losses in efficiency. Rural industry tends
to be more labor-intensive, for example, thus providing income-generating
activities for a larger segment of the population. In addition, improved
rural economic organization tends to stimulate increased investment in agriculture
improved input and product markets and more widespread and rapid adoption of
innovations. There are problems, however. Although it has been shown in China
that comprehensive regional or rural development can be largely self-financed,
the typical LDC central government approach requires substantial resource
allocation (or reallocation). In order to conserve scarce resources, the
programs must be carefully integrated with pricing policies. In addition, it
may not be clear what specific strategy is optimum in a given situation.
As a result, few LDCs have made the necessary financial and manpower commitments.
*See ref 25 for general discussion of agricultural pricing policies
**See refs. 28, 29, and 30.
The reforms mentioned here, of course, have their less radical counterpartat
land tenure reform rather than land redistribution, agricultural subsidies
rather than basic pricing policy reform, and selected rural development project
support rather than comprehensive regional planning efforts. Government
action often takes these less radical forms, for both economic and political
reasons. And therefore the following discussions of the effects of alternative
producer strategies generally assumes existing patterns of resource holdings,
market imperfections, and organizational forms. This is not to imply that
change is impossible, only that it is slow in coming and, when it comes, it
is partial. *
To clarify somewhat further,it will be important to keep two points in
mind throughout the following discussion. One is that the assumption of no
major reforms lends a conservative bias to the discussion of small versus
large producer potential, i.e. more scepticism must be introduced about
the ability of the small farmer to contribute to production goals than if
major reform is assumed. In fact, in the terms of the following discussion,
if all three types of major reforms were instituted, there would be less
reason to specify a target group approach. The development process might
well exhibit major complementarities, rather than conflicts, between the production
and equity goals. The other point is that, despite the difficulty of reforms,
those mentioned probably constitute the most effective means of achieving
the minimum standard of living objective. Many development experts have come
to the conclusion that the forces of inequality are so strong that "pro-poor"
programs without the above reforms often end up primarily benefiting the non-
poor. Nothing short of major equity-oriented programs, undertaken on several
fronts simultaneously, will succeed in income redistribution.**
*Ref 31 provides an example of the economist's increasing interest in taking
into explicit account the institutional setting of development.
**See refs 9 and 16 for additional observations on the "elusiveness of equity.
Thus, where LDC governments are interested and willing to undertake reforms,
AID might wisely adjust its funding priorities in favor of support efforts,
involving, for example, multiple goal sector analysis, regional planning, Land
reform and revision of agricultural policies.
III. Alternative Producer Strategies
The discussion in this section is confined to alternative producer strategies.
Producer strategies are considered important because they can affect both the
supply and the demand side of the minimum living standard objective and because
they directly involve the majority of the active agricultural population. No
further value judgment regarding their superiority over other possible strategies
-- focusing on say landless agricultural laborers or on increasing food
availabilities to the consumer by reducing losses from off-farm handling and
storage -- is intended. In particular, this section will analyze the food
production and equity implications of a small farmer strategy versus a strategy
which is neutral in intent with rerard to its beneficiaries. The latter, in the
absence of malor reforms, turns out in practice to be a large farmer strategy
and will be referred to as such below.
The selection of a small farmer target group as one strategy option is
an explicit attempt to introduce an equity bias into a production-oriented
program. The pros and cons of such a bias have been hotly debated among aid
donors and within LDCs for the last few years, although not necessarily in
the present context. It is sufficient to note here that although a small
farmer strategy by definition entails greater direct equity effects than
a large farmer strategy, the ultimate impact on a combined goal of food
production and equity (including indirect equity effects) is not obvious.
Section A below discusses alternative criteria for defining a mall
farmer target group and proposes farm size as perhaps the best available proxy
for income and wealth. Some rough target group definition is a prerequisite
to any further analysis, even at the general level. Section B Dren nfe w* iA--
on relative productivittle by farm size, and Section investigates the
total equity effects of small versus large farmer strategies. Section D
attempts to draw the major implications of alternative stratiie for oera-i -
rural and national development.
A. Target Group Definitions
As noted earlier AID's Congressional Mandate requtirir -priority
attention to assisting the "poor majority", in LDCs. It is not entirely
clear, however, what constitutes a "small farmer" by reference to the poor
majority. The working definition of the poor majority elaborated by the
Congressional Mandate Implementation Task Force includes all individuals
with a per capital income less than $150 (alternative definitions pertaining
to nutrition and health status have also been offered -- see ref.l), however
income data are difficult to acquire in the rural areas. Moreover, the
ideal equity index in the agricultural sector would include a measure of
assets, which is even more difficult to acquire.
Farm size is a poor proxy for rural poverty for a variety of reasons:
(a) It does not reflect non-agricultural income obtained
by rural households;
(b) It may not reflect differences in yields resulting from
differences (1) in land quality (2) degrees of irrigation
and (3) cropping patterns;
(c) It may not reflect differences in price of given yields
resulting from market imperfections -- monopoly monopsony
problems in agricultural markets.
Nevertheless, it is probably the best income/assets proxy for which data
are available across countries,.and it has the advantage of being a convenient
definition for measuring the evidence on productivity effects of alternative
CLASSIFICATION OF AGRICULTURE. POPULATIONS BY FARM SIZE
Target Groups Relative Area Limits Proportion of Agricul- Proportion of Land
(hectares per holding) turalpopulation
SuL r-ar4 inals*
2.5 to 14.9
Medi',L:,' large farmers
Mt'.i.-- large fanners
2.5 to 7.49
!!E- 1 1
1 to 4.9
-".- : .-" inr.-lude landless agricultural workers, whi-ch
nag ;:..1 ral population in some countries.
constitute 30 per cent or more of the active
Source: .'apter V of ref. (6). The farn-size limitation for 'small farmer" has been expanded upward
.5 hectares for both South Asia and Tropical Africa in order to increase inter- gional
-'.parability of farm size and income.
The accompanying table gives a classification of agricultural populations
by farm size for three of the major developing regions of the world -- Latin
America, Tropical Africa, and South Asia. The agricultural population in
each region is divided into three groups:
submarginals (including landless agricultural workers), which constitute
approximately 50 percent of the populations; small farmers, which
constitute another 30 percent; and medium/large farmers, which make up
the remainder. The relative farm-area limits for the category "submar'ginals"
have been dictermined by using country-level estimates o' the size of
holding necessary to achieve minimum accepted standards of family
subsistence income. The smaller estimated subsistence holding in
Africa, for example, can be explained in part by the typically smaller
family ize. The "small farmer" category is also a relative concept,
base'i on regional norms.
Relating this table to AID's mandate to give direct assistance to
the "poor majority", we might include both the "submarginals" and the "small
farmers" in the target group. This would bring the percent of the agricultural
population included in the target group to about 80 percent in Latin America
and South Asia and 90 percent in Africa. (The percent directly covered by
the small producer strategy would be less, however, by the percent of
landless agricultural workers included under the submarginal category) The
percent of cultivated land eligible would amount to approximately 20 percent
in Latin America, 30 percent in South Asia, and perhaps about 30 percent
in Tropical Africa.
These regional averages disguise wide differences among countries and
among subregions within each country. They are only illustrative and have
limited operational significance. A small farmer strategy ultimately has
to be defined within the specific context of the individual country and
program examined. In such a definition, soil fertility and water availability
(rainfall or irrigation) characteristics might establish a "standardized
hectare" dividing line between small and large farmers. Besides farm size
other distinguishing behavorial characteristics of farm groups might also
be useful, e.g., a submarginal farmer might be distinguished from a small
farmer as a net buyer of food and a net seller of labor; "medium" farmers
might be defined to include only those interested in selective mechanization
to smooth out their seasonal demands for labor, while the definition, "large
farmers, might be reserved for those engaged in generalized labor-displacing
mechanization. Corporate plantations using traditional, labor-intensive
techniques for growing tea, rubber or sugar might form a fifth category.
The discussion in the following sections makes an arbitrary distinction
between the polar approaches of "small farmer" and "large" farmer" strategies,
though it is recognized that in practice there is a continuum of farm sizes
and groups, as well as of strategies that affect them. It does this for
purposes of presenting the broad conclusions from available investigations
which are typically country specific. Thus, the conclusions offered here
must be viewed only as broad tendencies and interpreted with care when they
are to be applied to specific countries or sub-regions.
R. Kvldenc:. on krel;at1vv Productivltilen
:itdli'. bnatie ,,r, ; irlcultural cetnrun data and sample surveys suggest
rhat net benefit (or value-added) from food production in the LDCs is similar
across farm size for a given value of resource inputs.* The findings draw
heavily on studies of foodgrains production but include some data on total
production by farm size. This emphasis is appropriate in view of our
concern with the food consumption patterns of the poor. Thus, in the
poorest countries, our primary concern might be for basic foodgrains
production. In somewhat more developed countries, this concern might
be extended to the supply of protein and food items satisfying other
basic nutrition needs.
The findings are typically reported in terms
of product per unit of labor, land, or capital (labor, land and capital
productivity, respectively). There is a clear tendency for labor
productivity to be higher on large farms, on the one hand, and for land
productivity to be higher on small farms, on the other. While the
relationship between capital productivity and farm size is less clear,
available evidence suggests a peak in the small to medium farm range
and a gradual decline as size increases. These countervailing relation-
ships reflect different technologies employed on different size farms
which, in turn, may be due to (a) different input and output prices
facing the small and large farmer, (b) varying degrees of risk aversion
and/or (c) different production possibilities due to different amounts
of fixed factor endowments (especially land/family labor). They thus
reflect a tendency for farmers to husband most carefully the resources)
which they perceive to be in relatively scarcest supply (clearly land
*See references 2, 4, 6, and 34 for cross-country evidence and 8 for
the Indian case.
for 'small farmers, and labor, especially in peak seasons, for large
farmers). These considerations lead small farmers to choose types of
crops and cropping systems (e.g./ multiple cropping) which require
intensive use of labor and result in relatively high yields per acre.
By contrast large farmers may engage in land extensive activities,
such as cattle raising, and enhance their labor input by combining
it with substantial amounts of agricultural machinery. At any rate,
the overall relationship, in terms of net returns per unit of total
resource costs, appears to be roughly one of constancy across farm
The significance of this constant ratio of net output to cost lies
in its implication of similar levels of economic efficiency across
farm size, and hence similar levels of food production for a given value
of resource inputs.** The ratio might in fact be better described as a
profitability measure rather than an economic efficiency measure, since
it includes the effects of crop choice as well as the effects of crop-
specific technology and input choice. In any case, the finding does bring
into question any unqualified pessimism about the viability of the small
farmer due to some inherent inefficiency.
*Reference 7 points out statistical biases in the U.S. farm data which
have led to the commonly held view that this relationship does not
hold true in the U.S. The author shows that when these biases -
related to farm class definition, estimated labor costs, and variations
in land quality -- are adjusted for, the U.S. data do show a rough constancy
of net returns to total resource inputs across farm size. (Note
that higher yields per hectare on large farms do not necessarily imply
higher net returns per.unit of cost.)
**There is a sizeable literature on the measurement and interpretation
of "efficiency" by farm size. The measure discussed here is a crude
one and is sensitive to the prices, or imputed prices, used to weigh
the components of the index. See ref26 and 32 for a more precise Mmeaure
which distinquishes price and technical efficiency.
A m;ji ]or.i., !.tatli of all the -fndings on productivity discussed
ilt,,ve I Iit i l# y dt.o I iL menjLur,e thl r lhin e in food production in
respond to a change In policies, resource inputs, institutions or
techniques. ( Kost'of them measure average productivity or yield per
hectare or per worker over a given time period.) One study attempting
to examine this question finds the results to vary depending upon the
input. Specifically, in the Punjab and Uttar Pradesh, India, production
response to increased fertilizer application was found to be higher for
small farmers, but it was found higher for large farmers in the case of
the provision of.irrigation.* A recent, less rigorous study looking at
adoption rates and productivity.changes in several Asian countries, due to
the introduction of new rice technologies, concludes that:
Farm size per .se has little meaning." It acquires significance
only when viewed within the context of the community, the
productivity of the land, the infrastructure, the services
available, the intensity of land use, population pressures, the
tenure system, and the social and economic values attached
to land ownership.**
Thus, it is apparently not farm size, but rather the conditions and
circumstances associated with farm size, that we must consider in any
assessment of potential production levels.
Perhaps the best available way to assess the potential change in relative
productivities is by looking at case history type studies of individual
projects and of *specific,.agricultural development efforts.*** These
studies often have a regional focus,' such as studies of the "Green
Revolution" in parts of Asia and Latin America. The time period turns
*See reference: to.study by'G.R. Saini In 6.
**See reference 10.
***See for example refs. 12, 15, and 21.
out to be an important consideration in analyzing the evidence.
The length of time it takes for program or policy actions to have an impact
will depend on several factors, including: (1) the differing technical
gestation periods of various inputs (application of fertilizer will have
a more immediate impact on production than construction of major infrastructure,
for example), (2) the time required to create or restructure institutions
to serve the agricultural sector or a particular group of farmers,and (3)
possible differences in the speed of adoption of new inputs and techniques
between small and large farmers, owing to lack of familiarity or skill in
employing new techniques, conflicting habits and customs, or aversion
to perceived risk.
In the short run (say 9 to 0 years), these studies suggest that
food production will be higher under a large farmer strategy than
under a small farmer strategy. The reasons for this lie largely in
the concept of access. Longer term inputs, such as irrigation, may
be available for large farmers but not yet completed for small farmers.
While seasonal inputs such as seeds, fertilizer and pesticides may
be theoretically available to small farmers, the lack of rural roads
and market facilities may inhibit physical access of the small farmer
to off-farm inputs. The large farmer orientation of the agricultural
credit and extension services may inhibit small farmers' institutional
access. Finally, although the evidence is mixed, small farmers may
be slower to adopt new practices than large farmers. The lag may be
due to labor or financial bottlenecks, risk aversion, or acculturation
problems -- unfamiliarity with new procedures, conflicting cultural
habits and customs. Improved analysis of factors determining small farmer
response requires additional empirical research.
Another factor related to access consideration is the degree of
divisibility of inputs. The greater the degree of divisibility, the
greater the accessibility of more productive inputs to small farmers.
Seeds,fertilizers,and pesticides are obviously highly divisible. Other
inputs may not be but can be rendered "divisible" with time and the proper
institutional arrangements. Irrigation and basic farm implements, for
example, can be made sufficiently divisible if appropriate technology is
employed (e.g., feeder canals and more efficient animal drawn implements).
Heavier equipment, where required (not for rice but perhaps for wheat
in some types of soils) can be made divisible through rental arrangements.
In sum, the implementation problems of helping small farmers obtain
increased access -- and in particular the requirement of sequential
programming -- may make it difficult to introduce changes in the short
run. Rural road construction, manpower training, and introduction of
new organizational modes all require both planning and execution time.
On the other hand, some experts believe yith proper extension advice,
there is considerable scope in the short run for increasing total farm
output through improved on-farm storage and more intensive land use. Also,
there is evidence in some areas that the small farmer -- due to his
intense commitment to the land -- may be more of a profit maximiser
than the typical large farmer and that, where ecological conditions
favor small-scale, labor-intensive farming, the small farmer will
invest proportionally more of his available resources.* Thus, one
might say that in the short run it is possible to get an increase in
agricultural production under a small farmer strategy equivalent to
that under a large farmer strategy but that it is not probable.
In the medium run, say from two to five years, where basic rural
infrastructure and trained manpower already exist or can be put in place
or trained quickly, and where appropriate organizational modes which
Sie reference 18
ke.ep corqt down ':nn bt. Iventilied, progress can be relatively quick.
1alti ircan re'iult. w.-re obtained f,:r example, in the early phase6of
the Comilla Program (East Pakistan) and the Ethiopia Minimum Package
Program in two or three years. It should be noted, however, that a
high-quality, intensive, pre-implementation planning effort was
involved in both cases. Also, we must be concerned with factors -- such as
manpower availabilities and political reaction from the elites -- required
for sustained success.
In the longer run, say 6 to 15 or 20 years, where physical
infrastructure can be extended to remote areas, where institutions --
for example, agricultural research institutions -- can be re-oriented toward
the small farmer, and where information, education, and new organizational
modes have had a chance to permeate the rural areas, food production
under a small farmer strategy should be as large for a given amount of
investment as under a large farmer strategy. Indeed, as explained below
in Section D, it may well be greater.
Despite these findings, the difference in the financial costs of credit
for small and large farmers is often cited as evidence of the inability of
small farmers to compete in the market place. Indeed project analysis shows
that interest rates charged to small farmers must clearly exceed those
currently paid by large farmers if the value of the portfolio is to be
maintained.* This may be a specious comparison, however, for several
reasons. First, sources of credit to which only large farmers have access
may be substantially subsidized below the opportunity cost of capital. An
observed difference between interest rates charged to large vs. small
*Note that this does not mean that small farmers will not be able to borrow
at the higher rate. If capital productivity is higher on small farms,
they may be able and willing to pay the premium.
farmers would in this instance be misleading. Second, assuming that no
subsidies exist, there are ways of reducing costs by administration of
small farmer credit through group approaches, as is apparently the case
for example in China. And third, even if higher costs of lending to
small farmers cannot be absorbed through group approaches, the difference
between total costs of all inputs per unit of output for small farms, as
compared with large farms, may not be as great as the lending cost difference
would suggest. This is because a smaller proportion of total inputs on
small farms may be credit financed.
Finally, it is worth noting under the general heading of cost
comparison that there are considerable "sunk costs" associated with
large farm development which are typically already in place. These
include roads to the farm gate, institutions oriented toward large producers,
fixed capital investment (including, for example, irrigation) and investment
in human capital There are few "aunk costs" associated with small farm
production however. Thus the short-run or "marginal cost" of financing
small farmer development is greater because it includes virtually all fixed
costs as well as variable costs, whereas the "marginal cost" of financing
large farm development may consist primarily of variable costs, e.g.,
seasonal production or marketing credit. In the long-run, however, total
costs of small farmer development need not exceed total costs of large
C. Equity Effects under Alternative Strategies
This section considers the impact of small and large farmer strategies
on equity; i.e., on increases in the real incomes of the poor. It concludes
that the ability of a large farmer strategy to match the equity effects of a
small farmer strategy in the short-to-medium run is open to question and even
more doubtful in the long run.
Sir.r.e smal farmers are taken to be included in the "pocr majority" (there
z=Iy Le exceptions in some countries, of course), a small farmer strategy, to
the extent it succeeds in raising the real incomes of small farmers, would appear
to be superior to a large farmer strategy on equity grounds by definition.
This ignores, however, the distinction suggested at the beginningof the paper
between a "direct" and an "indirect" equity impact. For a small farmer strategy
(and similarly for a large farmer strategy), the direct impact is the change in
real income accruing to participating small farmers from increased productivity
and returns.* The indirect impact is the change in real income accruing to
other groups in the poor majority (non-participating small farmers, the rural
landless poor, and the urban poor) who as (1) net buyers of food, benefit from
relative declines in food prices resulting from increased food production, as
(2) net sellers of labor, benefit from increases in real wages through increased
employment, and/or as (3) net sellers of food, suffer from a decline in product
price. The adverse/of (3) should be less under a small farmer strategy since
non-participating farmers (who are net sellers of food) will be fewer and
production increases (and henceproduct price declines) will be slower in coming.
Indirect equity effects through (1) and (2) are, however, uncertain.
In principle, it should be possible to (a) increase food production with a
large farmer strategy and (b) benefit those poor who are net buyers of food through
the income effect of lower food prices -- since food expenditures constitute a
large proportion of total expenditures by the "submarginals" and the utban poor --
or, (b') to the extent that food prices did not decline sufficiently to reduce
*Looking at the direct equity impact of a small farmer strategy, no distinction
need be made in financing food versus non-food crops, much less high value food
crops versus basic food crops. The indirect equity impacts discussed here, however,
do assume that primary emphasis in both strategies would be cubasic food crops.
profits to former levels, tax away either in income or in kind part of the large
farmer profits, using this revenue to finance poor-oriented production and/or
welfare projects.* In addition, in theory, since wages are linked to subsistence
needs, a decline in the price of food may result in a decline in wage costs
relative to other factor costs and possibly encourage a substitution of labor
for capital throughout the economy. This could generate additional employment
opportunities for the rural and the urban poor. Thus, both the food price decline
and the profits produced by a large farmer strategy could be linked to the equity
goal of benefiting the "poor majority."
The critical assumption of this approach is that retail food prices will
decline significantly or that part of the interim profits will be taxed away for
development purposes. However, where agricultural price supports are maintained
(at the retail as well as farm level) in the face of declining production costs--
as they were at least in the early phase of the Mexican wheat production program--
food prices will not decline and food buyers among the poor will not be benefitted.
In addition, few countries have had the political will to tax away large profits
reaped by the rich under any circumstances. Referring again to the Mexican wheat
production program, after the initial period when large farmers benefited enormously
from (1) the new seed varieties, (2) the government-financed irrigation investment
concentrated on large farms in the northwest, and (3) government-maintained prices,
price supports were removed and product prices declined substantially. High
interim profits of large farmers -- due in part to subsidies on capital inputs -
were not taxed away for development purposes, however, and the net result of
the early years of the program was to further skew the distribution of income in
the wheat growing areas.
*Another possibility, which may be more politically feasible, is for the government
to finance say rural public works in part through deficit financing. Success
would depend on the inherent income elasticity of the tax system and the ability
of the economy to absorb inflationary pressures.
Thus, t'ie theoretically favorable indirect equity impact of a large
f::rmer ftratf-y Ir the .hr.rt-to-medium term is questionable. In the long-run
.t is even more doubtful,.since demand for staple foods is income inelastic.
fI:ce basic calorie needs are satisfied, staple food demand will grow only as
fast as population growth. An increase in staple foods obtained with short-run
production increases on large farms may, in fact, pre-empt the small farmer
for years to come. Not only will current incomes of small farmers fall with
the'decline in product price (depending on the relative size of their marketed
surplus), but also in the absence of substantial cost-reducing innovations, future
production incentives will be depressed.
Distinguishing the rural from the urban poor highlights another
fundamental flaw in the equity case for a large farmer strategy. Private and
public food distribution networks and programs in developing countries are largely
aimed at urban rather than rural areas. This may reflect the presence of more
efficient transportation and marketing channels for linking rural with urban
treas than for linking one rural area with another. It also reflects the fact
-haL the urban poor typically have greater political power than the rural poor.
lius, increased food production from large farmers is not as likely to reach the
.ural poor as it is the urban poor (or if it does, the price may not be much lower)..
similarly among the rural poor, the submarginals are most likely to be excluded
rom the food distribution networks.
Given the important relationship between employment and income, an equity
Sise for a large farmer strategy would exist if large farmers made intensive use
)f hired labor. Available evidence does not support this conjecture, however,
Farm management studies show that not only do small farmers use more labor
per acre initially, but, when large farmers are provided profit increasing
inputs, they tend to use less rather than more labor per acre (by investing
in labor displacing equipment -- often made artificially attractive through
subsidized credit and import advantages). Also, there is much controversy
about the yield-increasing contribution of mechanisation. Many studies
showing increased yields from tractorization, for example, have not adjusted
for increased use of other inputs. It is rather clear, however, that
the net effects of mechanisation on employment and income distribution have
In sum, while conclusive quantitative evidence of equity effects
does not exist, a priori analysis of the various components of equity
impact and some historical evidence suggest that, although there are
potentially favorable indirect equity effects in the short-run, a large
farmer strategy will not have a favorable equity impact in the long run,
in either the direct or the indirect sense. The small farmer strategy,
on the other hand, clearly has a favorable direct equity impact since
the beneficiaries are members of the poor majority. The higher the
proportion of the poor majority accounted for by small farmers, the
greater the direct equity impact of the small farmer strategy.
If such a strategy included land reform, the impact would be even greater.
In addition, the long-run indirect equity impact of a small farmer strategy
should also be favorable. As discussed below, the small farmer strategy
appears to contribute more to sustained rural income generation in the
*See ref. 8, ch. 5, and ref 22 for evidence and further discussion.
D. Impact on Key Factors in Rural/National Development
The contribution of alternative producer strategies to food
production and equity objectives can be assessed at 3 given point in
time; however we must also be concerned with the establishment of a
self sustaining development process. This section discusses the probable
impact of these strategies on key factors in overall rural and national
development, namely on (1) the level of demand for domestic goods,
(2) the labor intensity of the prevailing consumption pattern, (3) savings
and (4) demographic trends.* The evidence suggests that the optimum
strategy may/depend on the time frame. In the short run, the high savings
rates and the large marketable surplus of large farmers/ As small farmer
incomes rise above the subsistence level,however, both savings and
marketable surplus will increase, thus eroding the principle arguments
for the large farmer approach and enhancing the superiority of the small
farmer strategy as a means of obtaining increased rural income generation
and sustained national growth.
Deficient demand, particularly in smaller countries, is recognized
as a major deterrent to economic growth in some LDCs. For a given level
of national income, the overall level of demand for domestic goods will
be determined by the amount of income spent at home versus abroad.
Typically, upper income agricultural producers have a higher marginal
propensity to import -- to satisfy both consumption and investment demand --
than lower income agricultural producers; thus, a small farmer strategy
is likely to generate more value-added activity in the domestic non-agricultural
sector than a large farmer strategy. Assuming equal export performance,
the balance of payments effect of the small farmer strategy would also
*Another key factor might be education of the labor force. A small farmer
strategy would appear superior here because of the broader participation
By contrast, because of the higher "reservation demand" of small
larnmers for their own f.rm products, a smaller proportion of their output
is actually sold on the market. This means that the income generated
through ancillary activities such as transportation, handling, and
processing is less, as is small farmer cash expenditures on other goods.
The smaller marketable surplus tends to be a temporary phenomenon, however;
research in the Philippines, for example, shows that annual per capital
consumption of rice rises with income up to about 264 kilograms ofpaddy
and then levels off.* Beyond this point, reservation demand declines and
marketable surplus rises.**
Another important factor in the establishment of self sustaining
growth in labor surplus economies is the labor-intensity of the prevailing
consumption patterns. The labor-intensity of the goods demanded is important
because of the employment generating effects, the higher propensity to
consume out of income to labor than out of income to capital, and the need
toconserve scarce capital. Most studies show that, except for the very low
income groups which consume foodgrains (which might be produced with capital
intensive methods), the labor-intensity of average consumption patterns
decreases with income.*** This implies that a small farmer strategy would
promote a more labor-intensive consumption pattern than a large farmer
*See ref 23 p. 73.
**Ref 17is a recent study of the effects of alternative farm sizes (small,
medium, and large) in the U.S. According to that study, the small-farm
structure generates an estimated 30 percent more in off-farm income,
primarily in the areas of agribusiness and consumer goods and services,
than the large-farm structure. Of course in the U.S., the typical small
farm is perhaps 100 hectares where it might be 1 or 2 hectares in the
LDCs. No comparable LDC studies are known.
**See rcfs 5, 27, 33, and 35.
Regarding savings, recent studies of small farmers in Korea, Taiwan,
Bangladesh, and Zambia show that their marginal private savings rates can be
as high as 50 percent if the incentives to save -- generally considered
to involve a profitable technology plus attractive deposit rates and
convenient savings institutions -- exist.* Of course, average savings
rates of large farmers are likely to be even higher than those of
small farmers, although the productivity of large farmers' capital
investments may be lower. On the other hand, heavier'public taxation
cf the agricultural surplus for development purposes is considered
more politically feasible under a small farmer approach than under a
continuing emphasis on large farmers since the latter are in a stronger
position than the former to exert restraining influence on tax rates
and, in addition, on the distribution of government expenditures. If
these behavior patterns are typical, reinvestment in the agricultural
sector could, on balance, generate as high a growth rate under a small
farmer approach as under a large farmer approach. One should note, however,
that the incentive and ability to save by tenant farmers or share croppers --
not to mention landless agricultural workers -- is considerably less than
for the owner-operator; thus the savings/investment potential will vary
with tenure patterns.
Finally,regarding demographic trends, migration to cities should be
discouraged by a small farmer strategy to the extent that the latter leads
to a relatively higher standard of living for the poor in the tural areas.
The increase in urban overhead costs associated with relieving congestion
and extending existing urban service levels to new migrants may well render
the provision of urban services more costly in the long run than the
*See ref 3
provision of comparable rural services. In addition, rising incomes
among a broad spectrum of rural families, when accompanied by other
modernizing influences -- such as education, mass communications, and
increased availability of consumer durables -- appears to be accompanied
by a decline in fertility among those families.* Given limited non-human
resources, this spontaneous decline in fertility permits per capital
incomes, hence standards of living, to rise faster than would be other-
IV. Conclusions and AID Program/Policy Implications
In summarizing the preceding discussion it is perhaps useful to begin
with the equity objective. While something of a case can be made for a
large farmer strategy serving the equity objective, it is a shaky one,
especially when the two following considerations are taken into account.
First, it may be more difficult (and costly) to shift to a small farmer
strategy in the future after programs and institutions biased toward
large farmers have been strengthened and after real incomes and production
incentives of small farmers have been reduced -- than to begin with a
small farmer strategy in the first place. Second, the case fcr the large
farmer strategy serving the equity objective is an indirect one based
essentially on its effect on lowering food prices to the consumer; however,
the price decline may not occur or may occur with some delay. The indirect
case is further weakened when the previously discussed problems entailed
by food distribution to the rural poor are taken into account. It
*See ref 24
follows that if a small farmer strategy has a roughly equivalent impact
on food production, it must *- since small farmers are the primary bene-
ficiaries -- have a superior impact on equity.
Thus, the assessment of the small farmer strategy in meeting the
equity objective rests in part on Its success in meeting the food production
objective. The main conclusion from the analysis on relative productivities
is that the choice of a small or large farmer strategy will depend largely
upon the time period for achieving the food production objective. Consideration
of the strategy choice impact on key factors in rural and national development
reinforces this conclusion. The shorter the time period, or the more
urgent the food crisis in a given country, the less appropriate strategies
at the small farmer end of the spectrum are likely to be.
But this conclusion rests on two important assumptions. First, it
rests on the assumption that existing physical and institutional infrastructure
are seriously deficient for small farmers and/or a profitable technology
does not exist in a readily usable form. To the extent that this is not
true, the time period becomes less of a constraint. Second, it rests on
the assumption that there are not alternative ways of meeting a short run
food crisis. Countries with high-cost domestic production and/or limited
agricultural resources might find it more economical to meet'expanded needs
through imports than through expanding domestic output. Thus, increases
in these cases
in domestic food production can be seen/as a long-run objective, not a
short-run requirement. Also, a relatively small country in a favorable
foreign exchange situation has the option of importing some food and,
assuming it has some physical potential for food production, can accordingly
be more free to pursue a small farmer strategy even in the short run than
a large country in an unfavorable foreign exchange situation. Even the
latter type of country could be provided some "degrees of freedom", however,
:. pur,. .. t. A trmei slrnteg\ in the hort run, h 1 eing provided foo~
aid until growth in .lomeAtic fnnol prndu'tit.n rprArh.q nn Arcptahle levl.
Ilh pirecetlding analysis suggests that, to the extent possible, AID
should support major reforms which might allow a more rapid attainment
of both food production and equity objectives. Where such reforms are
not in progress, a small farmer strategy better meets a combined food
production and equity objective. Thus, as is consistent with the Congressional
Mandate, AID should concentrate its assistance to agricultural producers
in the small farm sector. The implications for AID programs and policies
arising in this context are even stronger if we keep two points in mind.
The first is that other aid donors and LDC governments will continue to
finance generally unfocused agricultural production programs which do not
have a target group orientation; and the second is that, although small
farms may occupy a small proportion of total agricultural land (as small
Sas 20 percent according to Table 1), AID could not expect to affect a
much larger area in any case, in view of the Agency's limited development
budget. Thus, AID can focus on small farmers while other resources are
allocated more broadly within the sector. It is also worth noting that
a small farm strategy appears to better ensure the participation of women
as decision-makers. Individual country studies in Africa, for example,
show a significant number of female heads of farm in small farm areas
where one or more males seek supplementary income through off-farm
At the agricultural project and program level, the weight of the
Congressional Mandate and the evidence discussed in this paper would appear
sufficient, notwithstanding the various qualifications, to suggest that the
Agency should, indeed, require agricultural production projects to be
small-farmer oriented (some elements of such a strategy may also benefit
Of course the ease of implementation will vary by country. There may be
some countries, for example, which appear to have only small producers.
Must we establish a cut-off point? There may be other countries where
AID is financing large resource transfers. Is it possible to program
large resources in the small farm sector in short time periods? AID
must find reasonable answers to these questions which uphold the spirit
of the Congressional Mandate.
Another obvious policy implication for AID is that carefully programmed
PL 480 food aid -- certainly Title II and perhaps Title I -- be provided
to those countries which, of course, have a need for the food aid for both
domestic and balance of payments reasons, but also where AID is collaborating
in mounting a small farmer-oriented development strategy. A major concern
with food aid programs has been possible disincentive effects on domestic
production. But if food aid is linked to a major small farmer oriented
food production strategy, including provision of relevant physical
infrastructure through labor-intensive rural works programs, which would
increase the demand for food, prices and incentives for producers could
It is recognized that small farmer assistance is more difficult to
program and will therefore require maximum programming flexibility, including
local currency use in DLs, freedom to choose the most appropriate inputs and
outputs, more use of technical assistance funds for local hire and local
procurement, and rapid response and assistance from AID/W. A first
operational step in pursuing this policy would be for each Mission to define
more precisely, ideally in collaboration with the host government, the "small
/target groups appropriate to that country, employing some of the notions
discussed in Section III A above suitably amended to fit individual country
situations. This definition would be employed as one criterion for identifying,
designing and approving agricultural project proposals.
Additional research, analysis of existing research, and guidance is clearly
in order. In devising specific strategy options attention should be given
to decreasing small farmer costs'of production relative to those of the large
farmer. This can best be accomplished through a combination of small farmer
oriented agricultural research and farm management studies for typical small
farms. We need in particular to address the difficult problem of helping
"submarginal" small farmers to increase their food production in cost-effective
ways. Although the limited size and/or quality of their land holdings may preclude
their ever being full-time farmers, and their inaccessibility by reason of
location or fragmented holdings may make the cost of reaching them directly with
small farmer institutions prohibitive, financial and technical assistance on
pricing policies, markets, input supplies and research could help them, as well
as more "viable" small farmers, substantially increase food production. The
problem of submarginal farmers takes on particular importance in view of the
fact that they, along with completely landless laborers (who constitute 30
per cent or more of the active agricultural population in some countries),
constitute the poorest and worst-nourished of the poor majority, and that their
position, if not taken explicitly into account, could be worsened. Small farmer
projects should be carefully designed to facilitate the participation of submarginals.
List of References*
1. AID/PPC. A-263, 4-30-75, Congressional Mandate and attached paper,
"The Congressional Mandate: Aiding the Poor Majority."
2. AID/TAB. "Small Producers Economics" (Rough draft of AID paper).
3. AID. Spring Review of Small Farmer Credit (1973). Papers by Dale
Adams and Richard Roberts.
4. Bachman, Kenneth L. and Raymond P. Christensen. "The Economics of
Farm Size" in Southworth and Johnston, editors, Agricultural Development
and Economic Growth, 1967.
5. Ballentine, Gregory, and Ronald Soligo. "Consumption and Earnings Patterns,
and Income Redistribution: The Case of Colombia: (unpublished draft, 1974).
6. Berry, Albert. "Cross Country Evidence on Farm Size/Factor Productivity
Relationships," (unpublished paper presented at The Purdue Workshop
on Empirical Studies on Small Farm Agriculture in Developing Nations,
Purdue University, November 1972).
7. Berry, R.A. "Farm Size and Factor Productivity in U.S. Agriculture"
(unpublished draft, January 1973).
8. Bharadwaj, Krishna. Production Conditions in Indian Agriculture: A
Study Based on Farm Management Surveys. University of Cambridge,
Department of Applied Economics, Occasional Paper No. 35. (London
and New York: Cambridge University Press, 1974).
9. Blair, Harry W. "The Elusiveness of Equity: Institutional Approaches
to Rural Development In Bangladesh", Cornell University Special Series
on Rural Local Government, November 1974.
10. Castillo, Gelia T. "Diversity in Unity: The Social Component of Changes
in Rice Farming in Asian Villages" in Changes in Rice Farming in Selected
Areas of Asia, International Rice Research Institute, 1975.
11. Chenery, Hollis et al. Redistribution with Growth, 1974.
12. Dalrymple, Dana and William I. Jones. "Evaluating the Green Revoluticn"
13. Dorner, Peter. Land Reform and Economic Development, 1972.
*There is an extensive literature on most of the topics discussed in this paper.
The "List of References" is just that. It does not constitute a selected,
much less an exhaustive, bibliography. An effort has been made, however,
to refer to (1)recent research and (2) publications of particular interest
to AID as well as to (3)summary material.
14. Ford Foundatuon. "Employment in Developing Countries: (July 1974).
15. Crlff I K.e th. 'lhe Political Pconomy of Agrarian Change: An Essay
on the (;reen Revolution (1974)
16. Harberger, Arnold C. "Fiscal Policy and Income Redistribution"
(unpublished draft for AID Princetor-Brookings Contract, 1974).
17. Heady, Earl 0. and Steven T. Sonka. "Farm Size, Rural Community Income,
and Consumer Welfare", American Journal of Agricultural Economics,
18. Herz, Barbara. Demographic Pressure and Economic Change: The Case of
Kenyan Land Reforms. (unpublished Ph.D. dissertation, Yale University,
19. House of Representatives, Foreign Affairs Committee, Report on Foreign
Assistance Act of 1974. House Report No. 93-1471, October 25, 1974.
20. House of Representatives Bill H.R. 9005 (August, 1975) and Committee on
International Relations Report, "International Development and Food
Assistance Act of 1975" (Report No. 94-442).
21. IBRD. Rural Development: Sector Policy Paper (February 1975)
22. International Labor Office. Mechanization.and Employment in Agriculture,
23. IRRI Research Highlights for 1974, 1975.
24. Kocher, James E. "Rural Development, Income Distribution and Fertility
Decline," published by the Population Council, 1973.
25. Krishna, Raj. "Agricultural Price Policy and Economic Development in
Southworth and Johnston, editors Agricultural Development and Economic
26. Lau, L.J. and P. A. Yotopoulos, "A Test for Relative Efficiency and
Application to India Agriculture", American Economic Review (March
27. Mellor, John and Uma J. Lele. "Growth Linkages of the New Foodgrain
Technologies." AID Employment and Income Distribution Project -
Occasional Paper No. 50.
28. Miller, James C. "Regional Devel opment: A Review of the State-of-
the-Art" TAB/UD, Agency for International Development (August 1974)
29. Mosher, Arthur. Creating a Progressive Rural Structure, 1969, and
To Create a Modern Agriculture, 1971.
30.' Owens, Edgar, and Robert Shaw. Development Reconsidered: Bridging
the Gap Between Government and People (1972).
31. Sen, Amartya. Employment Technology and Development, 1975.
32. Sidim,, 'urjit S. "Relative Efficiency in Wheat Production in the Indian
Punjab", American Economic Review (September, 1974).
33. Soligo, Ronald. "Factor Intensity of Consumption Patterns, Income
Distribution and Employment Growth in Pakistan", Rice University
Program of Development Studies, Discussion Paper No. 44 (1973).
34. Statement of Edgar Owens, House Foreign Affairs Committee Sub-Committees
on International Organizations and Foreign Economic Policy (September
35. Summan, Tucay M. "Short-run Effects of Income Distribution on Some
Macro-Economic Variables: The Case of Turkey" Rice University Program
of Development Studies, Discussion Paper No. 46 (1973).
36. USDA, Economic Research Service, Foreign Agricultural Economic Report
No. 98, The World Food Situation and Prospects to 1985 (1974).
37. World Food Conference (November 1974). Draft Resolutions.