World Development Vol. 38, No. 10, pp.

13491361, 2010
2010 Elsevier Ltd. All rights reserved
0305-750X/$ - see front matter
www.elsevier.com/locate/worlddev

doi:10.1016/j.worlddev.2009.06.012

The Future of Small Farms: Trajectories and Policy Priorities

PETER HAZELL, COLIN POULTON *
University of London, UK
STEVE WIGGINS
Overseas Development Institute, London, UK

and
ANDREW DORWARD *
University of London, UK
Summary. While smallholder development has, in the past, led to reductions in poverty and hunger, does this still apply in todays
more globalized world? This paper reviews the debates on the contemporary role of agriculture in development and the case for small
farms in light of the rise of supermarkets, lower commodity prices and liberalized trade, agricultural research funding, environmental
change, HIV/AIDS, and changing policy ideas. Although the answers vary greatly by context, for many low-income countries, smallholder development remains a key option. The policy agenda, however, has changed. In addition to providing public goods, the growing
challenge is to overcome market failures, which is largely a matter of institutional innovation.
2010 Elsevier Ltd. All rights reserved.
Key words small farm, growth, poverty

1. INTRODUCTION

And for those that do not, what alternative opportunities

can be created, and what policy interventions are needed to
help manage the transition to fewer and larger farms while
avoiding worsening poverty and social inequalities at regional
and household levels?
This paper addresses these issues. It is based on a literature
review and the deliberations of an international workshop on
the future of small farms convened at Imperial Colleges campus at Wye, Kent, England, in June 2005 by the International
Food Policy Research Institute (IFPRI), Imperial College,
London, and the Overseas Development Institute (ODI).
The rest of this paper consists of four sections dealing with,
in turn, the role of agriculture as a whole, the case for small
farm development, policy options, and a short conclusion.

Of the developing worlds three billion rural people, over

two-thirds reside on small farms of less than two hectares; there
are nearly 500 million small farms. These people include half of
the worlds undernourished people, three-quarters of Africas
malnourished children, and the majority of people living in
absolute poverty (IFPRI, 2005). Despite recurring predictions
that small farms will soon disappear, they have proved remarkably persistent. Indeed, an increasing part of agricultural land
in the developing world is being operated in small farms. The
importance of farming in household incomes may have declined, but the number of rural households that use farming
as a platform for their livelihood strategies continues to grow.
Agricultural growth that improves productivity on small
farms has proven to be highly eective in slashing poverty
and hunger and raising rural living standards, as demonstrated
in large parts of Asia during the green revolution. Moreover,
most of the countries that have failed to launch an agricultural
revolution remain trapped in poverty, hunger, and economic
stagnation. Even so, the conventional conclusion that developing countries should continue to invest in their agricultural
development and in small farms, in particular, is being challenged. The world has changed since the 1960s when the green
revolution started. Liberalized international trade and falling
prices for agricultural produce on world markets; the vigorous
entry of supermarket chains in some developing countries and
their exacting demands for quality, consistency, and timeliness
from potential suppliers; the scourge of HIV/AIDS; and
mounting pressure on natural resources from population
growth, all pose serious challenges to the viability of smallscale farming.
Just how serious are these threats to small farms? Under
what conditions do small farms help generate incomes above
the poverty line with little risk of slipping back into poverty?

2. THE ROLE OF AGRICULTURE

Questions about the future of small farms are embedded
within wider debates on the roles of agriculture in development, so these need to be reviewed before looking at the specic case of small farms.
Historically, few countries have ever achieved rapid economic growth at the early stages of development without a
substantial growth in agriculture. Circumstances change, however, calling into question whether substantial agricultural
* We are grateful for the comments from anonymous referees and from
Johann Kirsten. The workshop on Future of Small Farms was made
possible by grants from the United Kingdom Department for International Development, the International Food Policy Research Institute,
and USAID. Neither the commentators nor the sponsors are responsible
for the opinions and interpretations presented here, for which the authors have sole responsibility. Final revision accepted: June 23, 2009.
1349

1350

WORLD DEVELOPMENT
Table 1. Summary of the debate about the role of agriculture in development

Type of argument

Case for agriculture

Case against agriculture

Engine of growth

Agriculture is large enough in many countries that its

Agriculture has become a relatively small sector in
growth can make a real dierence in rural living standards. successfully growing countries, and other faster growing
Moreover, agriculture has powerful growth linkage eects sectors should now be prioritized. In many poor countries
on the rest of the economy, including providing a growing
the low productivity of agriculture and unfavorable
demand for nascent industries
market prospects undermine its potential. Moreover, the
growth linkages in agriculture are weaker in todays
liberalized economies and may not be any larger than the
linkages associated with labor-intensive manufacturing
and services
Alternatives to agriculture Many poor countries do not have viable alternatives to
Trade liberalization and FDI have opened up new
agriculture. They have few minerals to export, their
opportunities for developing countries to become early
manufacturing sectors are small and internationally
exporters of manufactured goods and some services and to
uncompetitive, and their service sectors are demand
rely more on low cost food imports
constrained
Technical feasibility
Modern science is opening up new opportunities to
The best technology opportunities have already been
increase agricultural productivity, even in countries and exploited, and agricultural research now faces diminishing
regions that have not beneted much from new
returns in the better agricultural areas and costly and risky
technologies in the past
prospects in lagging regions. Modern intensive farming
also leads to environmental degradation in many
developing countries. The shift toward private funding of
research means that problems of poor farmers are less
likely to receive priority
Poverty impact
Agricultural growth has proven to be powerfully pro-poor
Changes in market systems mean that there are limited
when based on small farms and the products they grow, market opportunities for small farms today, and the prices
especially food staples
of the products they grow are at historic lows. The
combination of lower prices and smaller farm sizes reduces
the impact of direct poverty on agricultural intensication.
The rural poor have also diversied away from agriculture
as their main source of livelihood. Commercial farms and
high value market chains oer better prospects for creating
employment and reducing poverty
Policy environment
Structural adjustment programs have removed the worst
There is no tolerance today for the kinds of big public
of the biases against agriculture and opened the way for
spending on agriculture, including subsidies, that were
more successful agricultural investments
characterized by the green revolution. Many countries also
lack the governance and administrative capability to
implement ambitious agricultural development programs

growth is necessary for development. Key arguments made for

and against agriculture as a central plank in development are
summarized in Table 1.
Some of the dierences in views can be resolved by recognizing that there is no single role for agriculture, and country
context plays an important role in dening opportunities, constraints, and hence roles for the agricultural sector. A key
dimension of context is stage of development. With increasing
economic growth and urbanization, agriculture as a sector declines in relative importance. Arguments for prioritizing agriculture, especially smallholder farming, correspondingly
weaken.

Our main concern is those low-income countries that are

still far from transforming their economies, where the majority
of people live in rural areas, and agriculture remains the largest single source of employment. Two variables aect considerations in these early stages of development: the
opportunities for other sectors to provide growth and jobs
and the potential for agricultural development. Table 2 summarizes how the role of agriculture depends on dierent circumstances.
Countries with minerals may be able to earn signicant export revenues and government income without agricultural
development. When there is potential for farming, it may be

Table 2. The role of agriculture during the early stages of development, by country context
Agricultural potential

Nonagricultural engines of growth

Minerals

High

Low

Manufacturing

Agriculture as a secondary growth

Agricultural growth can speed up
sector and a means of spreading the manufacturing development by freeing up
benets from minerals to a broad labor and capital, reducing food costs and
rural base
supplying raw materials for agriculture
based industries
Means of spreading the benets from
Subsistence for the rural poor
minerals to a broad rural base

Limited
Lead sector for growth and poverty
reduction

Overall economic prospects bleak, but

exploitation of niche agricultural
opportunities important for growth
Subsistence for the rural poor

THE FUTURE OF SMALL FARMS: TRAJECTORIES AND POLICY PRIORITIES

possible to invest mineral revenues in roads, irrigation and

drainage, research, and extension to promote a competitive
farm sector despite exchange rates buoyed by mineral
revenues. A good example is Indonesia where oil earnings
allowed heavy public investment in agricultural and rural
development. On the other hand, when potential is poor, agriculture will remain extensive, functioning as a subsistence reserve for those on the land.
Some countries that can reach international markets at low
cost may have prospects for export-oriented industries at an
early stage. It is still likely that agriculture will play an important part in their development as an initial source of capital,
foreign exchange, and labor. Moreover, the early stages of
manufacturing may be based on processing farm production.
In countries with low agricultural potential, agriculture will
inevitably play a smaller role, particularly if there are minerals
or potential for export manufacturing or tourism. The most
challenging cases are countries that have low agricultural
potential, no minerals, and limited prospects for alternative
growth sectors. Agriculture in these countries is likely to be
rst and foremost a subsistence reserve where the poor can
build livelihoods with little dependency on the state particularly when land is distributed equitably. Even here, some
may be competitive, at least on the domestic market. Even
where land resources are generally poor, some pockets of land
with reasonable soil and a water supply exist. Prominent
examples here are Sahelian countries that have established
themselves as major cotton exporters in the past two decades,
as well as developing a modest level of irrigated rice production.
The extent to which agricultural growth will be pro-poor
also depends on context, the major distinction being land distribution: relatively even (unimodal) or relatively unequal with
marked dierences between large and small farms (bimodal).
When land is distributed relatively evenly, agricultural growth
can be powerfully pro-poor. It not only raises small farm incomes and employment but also contributes to lower food
prices and generates strong growth linkages in the nonfarm
economy, which in turn help the poor.
Asias green revolution demonstrated how agricultural
growth that reaches large numbers of small farms could
transform rural economies and raise enormous numbers of
people out of poverty (Rosegrant & Hazell, 2000). Recent
studies also show that a more egalitarian distribution of land
not only leads to higher economic growth but also helps ensure that the growth that is achieved is more benecial to the
poor (e.g., Deininger & Squire, 1998; Ravallion & Datt,
2002). 1
In contrast, agricultural growth has proven much less propoor in countries that began with an inequitable distribution
of land. Good examples of this case can be seen in many parts
of Latin America, South Africa, and Zimbabwe.
(a) Recent changes
Rapid growth in international agricultural trade, low world
prices, and increasing competition in agriculture around the
world are making it more dicult for farmers in countries with
poorly developed agricultural sectors to compete. The pressure
on developing country farmers is exacerbated by the hefty subsidies that farmers receive in most OECD countries. 2 In this
environment, some analysts, such as Maxwell and Ellis, ask
if it is realistic to continue to prioritize agriculture in poor
countries (Ellis & Harris, 2004; Maxwell, Urey, & Ashley,
2001), an especially pertinent question for countries lacking
either minerals or manufacturing potential, as applies in much

1351

of sub-Saharan Africa. Four reasons for no longer prioritizing

agriculture in poor countries are put forward.
First, agriculture in many poor countries, and especially in
Africa, has fallen so far behind the rest of the world in productivity that it would be very dicult and expensive to bring it
up to levels at which it could compete in the market at todays
historically low prices. 3 Rather than prioritizing agriculture,
countries might do better by taking advantage of trade liberalization and private sector capital ows (FDI) to develop new
industries and rely on food imports as needed.
Second, linkages from agricultural growth are much weaker
in todays more open economies, especially in small countries.
For example, when imports can enter freely, food prices will
be determined more by border prices than domestic agricultural production, and industry can sell directly into foreign
markets without having to wait for growth in domestic demand.
Third, the rural poor have already diversied away from
farming, making agricultural growth less important for poverty reduction.
Finally, there is no tolerance today for the kinds of big public spending on agriculture, including subsidies that characterized the green revolution. Many countries also lack the
governance and administrative capability to implement ambitious agricultural development programs.
There are three responses to these observations. One, what
alternatives to agriculture exist in the early stages of development? Other sectors are not necessarily easier options.
Launching manufacturing exports is challenging when countries such as India and China are ooding world markets with
cheap goods, especially for countries that face high transport
costs, cannot attract much FDI, or both. Reviewing the prospects for economic growth in Africa, Fafchamps, Teal, and
Toye (2001) favor manufacturing owing to its potential to
grow much more quickly than agriculture, but they recognize
that only a few countries in Africa have the conditions where
rapid growth of manufacturing on a substantial scale can be
expected in the short term.
If not manufacturing, then what about services, a sector that
is growing quite rapidly in many developing countries? Most
services depend on domestic demand, so unless some other
sector is growing, then new service sector jobs are likely to
be low productivity activities that simply supplement, rather
than replace, existing incomes, which Lipton (2004) calls jobs
of distress.
Two, there are costs to not developing agriculture. Although
open international trade should allow some countries to obtain food and raw materials at low cost no matter what the
performance of their agriculture, this does not apply for three
important groups of poor countries: the half dozen or so most
populous developing countries whose total food needs dwarf
world trade volumes such that even relatively modest production shortfalls could lead to large increases in world prices;
landlocked countries that face high transport costs; and countries with low foreign exchange earnings that can ill aord to
divert these earnings away from essential imports and capital
goods to food that could be grown at home.
More important, to ignore farming in the absence of other
opportunities is to condemn the rural majority to poverty.
This may in turn lead to heavy expenditure on welfare; protection of the very poor and destitute in rural areas can be an
expensive business. It may also invite political instability arising from social inequalities.
Three, not all changes are negative. Some facilitate agricultural development. Modern science is opening up new opportunities to increase agricultural productivity, even in countries

1352

WORLD DEVELOPMENT

and regions that have not beneted much from new technologies in the past. Developments in information technology and
energy generation can overcome some of the constraints of
poor infrastructure. As a result of structural adjustment and
economic liberalization, many countries have created conditions in which the private sector and civil society can play a
greater role in agricultural development than in the past, thus
reducing the burden on the state.
3. THE CASE FOR AND AGAINST SMALL FARMS
(a) The case for small farms
The case for setting small farms at the center of agricultural development can be made on two grounds: eciency
and equity.
Regarding eciency, studies have often reported an inverse
relationship between farm size and production per unit of
land. This shows a common tendency for larger farms to yield
lower gross and net returns per hectare of land per year than
smaller farms. These results are generally strongest in Asia
where land is scarce compared to labor. 4 The explanation
may be found in the dierences in transaction costs faced by
small and large farms for dierent operations, as summarized
in Table 3.
The varying transaction costs stem from asymmetric information in rural markets for labor, credit, and insurance, in
turn originating in the material conditions of agricultural production that typically give rise to specic institutions designed
to mitigate the imperfections (Binswanger & Rosenzweig,
1986). Of particular relevance is the high cost of hiring labor
when the costs of selection of workers and their supervision
are added to those of wages. The household farm with labor
that largely supervises itself is the institutional response to
these costs. As a result, most farms operate at the scale of
the family. In the absence of machinery, the size of the farm
will be small. 5
The implication of this is that when labor costs are an
important part of agricultural costs, small farms may have signicant advantages over larger units. Conversely, once agriculture becomes more intensive in transactions beyond the
farm gate, for example, buying substantial quantities of input
and selling most of the output, larger farms may have the
advantage. Thus, small farms have the edge for less technologically advanced agriculture with low labor costs, but as an
economy develops with increased use of capital intensive technology and hired labor, the advantage shifts to larger farms. 6
Table 3. Transaction cost advantages of small and large farms Source:
Poulton et al., 2005.

Unskilled labor supervision, motivation, etc.

Local knowledge
Food purchases and risk (subsistence)
Skilled labor
Market knowledge
Technical knowledge
Input purchase
Finance and capital
Land
Output markets
Product traceability and quality assurance
Risk management

Small
farms
p
p
p

Large
farms

p
p
p
p
p
p
p
p
p

Has economic development tipped the scales from small

to large farms? Apparently not yet, in most countries, judging by the evidence of the decline in farm sizes in the developing world (Lipton, 2005). While farm sizes have been
rising in most OECD countries over recent decades, they
have fallen in many developing countries (see Figure 1).
While the declining farm size reects population growth
and the sub-division of land holding, it should not happen
if there were economies of scale in farming, since then it
would make sense for small farmers to rent out plots to larger operators. However, as Lipton comments, such cases are
rare. Most tenancy involves letting out parts of larger farms
to smaller operators.
An alternative explanation is that land markets are imperfect. Selling prices for land may be inated well above the discounted value of the future production on account of the value
of the land as collateral against bank credit, the social prestige
of land ownership, or the expectation that land prices will
raise. People may also retain and manage their own farms
rather than renting them out for cultural reasons (Singh,
2005). Both large and small landowners may also be reluctant
to rent out elds for fear of not being able to regain their land
promptly, if ever. Moreover, in situations where households
with small plots of land have few, if any, alternatives to farming, small farms may be operated for want of better options,
regardless of considerations of scale.
Thus, the declining average farm size in developing countries may not demonstrate any superior economic eciency
of small farms. It does, however, indicate that even tiny landholdings remain a valued component of a diversied livelihood
in the presence of highly imperfect land, labor, and capital
markets.
Regarding equity and poverty reduction, there is a strong
case for preferring small to large farms. Small farms are typically operated by poor people who use much labor from both
their own households and their (equally or more) poor neighbors. Many farm surveys have shown that the smaller the
holding, the more labor per unit area is applied ( Cornia,
1985; Heltberg, 1998). If there were no transaction costs in labor markets, this would not happen, but given the costs of
supervising hired labor, larger farmers tend to employ fewer
workers than would otherwise be optimal.
Moreover, small farm households have more favorable
expenditure patterns for promoting growth of the local nonfarm economy, including rural towns. They spend higher
shares of incremental income on rural nontradables than large
farms (Hazell & Roell, 1983; Mellor, 1976), thereby creating
additional demand for the many labor-intensive goods and
services produced in local villages and towns.
Notwithstanding the advantages of small farms in developing countries, policy has often favored large farms,
through access to subsidized credit, protection for the output of such farms, and infrastructure provision in areas of
large farms, among other measures. 7 Policy makers have
often seen large farms as modern, technically advanced,
and ecient, a view reinforced by large-scale farmers themselves who are often better organized to lobby for public
support. In some cases, the large farms are state enterprises. Small farms have persisted more often than not despite such biases.
(b) Changes and threats to small farmers
The above-mentioned arguments are well known and widely
accepted. What concerns some observers (see, e.g., Ellis 2005;
Maxwell 2003) is that in a changing world, the prospects for

THE FUTURE OF SMALL FARMS: TRAJECTORIES AND POLICY PRIORITIES

1353

4
3.5
3
hectare

Median farm size, Africa & Asia

2.5
2
1.5
1
0.5

Th

19
80
19
91
rke
y2
00
1

ista
Pa
k

Ko
re

Tu

19

19
78
1
99
aila
3
nd
20
03

80
19
89
n2
00
0

02

82

l2
0

19

Ne

Re
a,

on
Ind

pa

90

p.

19

70
19
80

es

ia

19

19

73
19
93

71
19
77
Ind
ia 1991
19
95
/96

19

69
99
3

19

i1

law
Ma

00

99

/20

70

19
Le

so

tho

pia
Eth
io

ng
Co

90

19

19

/92
01
/02
20

77
19

89
19

oD

19
70
R1
99
0

12

Median farm size, Latin America

hect are

10
8
6
4
2

01

88

20

19

94

bia
lom
Co

72

19
ru
Pe

y1
ua
rag

Pa

19

99

81

20

19

01

91
19

ma
Pa

na

81
19

71
19

96
19

85

Bra

zil

19

80
19

19

70

Figure 1. Median farm sizes in the developing world. Source: FAO Data from agricultural censuses.

smallholders are deteriorating. Contemporary challenges include changing production methods and increased concentration in supply chains; low world prices and more open markets
to international competition; changes in agricultural research;
environmental degradation and climate change; the impact of
HIV/AIDS; and changes in the policy environment. 8
(i) Changing production methods and greater market concentration
Green revolution technology centered on seeds was largely
scale neutral. Small farmers could participate, especially as
new rounds of crop breeding made the modern varieties less
variable in yield and thus less risky. 9 Where new technologies
require access to information, machinery, or higher capital input, and many high value crops require considerable up-front
cash investment in seeds, fertilizers, and pesticides, small-scale
farmers may be disadvantaged since they cannot obtain credit
and input on the same terms as larger operators.
Of more concern are changes to marketing chains. Supermarket operators or their agents are becoming increasingly
important in parts of the developing world, especially in Latin
America. Buying power is being concentrated in a few hands.
The supermarkets require stricter standards for the quality,
consistency, and timeliness of supply. They may also need to
trace consignments back to source and to arm the conditions
under which they have been produced, in terms of pesticide
applications, organic cultivation, use of child labor, or animal
welfare. Meeting the requirements for these credence characteristics, which cannot be proved by examining the produce,

can be particularly onerous for smallholders. Auditing and

certication costs have strong economies of scale (Raynolds,
2004).
Supermarkets also often require suppliers to adjust rapidly
to changing consumer demands with new investments in
equipment and knowledge. Small-scale, under-capitalized,
and often under-educated farmers nd it particularly dicult
to meet the quantity, timeliness, traceability, and exibility
requirements of the new supply chains even if family labor is
well suited to delivering quality products.
By and large, smallholders have yet to nd widely replicable
institutional solutions to the new demands (Boselie et al.,
2003; Reardon & Timmer, 2007).
The importance of this to smallholder farmers depends on
three considerations. One is how quickly supermarkets are
capturing the marketing chains, particularly those large
domestic channels that deliver food to households of modest
means. This seems to have happened rapidly in Latin America
and parts of South-East Asia, and there are signs of the same
events occurring in China. It appears to be a more patchy and
slower process in Africa and South Asia. Traill (2006) found
that increased incomes, urbanization, and income inequality
tend to increase the supermarket share of food sales. The advance of supermarkets will probably thus continue to be rapid
where they have gained a signicant foothold, that is, in the
industrializing and middle income countries of East and
South-East Asia and Latin America. In other regions, and
above all in Africa and South Asia, however, the advance
may be quite slow, and hence the threat to small farms may

1354

WORLD DEVELOPMENT

be limited in the near future. 10 That said, it is devilishly dicult to predict such changes since key processes are discontinuous and nonlinear.
Another question is whether the supermarket buyers have
alternatives to dealing with smallholders. In cases of bimodal
land distribution, the buyers may be able to obtain the supplies they need from a relatively small number of large-scale
growers, thus cutting down on transactions costs. When,
however, supermarket buyers have no alternative to sourcing
supplies from smallholders because there are insucient large
farmers in a country, and importation is uneconomic or restricted by import regulations, they have sometimes proved
willing to invest in technical assistance and credit systems to
small farmers to improve the quantity, quality, and reliability
of supplies (Reardon, Timmer, & Berdegue, 2005). 11
The third consideration is how much credence attributes
matter for agricultural produce, which varies by market and
product. For many staple foods and traditional export crops,
credence attributes are not important while for higher-value
produce, such as horticulture and livestock, 12 they typically
are.
Based on these latter two considerations, a typology of situations can be constructed to assess the prospects for smallholders (Table 4). The two axes distinguish between goods
where credence matters or not and between situations where
buyers have to deal with smallholders since land distribution
is relatively equal, or where land is unequally held and buyers
can deal primarily with large farmers.
Staples and traditional cash crops tend to be in cells A and B
with opportunities for smallholders to compete, especially in
cell A. By contrast, commodity chains for higher-value produce are increasingly located in cells C and D. The well-documented cases of smallholder exclusion from evolving
marketing channels occur particularly in cell D (e.g., Carter
& Barham, 1996; Dolan, Humphrey, & Harris-Pascal, 1999)
and sometimes also in cell B (Latin American supermarket
systems summarized by Reardon & Berdegue, 2002).
The few successes for smallholders in zone D tend to involve
some form of donor or NGO support and subsidy. History,
however, oers some hope. When global buyers began to
source tropical produce from smallholders in the developing
world one hundred or more years ago, similar challenges of
quality and consistency were faced. In most cases, solutions
were found, and much of the tropical exports came from small
farms. Again, policy may be important here if it can encourage
(rather than deter) buyers or large-scale producers to search
for innovations that will draw in smallholder producers.
Will the supermarkets and other high volume buyers turn
the staple chains into those with high credence attributes?
Supermarkets are most likely to try to change supply chains
when there is a strong demand for foods from a thriving market, a situation associated with economic growth and rising incomes. Such circumstances are promising for small farms as

they are likely to oer expanded opportunities for selling farm

produce outside supermarket chains. They also oer increased
demand for labor in nonfarm activities: a boon for marginal
farms where households already depend heavily on o-farm
sources for their incomes.
It is not yet clear how current changes will aect small
farms; almost certainly, the impact will dier considerably
by context. The policy challenge, however, is clear: how to
make institutional innovations that will allow at least some
small farms to overcome increased transaction costs and take
advantage of the emerging supply chains.
(ii) Decline in commodity prices and more open domestic
markets
The prices of most agricultural commodities have, in real
terms, been falling in the long run, and even the increases seen
in 2007 and 2008 still leave most prices below levels seen in the
1990s. Increased openness of domestic markets also means
that producers are much more exposed to competition from
imports.
The consequences for all producers are clear. If they cannot
raise their productivity or otherwise reduce their unit costs of
production faster than prices fall, they will lose income. How
much smallholders are more vulnerable to falling prices than
larger farmers is a moot point, hinging largely on whether
small farmers produce at higher costs than larger operators.
Evidence on this is scant. Data on costs of production on
farms are not regularly collected in most developing countries.
When they are, estimates of land and labor costs are fraught
with diculties, and dierent conventions for treating these
as well as for presenting summary measures may be followed
in dierent countries. However, in the cotton sector, West
African producers, almost all smallholders, are believed to
have some of the lowest costs in the world. Arguments for
the inverse relationship might also suggest that smaller farms
are, on average, lower cost producers.
Notwithstanding smallholders present costs of production,
they may be less able to adapt to falling commodity prices
than large farms, for two reasons. One, since marginal costs
of capital are generally considered to be higher for small farms
than large, they may be more disadvantaged when raising productivity requires increased capital investment in purchased
input or equipment, as often applies.
Two, smaller farms may also be disadvantaged if prices are
more variable since they are less able both to insure themselves
against price risk and to access capital to take them through
low prices within or between seasons, compared to larger
farms. 13
The converse is the question of how much small farmers
could benet from higher commodity prices if these were to
be achieved through world trade reform. The diculties of
small farms in accessing services and credit means that they
are often constrained in their ability to take advantage of

Table 4. Commercial interest in sourcing supplies from small farmers. Source: Poulton et al., 2005.
Demand for output
from small farms
and inequality in farm structure

Comparative
advantage of
small farms

Unimodal land distributionhigh

demand for smallholder produce

Bimodal land
distributionlow demand
for smallholder produce

Highcredence attributes not important

Lowcredence attributes important

THE FUTURE OF SMALL FARMS: TRAJECTORIES AND POLICY PRIORITIES

higher prices by expanding production. Exceptions to this include the limited number of areas where (a) it is still possible to
expand the total land area planted, or (b) contract farming
systems provide smallholder growers with all the services that
they need. 14 Large farms, on the other hand, with better access to markets, information, and capital will often be better
placed to take advantage of any price gains.
(iii) Agricultural research
There are concerns that research systems in developing
countries are generating fewer innovations to raise yields than
a quarter century ago. Funding to the international agricultural research centers has fallen in real terms. Moreover, the
centers have devoted more of their resources to countering
changing pest, disease problems, and environmental problems
than to furthering yield increases.
Fewer innovations for yield increases aect all farmers, large
and small, but there is one way in which smallholders may be
harder hit. There has been a dramatic shift in the balance of
research funding from public to private sources, particularly
in biotechnology, where there is the greatest potential for more
dramatic advances. Smaller farmers may lose out as private research lacks incentives to address their concerns (Pingali &
Traxler, 2002) and focuses more on the concerns of larger
farmers.
(iv) Environmental degradation and climate change
Farming can degrade the local environment. However, evidence of what is happening is patchy. Historically, careful
studies of changes to soil and water quality have usually had
to be carried out on a small scale although the development
of near infrared spectroscopy (Shepherd, Palm, Gachengo, &
Vanlauwe, 2003) looks set to change this. Extrapolation from
such studies to make estimates for larger areas is fraught with
problems. Studies of environmental change, moreover, tend to
focus on damage and do not always take into account
improvements made by farmers, such as soil conservation
works and tree planting.
Climate change represents a global phenomenon to which
farming contributes to in part and to which it is especially vulnerable since most farming depends on the weather. While the
science of climate change may be reasonably well established
broadly, the precise impacts of processes that play out over
decades are as yet only vaguely discernable. For example, attempts to predict changes in rainfall for large regions have
substantial margins of error. At country scale and below,
much more work is needed to improve predictions of the impact of climate change, but there is increasing evidence that
crop production will be hardest hit in tropical areas, particularly in Africa (Hulme, Doherty, Ngara, New, & Lister,
2001; Royal Society, 2005).
The impact on small farmers of both environmental degradation and climate change are usually assumed to be more severe than for larger holdings 15 since small farmers lack access
to human, social, and nancial capital and information compared to larger farmers. This is plausible but not proven. It
might equally be argued that larger farmers with heavy investments in xed capital are also vulnerable to changes in the
environment. Smallholders whose major asset is their labor
power may more easily adapt their production patterns and
practices to new environmental conditions.
(v) The impact of HIV/AIDS
In Eastern and Southern Africa, where the prevalence of
HIV infection is highest, the direct eects seen on households
aected include loss of labor to sickness, death, and caring,

1355

and erosion of capital and assets to pay for drugs, treatment,

and transport to the hospital. For the households in question,
the consequences for their agriculture may be less land tilled,
less use of purchased input, and crop substitution: from cash
to food crops for subsistence (and survival) and from crops
with high peak demands for labor to those less demanding
of labor. Cash crops are particularly likely to be abandoned
when adult males fall sick since men typically control such
crops and have contacts to market the produce. Agricultural
skills and knowledge, including highly specic knowledge of
the local ecology and plants, may not be passed down to younger generations (Jayne, Villarreal, Pingali, & Hemrich, 2004;
Mutangadura, Mukurazita, & Jackson, 1999).
Is the impact of the epidemic on smallholders more severe
than on larger farmers? Some smallholders may be particularly aected: those that are poor and lack assets. Studies to
date (Mather et al., 2004) show that poor households lacking
assets, savings, and other means to cover the costs of the disease are more vulnerable to reduced output, loss of productive
assets, and eventual destitution. Perhaps the most salient point
about the impact of the epidemic is the sharp discrimination
that it exerts. Those households aected by the disease usually
incur heavy losses. Their more fortunate neighbors may be little aected, 16 and this distinction is largely epidemiological
rather than a matter of scale of farming.
(vi) Changes in the policy environment
Since the 1980s, the international community has moved
away from supporting government intervention in agricultural
development. 17 In Africa, early green revolution successes,
like hybrid maize, that depended on the state supplying seed,
fertilizers, and undertaking marketing proved unsustainable
because of their high scal costs, contributing to eventual debt
crises and stagnation in many of the countries where it spread
(Smale & Jayne, 2003).
The shift in preferred policy from extensive interventions to
a narrower state role, leaving private actors in the market to
provide input, services, credit, and marketing, has left many
smallholders at a disadvantage since they face higher transaction costs in the markets than larger operators.
In summary, not all of the changes that might be thought to
be particularly harmful to small farmers are necessarily any
worse for them than they are for large-scale farmers, but some
clear threats to small farms emerge. In large part, they arise
from market failures amplied by the policy retreat from intervention that has left the private sector responsible for input
supply, nancial services, marketing, and even technical advice
and innovations. If smallholders are to survive and prosper,
then they have to nd ways to meet new demands in supply
chains. They have to obtain input, credit, and technical knowledge from private agents at the same prices paid by large-scale
farmers.
How far this can be achieved and how much public policy
can make it happen are the main dierences between those
who believe that small farmers have a future and those who
do not.
4. STRATEGIES FOR SMALL FARMS
Options for smallholder development will vary by context.
Table 5 presents these, building on the typology presented in
Table 2.
Two major sets of options emerge, one concerning growth,
the other distribution. Where agricultural potential exists,
and land is already distributed equitably, commercially

1356

WORLD DEVELOPMENT
Table 5. Priorities for small farms by country context

Land
distribution

Agricultural
potential

Unimodal

High

Low

Bimodal

High and low

Early stages of development

Nonagricultural engines of growth

Later stages of
development

Minerals

Manufacturing

Limited

All

Commercial
opportunities for small
farms selling high value
products in domestic
market. Social value in
retaining small farms to
spread mineral wealth
and provide subsistence
for the rural poor
Social value in retaining
small farms to spread
mineral wealth and
provide subsistence for
the rural poor
Social value in retaining
workers in agriculture
and providing
subsistence for the rural
poor

Commercial
opportunities for small
farms to sell food staples
and high value products
in domestic market.
Social value in retaining
small farms for
employment

Commercial opportunities
for small farms in export
crops, food staples, and some
high value products

Social value in retaining

small farms for
employment and
providing subsistence for
the rural poor
Social value in retaining
small farms for
employment and
providing subsistence for
the rural poor

Opportunities for small

farms to exploit niche
agricultural opportunities.
Small farms provide
subsistence for the rural poor
Opportunities for small
farms if land redistributed.
Otherwise, small farms that
exist exploit niche
opportunities. Social value in
retaining workers in
agriculture and providing
subsistence for the rural poor

Remaining small farms

gradually get squeezed
out and those that
survive focus on high
value products and/or
part-time farming. Social
value in retaining small
farms for employment
until sucient exit
opportunities have been
created

oriented small farms may be ecient and competitive. Because

many small farmers are also poor, these can be winwin
opportunities for growth and poverty alleviation. In countries
where agriculture is the lead growth sector, small farm growth
opportunities will primarily arise in the domestic markets for
food staples and in high value export markets at least during
the early stages of development when the domestic market
for high value products is still small.
Countries with large mineral or manufacturing sectors, on
the other hand, will probably have high exchange rates and
ready access to low cost food imports, so small farm growth
opportunities are likely to be limited to high-value domestic
markets.
A second option for small farms arises from its potential social contributions. Small farms can provide a way for governments to spread the benets from mining or manufacturing
during the early stages of development when most people
are still engaged in agriculture. As economies grow, small
farms can also serve as a useful reserve employer until sucient exit opportunities exist, a role that can be especially
important in fast-growing countries regardless of their primary engine of growth.
Social options matter most in countries with poor agricultural productivity potential, a bimodal distribution of land,
or abundant minerals or manufacturing. Small farms need
not be commercially viable. In fact, subsistence-oriented small
farms may be the most appropriate to target.
As the economic transformation proceeds, small farms have
a shrinking role to play in all kinds of countries. Rising real
wages within the wider economy tend to drive farm consolidation (as has occurred in many OECD countries and is now
occurring in parts of East Asia), and the small farms that survive nd niches in high value markets or become part-time
farmers.
Should governments support small farms? There is less debate about this issue when considering social roles since even
the most ardent-free market advocates only expect market

solutions to provide ecient outcomes and not necessarily

equitable or poverty-reducing outcomes. Direct support to
subsistence-oriented small farms may be more cost eective
than other income transfers and social safety nets. For example, food aid, a common response to distress from donors, typically costs more than US $250 for each ton of cereal delivered
in rural areas, compared to typical smallholder production
costs of US $100 or less. 18 Such support for nonviable small
farms may, however, encourage too many workers and poor
people to stay in agriculture or for too long.
The need for governments to help commercial small farms
grow is less obvious. Policy might be restricted to building
an enabling environment through provision of essential public
goods and services and maintaining a stable and undistorted
macro-economy. There are, however, many institutional and
market failures in poor countries that can lead to discriminatory and inecient outcomes. For example, if failures in input
and product markets aect small farms more than large farms,
as is likely, then large farms may be the only ones to take
advantage of market opportunities, leading to an outcome
that is less ecient and less equitable than smallholder development. In this case, targeted policy interventions to correct
underlying market failures might improve both eciency
and equity.
Failures in input and output markets are linked and spill
over from one market to another, as seen earlier in the
inverse ratio that can be explained by failures in both land
and labor markets. If farmers are to invest, they need reasonably reliable access to a range of factors, such as land, labor,
purchased input, technical skills, and information, on reasonable terms. If one factor is missing, then investments in others
will be lost or signicantly reduced.
Conventional liberalization policy tends to underplay both
the market failures and the need for complementarity in the
provision of input and services. In practice, potential suppliers
face highly uncertain demand for their services unless farmers
are assured of access to other complementary services

THE FUTURE OF SMALL FARMS: TRAJECTORIES AND POLICY PRIORITIES

(Poulton, Dorward, & Kydd, 2005). Such assurance is lacking

in many developing rural areas unless some external agent
either provides the full range of services directly or establishes
a way to coordinate other suppliers that is credible to both
farmers and suppliers.
These arguments are supported by observation that successful green revolutions for staple crops have generally been supported by considerable state action to ensure the supply of
credit, input, and technical assistance (Dorward, Kydd,
Morrison, & Urey, 2004). Similarly, intensive cash crop production by small farmers either has often been developed by
the state acting through marketing boards or has been promoted by the private sector when it has been possible to interlock credit, input supply, and marketing in schemes such as
contract farming (Best, Ferris, & Schiavone, 2005).
The potential for private coordination of complementary input and services for small farms usually depends on the market
characteristics for products. Private companies may take on
the costs and risks if (1) high xed costs in processing or other
downstream activities mean that processors need assurance on
the supply of high volumes of farm output; (2) small farms are
important suppliers either because they are the only farms,
have lower costs than large farms, or there are political benets in dealing with small farms; and (3) the company has some
degree of monopsony in buying farmers produce, so crop purchases provide collateral for seasonal loans and thus some
protection against strategic default by farmers. 19 Conditions
(1) and (3) are often related in that high xed costs lead to
economies of scale and represent an entry barrier to smallscale buyers.
The markets for many staple crops, and indeed for some
cash crops, do not have these characteristics. Private actors
are thus unlikely to solve coordination problems, but the state,
especially in Africa, has only been able to undertake this role
at high and usually unsustainable cost. A key challenge to
small farm development is therefore to develop new coordination systems involving combinations of government agencies,
civil society, farmer organizations, and agri-business rms.
Such mechanisms are being developed and tested on a small
scale with mixed success (see, e.g., Poulton, Kydd, Wiggins,
& Dorward, 2006), but greater eorts are needed.
The policy for smallholder development will change through
time, as implied in Table 5. In the poorest countries and rural
areas, deciencies, such as low road density, poor roads, poor
telecommunications, poor human health, lack of irrigation
infrastructure, and lack of productive agricultural technologies, combine with weakly developed markets. In such conditions, there are strong initial needs for investment in public
goods of infrastructure, agricultural research, and extension,
which are key elements in green revolutions (Dorward et al.,
2004). Once in place, attention moves to developing institutions to support business and market activity, including ways
to solve the coordination problems described, and kick starting markets (Dorward et al., 2004). Finally, once the markets
have been developed with supporting institutions, government
can withdraw, restricting its role to the supply of purely public
goods.
Policy thus needs to match circumstances and change
through time. Do governments have the capacity to make such
adjustments? Weaknesses in administrative and technical
capacity are nothing new but have been exacerbated by structural adjustments that neglected, rather than reformed, many
public institutions serving rural areas. Reform and strengthening of public agencies will often require overcoming vested
interests; otherwise, new forms of ineciencies and rent-seeking simply replace old. Innovations may be needed, for exam-

1357

ple, nancing arrangements that empower the users of public

services (e.g., vouchers, user fees, and other co-nancing
mechanisms) including appropriate institutional reforms to
improve mandates and performance. Even if the government
nances services, contracting out implementation to NGOs,
community bodies, and private rms can be more cost eective
and may oer better possibilities for involving local people
and communities.
Successful public policy is more than technical capacity.
Political will is also needed. The rst round of green revolutions in Asian countries began under strong national imperatives, such as avoiding the humiliating dependence on food
aid (India), establishing the basis of economic growth (China),
and building support for the government in the countryside
(Indonesia).
In Africa, agriculture is often seen as the backbone of the
economy, yet the share of the national budgets devoted to
agriculture remains consistently well below that in Asia (Fan
& Rao, 2003). 20 When signicant sums are spent, they tend
to fund subsidies rather than long-term investments in productive capacity. Despite structural adjustments, many African
countries have yet to reduce public intervention in markets,
owing to resistance from entrenched political and bureaucratic
interests that retain the control of policy levers that are useful
for patronage or rent-seeking purposes. Meanwhile, as budgets have contracted, long-term investment has been increasingly left to donors whose own funding for the agricultural
sector has been in decline.
Some changes, however, are underway that may improve
the prospects for reforms, including democratization, decentralization, and participatory processes (Birner & Resnick,
2005). Democratization may squeeze opportunities for private
rent-seeking in the long-term 21 and ultimately also strengthen
the voice of small farm households simply by virtue of their
numbers. The long-term, however, could be long indeed. In
the meantime, the need for presidents or ruling groups to
win elections may actually strengthen the incentives for the
exercise of patronage.
Decentralization oers the particular promise for more
eective local support to small farms given that agro-ecological heterogeneity means that general solutions need adaptation for local application (Foster, Brown, & Naschold, 2001).
The local level is also where much of the relevant information is available for holding front-line service providers to account for their performance. Decentralization is not without
risks; local administrations may be under-resourced, and
decentralized planning may be captured by local elites (Bardhan, 1996). Furthermore, it does not remove the need for
central funding of national public goods, such as agricultural
research.
Increased participation, as seen, for example, in the preparation of poverty reduction strategies, may also be a boon for
small-scale agriculture given the large number of stakeholders
involved, both within the government (where relevant ministries might include Livestock, Forestry, Water Resources,
Roads, and Finance, as well as Agriculture) and outside (Foster et al., 2001). Policy making can benet from more information and expertise, provided that the forums used encourage
participants to seek a consensus on ways forward, as opposed
to using them to propagate and entrench conicting viewpoints. Regular deliberative fora (Hall & Soskice, 2001)
may help to forge consensus even where participants begin
with polarized views. However, the challenge may be greater
in situations of major inequality (e.g., pitching unions of peasants and the landless against large landholders or corporate
interests in Latin America) or when the discussion concerns

1358

WORLD DEVELOPMENT

staple foods where local producer and consumer interests diverge, rather than export crops. Success with such forums
has been seen in southern and eastern African cotton sectors
(Tschirley, Poulton, & Boughton, 2006) due to the relatively
small numbers of key stakeholders involved and the reasonable coincidence of interests. The potential of forums to
debate issues and to hold public agencies to account for their
performance is particularly important in Africa, where political scientists are pessimistic about the ability of other measures
to push neo-patrimonial political systems in a more developmental direction.
Donors are major players in policy making in Africa even if
there are limitations to their inuence when strong domestic
interests are threatened (de Renzio, 2006). After almost two
decades of declining allocations of donor funds for agriculture
and rural development, there are signs of renewed commitment. There are dangers in increased funding, of course. Additional funds to public agencies could undermine incentives to
reform themselves or to adopt more eective policies. Correcting this by setting prior conditions has not proved eective in
the past. It may be better if donors make credible (and coordinated) commitments to reward better governance within such
agencies with additional resources. Conditionality could be focused on process, such as the encouragement of the forums discussed, rather than on the adoption of particular policy
reforms. 22 Donors may also consider supporting the emergence of farmer organizations for both service delivery and
advocacy. At the local level, they may form a counter-weight
to the power of local elites in decentralized planning processes.
Currently favored ways of providing international aid
through budget support linked to poverty reduction strategy
papers or sector-wide programs do not necessarily do agriculture any favors (Cromwell, Luttrell, Shepherd, Wiggins, &
Cabral, 2005). The kind of blueprint planning envisaged in
such processes is dicult for a sector when there are diverse
stakeholders, in which so much depends on private actors,
rather than the government, and general solutions need local

adaptation (Foster et al., 2001). The search has thus begun

for ways to adjust agricultural planning to these realities as
well as to modify the way new instruments are used to allow
them to take account of agriculture and indeed other productive sectors.
5. CONCLUSION
The case for smallholder development as one of the main
ways to reduce poverty in low-income countries remains
compelling. The policy agenda, however, has changed from
the early days of the green revolution, as the liberalization
of international trade and domestic markets has allowed private enterprise to become increasingly important with a consequent restructuring of supply chains. For small farmers, the
diculties and costs of accessing inputs and markets are
mounting.
Hence, while public goods need to be provided as they always have, the growing challenge is to improve the workings
of markets for output, input, and nancial services to overcome market failures. This calls for innovations in institutions,
for joint work among farmers, private companies, and NGOs,
and for ministries of agriculture and other public agencies to
take on new, more facilitating roles.
New thinking on the role of the state in agricultural development, wider changes in democratization, decentralization, the
introduction of participatory policy processes, and a renewed
interest in agriculture among major international donors presents opportunities and gives ground for hope that greater and
more eective support to small farm development can be delivered. However, unless key policy makers adopt a more assertive agenda toward small farm agriculture, there is the growing
risk that there will soon be a dramatic increase in rural poverty
and waves of migrants to urban areas that could overwhelm
available job opportunities, urban infrastructure, and support
services.

NOTES
1. There is a large body of econometric literature that uses cross-country
or time series data to estimate growth-poverty elasticities by sector. These
studies generally nd high-poverty reduction elasticities for agricultural
productivity growth, especially in the early stages of development and
relative to other sectors. For example, in a cross-country study, Thirtle,
Lin, and Piesse (2002) estimated that a 1% increase in crop productivity
reduces the number of poor people by 0.72% in Africa and by 0.48% in
Asia. In India, Ravallion and Datt (1996) have estimated the elasticity of
poverty reduction with respect to agricultural value added per hectare at
0.4% in the short run and 1.9% in the long run, the latter through the
indirect eects of lower food prices and higher wages.
2. The World Bank (2002) estimated their total value at about US $330
billion per year.
3. Even if the 2007/08 food price spike raised food prices, in real terms,
they are still well below those seen in the 1960s and 1970s when the Green
Revolution began in Asia.
4. The evidence for the inverse relationship (IR) is not undisputed. There
are particular diculties with denitions of farm size and measures of
productivity. However, when studies tried to rene denitions of size and
productivity (for example, looking at size in terms of land area per worker
and at dierences in productivity per ha with an adjustment for land
quality), the IR has often been strengthened (Lipton, 1993).

5. Family-operated farms predominate in most OECD countries,

albeit that mechanization allows much larger areas to be cropped or
bigger herds to be kept. The exceptions arise in horticulture, pigs,
and poultry, where more industrial forms of agriculture are technically possible.
6. The more ecient use of labor by small farms arises as a result
of lower transaction costs, and some of these, relating to greater
self-motivation and lower supervision costs, arise as a result of the
low opportunity costs of labor for poor farmers and hence their
self-exploitation (Dyer, 1991, 1996). Under such circumstances,
marginal returns to labor may be lower than those on large farms
though total unskilled employment and labor earnings should be
higher.
7. Examples include the generous protection of sugar production in
Nicaragua, an estate crop; the high price paid for maize by the Cereal
Board in Kenya that takes deliveries mainly from medium- to large-scale
farmers; and the subsidized credit schemes oered in Brazil in the 1970s
and 1980s that were taken up mainly by large farmers in the more
prosperous regions of the country.
8. It could be argued that some of these challenges were present in the
later agricultural transformations in China and Vietnam or were specifically addressed by policies and public investments.

THE FUTURE OF SMALL FARMS: TRAJECTORIES AND POLICY PRIORITIES

9. As noted earlier, small farmers tend to be disadvantaged relative to
larger farmers by increased market transactions for input, nance, and
output (but not for labor). Green Revolution technologies increased
input use, nancial demands, risk, and output per ha (favoring the
eciency of large farms) but also labor demands (favoring the eciency
of small farms). These mixed benets to small and large farms are
consistent with the observation that the inverse ratio appeared to
weaken in the early stages of the Green Revolution as large farmers
adopted new technologies rst, but the inverse ratio was then often reestablished in Green Revolution areas as new technologies were
adopted on small farms.
10. Traill (2006) projected supermarket shares of retail sales to 2015.
While in many Latin American countries it is expected that supermarkets
will have between half and three-quarters of the market, and in China the
share will be more than one quarter; the corresponding share for Kenya is
forecast to be 16% and less than 10% for South Asia.
11. In Africa, it is intriguing to see that supermarkets have penetrated
furthest where there is access to large farms (South Africa, Zimbabwe,
Kenya, and Zambia).
12. Note that, when credence attributes are not insisted upon, small
farms can thrive as suppliers of horticultural produce because of their
advantages over large farms in terms of labor quality and motivation
(Reardon et al., 2005).
13. There is a more general point here regarding large farm advantages in coping with variability. They are also likely to do better from
years of good weather and yields if access to seasonal capital and
storage facilities allows them to store produce from good harvests until
prices improve. Small farms without storage facilities and capital are
more likely to be forced to sell soon after harvest when markets are
glutted and prices low.
14. Thus, Gillson, Poulton, Balcombe, and Page (2004) found that
African cotton production, especially that in Tanzania and Zimbabwe,
was highly responsive to the world cotton lint price. In fact, it was
more responsive to prices than US production. In both cases, however,
production responded with a one-year lag by which time world prices
had often changed again. This highlights the disadvantage that
smallholders face relative to large farms in terms of market
intelligence.
15. The impact of climate change extends beyond agriculture to include
other sectors, such as health, where poor and vulnerable communities are
likely to be hit hardest (IPCC, 2001).

1359

16. Indeed, they may conceivably benet from those aected selling
livestock, land, or other valuable assets at distress prices.
17. Although a ubiquitous feature of successful green revolutions
(Dorward et al., 2004), the high scal costs associated with many
marketing and input subsidies became an escalating burden as governments proved unable to phase them out once they had achieved their
initial purposes. India, for example, currently spends about US $10 billion
per year on subsidies that are largely unproductive (Dorward et al., 2004).
Similar problems persist in many other Asian countries.
18. For example, for smallholder farms in Zimbabwe, the estimated cost
of producing one ton of maize was under US $80 in 1995/96 (Sukume,
Makudze, Mabeza-Chimedza, & Zitsanza, 2000). Imports of maize from
the world market usually cost at least US $220 CIF Harare. If these
imports are then delivered to rural areas as food aid, additional transport,
and handling costs have to be added. In Malawi, in 200304, an informed
advisor on food security claimed that while the import parity price of
maize was around US $250 a ton, it cost a leading food aid agency as
much as US $450 a ton to deliver food aid to rural clients, a cost that
includes those of targeting.
19. This can also be achieved by horizontal coordination between buyers
who agree to share information about farmers who default on loans
(Stockbridge, Smith, & Lohano, 1998).
20. Agriculture has recently received a higher political prole in Africa.
In 2003, the Heads of African States of the African Union declared (the
Maputo Declaration) that they would allocate up to 10% of their scal
budgets to agriculture by 2008, and African governments are also working
together on the Comprehensive Africa Agriculture Development Programme (CAADP) through the New Partnership for Africas Development (NEPAD). Subsequently, there has been some, if not rapid, progress
toward this target.
21. This will depend, inter alia, on the rules of the governing party
funding under the evolving political dispensations.
22. Some (e.g., Lockwood, 2005) argue against conditionality of any
kind, suggesting instead that the majority of aid funding should be used to
reward key development outcomes ex-post. In his discussants comments
at the 2005 Small Farms workshop in Wye, Rob Paarlberg argued that
funding should reward measurable outcomes in terms of delivery of
specic rural public goods, such as increases in road density or agricultural
research output. This would leave the search for appropriate institutional
arrangements for the delivery of such public goods entirely up to local
stakeholders.

REFERENCES
Bardhan, P. (1996). Decentralised development. Indian Economic Review,
31(2), 139156.
Best, R., Ferris, S., & Schiavone, A. (2005). Building linkages and enhancing
trust between small-scale rural producers, buyers in growing markets
and suppliers of critical inputs. In Paper presented at crop post-harvest
workshop beyond agriculture: Making markets work for the poor (28
February & 1 March, 2005). London: Crop Post Harvest Programme.
Binswanger, H. P., & Rosenzweig, M. R. (1986). Behavioural and material
determinants of production relations in agriculture. Journal of Development Studies, 22(3), 503539.
Birner, R., & Resnick, D. (2005). Policy and politics for smallholder
agriculture. In Paper presented to The Future of Small Farms
research workshop. Wye.
Boselie, D., Henson, S., & Weatherspoon, D. (2003). Supermarket
procurement practices in developing countries: Redening the roles

of the public and private sectors. American Journal of Agricultural

Economics, 85(5), 11551161.
Carter, M. R., & Barham, B. L. (1996). Level playing elds and laissez
faire: Postliberal development strategy in inegalitarian agrarian economies. World Development, 24(7), 11331149.
Cornia, G. A. (1985). Farm size, land yields and the agricultural
production function: An analysis for 15 developing countries. World
Development, 13(4), 513534.
Cromwell, E., Luttrell, C., Shepherd, A., Wiggins, S., & Cabral, L. (2005).
Poverty reduction strategies and the rural productive sectors: Insights
from Malawi, Nicaragua and Vietnam. London: Overseas Development
Institute (Working Paper 258).
de Renzio, P. (2006). The primacy of domestic politics: Ethiopia and
Uganda, what next with aid? Opinions 65. London: Overseas Development Institute.

1360

WORLD DEVELOPMENT

Deininger, K., & Squire, L. (1998). New ways of looking at old issues:
Inequality and growth. Journal of Development Economics, 57,
59287.
Dolan, C., Humphrey, J., & Harris-Pascal, C. (1999). Horticulture
commodity chains: The impact of the UK market on the African fresh
vegetable industry. Working Paper 96. Brighton: Institute of Development Studies, University of Sussex.
Dorward, A. R., Kydd, J. G., Morrison, J. A., & Urey, I. (2004). A policy
agenda for pro-poor agricultural growth. World Development, 32(1),
7389.
Dorward, A., Fan, S., Kydd, J., Lofgren, H., Morrison, J., Poulton, C.,
et al. (2004). Institutions and economic policies for pro-poor agricultural
growth. Washington DC: Development Strategy and Governance
Division, International Food Policy Research Institute (Discussion
Paper No. 15).
Dyer, G. (1991). Farm size farm productivity re-examined: Evidence
from rural Egypt. Journal of Peasant Studies, 19, 5992.
Dyer, G. (1996). Output per acre and size of holding: The logic of peasant
agriculture under semi-feudalism. Journal of Peasant Studies, 25,
103131.
Ellis, F. (2005). Small-farms, livelihood diversication and ruralurban
transitions: Strategic issues in sub-Saharan Africa. In Paper, research
workshop on the future of small farms, organised by IFPRI. Wye:
Imperial College and ODI.
Ellis, F., & Harris, N. (2004). New thinking about urban and rural
development. UK: Department for International Development Sustainable Development Retreat.
Fafchamps, M., Teal, F., & Toye, J. (2001). Towards a growth strategy for
Africa. Centre for the Study of African Economies (Draft 2 October
2001).
Fan, S., & Rao, N. (2003). Public spending in developing countries: Trends,
determination and impact, EPTD Discussion Paper No. 99. Washington, DC: International Food Policy Research Institute.
Foster, M., Brown, A., & Naschold, F. (2001). Sector programme
approaches: Will they work in agriculture?. Development Policy
Review, 19(3), 321338.
Gillson, I., Poulton, C., Balcombe, K., & Page, S. (2004). Understanding
the impact of cotton subsidies on developing countries. London:
Overseas Development Institute (IEDG Working Paper; <http://
www.odi.org.uk/iedg/publications/online_papers.htm>).
Hall, P., & Soskice, D. (2001). Varieties of capitalism: The institutional
foundations of comparative advantage. Oxford: Oxford University
Press.
Hazell, P., & Roell, A. (1983). Rural growth linkages: Household
expenditure patterns in Malaysia and Nigeria (Research Report
No. 41). Washington, DC: International Food Policy Research
Institute.
Heltberg, R. (1998). Rural market imperfections and the farm sizeproductivity relationship: Evidence from Pakistan. World Development, 26(10), 18071826.
Hulme, M., Doherty, R., Ngara, T., New, M., & Lister, D. (2001). African
climate change: 19002100. Climate Research, 17, 145168.
IFPRI (2005). The future of small farms. Proceedings of a research
workshop. Washington, DC: International Food Policy Research
Institute.
IPCC (2001). Climate change 2001: Impacts, adaptation, and vulnerability.
Contribution of Working Group II to the third assessment report of the
IPCC. Cambridge: Cambridge University Press.
Jayne, T. S., Villarreal, M., Pingali, P., & Hemrich, G. (2004). Interactions
between the agricultural sector and the HIV/AIDS pandemic: Implications for agricultural policy, ESA Working Paper No. 04-06, March
2004. Agricultural and Development Economics Division. Rome: Food
and Agriculture Organization of the United Nations.
Lipton, M. (1993). Land reform as commenced business: The evidence
against stopping. World Development, 21(4), 641657.
Lipton, M. (2004). Crop science, poverty, and the family in a globalising
world. Plenary Session, Brisbane International Crop Science Conference.
Lipton, M. (2005). Crop science, poverty and the family farm in a
globalising world (Washington, IFPRI 2020 Discussion Paper). Washington, DC: International Food Policy Research Institute.
Lockwood, M. (2005). The state theyre in: An agenda for international
action on poverty in Africa. Warwick: ITDG Publishing.
Mather, D., Donovan, C., Jayne, T. S., Weber, M., Mazhangara, E.,
Bailey, L., et al. (2004). A cross-country analysis of household responses

to adult mortality in rural sub-Saharan Africa: Implications for HIV/

AIDS mitigation and rural development policies MSU International
Development Working Paper No. 82. East Lansing, MI: Department
of Agricultural Economics, Michigan State University.
Maxwell, S. (2003). Six characters (and a few more) in search of an author.
How to rescue rural development before its too late. In Paper
presented at the 25th international conference of agricultural economists.
Durban, South Africa.
Maxwell, S., Urey, I., & Ashley, C. (2001). Emerging issues in rural
development: An issues paper. London: Overseas Development Institute.
Mellor, J. W. (1976). The new economics of growth: A strategy for
India and the developing world. Ithaca, NY: Cornell University
Press.
Mutangadura, G., Mukurazita, D., & Jackson, H. (1999). A review of
household and community responses to the HIV/AIDS epidemic in the
rural areas of sub-Saharan Africa, best practice collection. Geneva:
UNAIDS.
Pingali, P., & Traxler, G. (2002). Changing locus of agricultural research:
Will the poor benet from biotechnology and privatization trends?.
Food Policy, 27(3), 223238.
Poulton, C., Kydd, J., Wiggins, S., & Dorward, A. (2006). State
intervention for food price stabilisation in Africa: Can it work?. Food
Policy, 31(4), 342356.
Poulton, C., Dorward, A. R., & Kydd, J. G. (2005). The future of small
farms: New directions for services, institutions and intermediation. In:
Conference on the future of small farms. Wye.
Ravallion, M., & Datt, G. (1996). How important to Indias poor is the
sectoral composition of economic growth?. World Bank Economic
Review, 10(1), 126.
Ravallion, M., & Datt, G. (2002). Why has economic growth been more
pro-poor in some states of India than others?. Journal of Development
Economics, 68, 381400.
Raynolds, L. (2004). The globalization of organic agro-food networks.
World Development, 32(5), 725743.
Reardon, T., Timmer, C. P., & Berdegue, J. A. (2005). Supermarket
expansion in Latin America and Asia. In A. Regmi, & M. Gehlhar
(Eds.), New directions in global food markets, agriculture information
bulletin, No. 794, economic research service. Washington, DC:
USDA.
Reardon, T., & Timmer, C. P. (2007). Transformation of markets for
agricultural output in developing countries since 1950: How has
thinking changed? In R. E. Evenson, P. Pingali, & T. P. Schultz (Eds.),
Handbook of agricultural economics, volume 3: Agricultural development: Farmers, farm production and farm markets. Amsterdam:
Elsevier Press.
Reardon, T., & Berdegue, J. A. (2002). The rapid rise of supermarkets in
Latin America: Challenges and opportunities for development. Development Policy Review, 20, 371388.
Rosegrant, M., & Hazell, P. (2000). Transforming the rural Asian economy:
The unnished revolution. Hong Kong: Oxford University Press.
Royal Society (2005). Food crops in a changing climate: Report of a royal
society discussion meeting held in April 2005. Policy Document 10/05.
London: Royal Society.
Shepherd, K. D., Palm, C. A., Gachengo, C. N., & Vanlauwe, B. (2003).
Rapid characterization of organic resource quality for soil and
livestock management in tropical agroecosystems using near infrared
spectroscopy. Agronomy Journal, 95, 13141322.
Singh, H. (2005). The future of small farms: New directions for services,
institutions and intermediation. In Conference on the future of small
farms. Wye.
Smale, M., & Jayne, T. S. (2003). Maize in Eastern and Southern
Africa: Seeds of success in retrospect. EPTD Discussion Paper
No. 97. Washington, DC: International Food Policy Research
Institute.
Stockbridge, M., Smith, L., & Lohano, H. R. (1998). Cotton and wheat
marketing and the provision of pre-harvest services in Sindh Province,
Pakistan. In A. R. Dorward, J. Kydd, & C. Poulton (Eds.),
Smallholder cash crop production under market liberalisation: A new
institutional economics perspective. Wallingford, UK: CAB International.
Sukume, C., Makudze, E., Mabeza-Chimedza, R., & Zitsanza, N. (2000).
Comparative economic advantage of crop production in Zimbabwe.
Technical Paper No. 99, November 2000, SD Publication Series, Oce
of Sustainable Development, Bureau for Africa, USAID.

THE FUTURE OF SMALL FARMS: TRAJECTORIES AND POLICY PRIORITIES

Thirtle, C., Lin, L., & Piesse, J. (2002). The impact of research-led
agricultural productivity growth on poverty reduction in Africa, Asia, and
Latin America. Research Paper No. 016, Management Centre Research
Papers. London: Kings College.
Traill, W. B. (2006). The rapid rise of supermarkets?. Development Policy
Review, 24(2), 163174.

1361

Tschirley, D., Poulton, C., & Boughton, D. (2006). The many paths of cotton
sector reform in Eastern and Southern Africa: Lessons from a decade of
experience, international development working papers 88, department of
agricultural economics. East Lansing, MI: Michigan State University.
World Bank. (2002). Institutions for markets: World development report
2001/2002. Washington DC: World Bank.

Available online at www.sciencedirect.com