Forty years of farming systems classification for enhanced food security and poverty reduction

Abstract

Further progress towards improved food security and rural incomes and realization of the Millennium Development Goals (MDGs) requires a substantial increase in agricultural production, particularly of cereal crops such as maize, rice and wheat. During the four decades of experience with farming systems analyses, there has been a progressive broadening of research and development agendas and enrichment of classification methods. The Food and Agriculture Organization - World Bank farming systems framework represents a recent synthesis that blends systems and\ livelihood perspectives in order to analyse, classify and map farming systems in developing regions. The framework characterizes farming systems in terms of their potential to improve food security and reduce poverty, based on the two dimensions of: (i) agricultural resource endowment, e.g. land quality, and (ii) access to institutions, e.g. market access. These characteristics shape farming system development pathways and household livelihood strategies that intensify production and augment farm incomes largely from cereal-based systems (the bedrock of national and global food security) and thereby accelerate progress towards the MDGs. Among the five principal livelihood strategies for reducing poverty, farm intensification and diversification are promising avenues for substantial poverty reduction in many cereal farming systems. The relative importance of povertyreducing intensification and diversification strategies in different farming systems provides a basis for priority setting and targeting of agricultural research.

Full text

Review
Forty years of farming systems classification for enhanced food security
and poverty reduction
John Dixon
1
*, Jon Hellin
2
, Xiaoyun Li
2
and Glenn Hyman
3
Address:
1
Australian Centre for International Agricultural Research (ACIAR), Canberra, Australia.
2
International Maize and Wheat
Improvement Center (CIMMYT), El Bata
´n, Mexico.
3
International Center for Tropical Agriculture (CIAT), Cali, Columbia.
*Correspondence: John Dixon. Email: dixon@aciar.gov.au
Received: 22 August 2009
Accepted: 22 September 2009
doi: 10.1079/PAVSNNR20094060
The electronic version of this article is the definitive one. It is located here: http://www.cabi.org/cabreviews
gCAB International 2009 (Online ISSN 1749-8848)
Abstract
Further progress towards improved food security and rural incomes and realization of the
Millennium Development Goals (MDGs) requires a substantial increase in agricultural production,
particularly of cereal crops such as maize, rice and wheat. During the four decades of experience
with farming systems analyses, there has been a progressive broadening of research and devel-
opment agendas and enrichment of classification methods. The Food and Agriculture Organization –
World Bank farming systems framework represents a recent synthesis that blends systems and
livelihood perspectives in order to analyse, classify and map farming systems in developing regions.
The framework characterizes farming systems in terms of their potential to improve food security
and reduce poverty, based on the two dimensions of: (i) agricultural resource endowment, e.g.
land quality, and (ii) access to institutions, e.g. market access. These characteristics shape farming
system development pathways and household livelihood strategies that intensify production and
augment farm incomes largely from cereal-based systems (the bedrock of national and global food
security) and thereby accelerate progress towards the MDGs. Among the five principal livelihood
strategies for reducing poverty, farm intensification and diversification are promising avenues for
substantial poverty reduction in many cereal farming systems. The relative importance of poverty-
reducing intensification and diversification strategies in different farming systems provides a basis
for priority setting and targeting of agricultural research.
Keywords: Farming systems, Livelihoods, Poverty, Classification, Intensification, Diversification, Markets
Review Methodology: Keyword search of CAB Abstracts and other bibliographic collections including ISI Web of Knowledge Service
for UK Education, the University of Florida’s farming systems literature collection of P. Hildebrand and the web using Google Scholar,
including high-citation articles, using the following keywords: farming systems, agricultural systems, systems. Review of key journals
and texts, including the Proceedings of International and Regional Farming Systems Association Symposia and Conferences. Authors’
wide-ranging professional experience that spans several disciplines, different research and development organizations and all
developing regions.
Introduction
Progress towards the Millennium Development Goal
(MDG) of halving hunger and poverty has been dis-
appointing [1]. Agricultural development is generally
recognized as a key driver of poverty reduction and
economic growth [2, 3], not least because the reduction
of hunger among smallholder farmers depends on in-
creasing their entitlements to food through expanded
production and/or increased purchasing power [4, 5].
Taking into account increased human consumption of
cereals arising from population expansion, income growth
and expanded non-food uses of cereals, it is expected
that the demand for cereals will double in the coming
decades [6].
Agriculture is evolving and differentiating at the local
level [7, 8] and faces new opportunities and risks associ-
ated with threatened natural resources [9, 10], changing
http://www.cabi.org/cabreviews
CAB Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources 2009 4, No. 060

agricultural markets and policies [11], volatile food prices,
biofuel production and climate change [1, 12]. Future
increases in agricultural productivity and adjustments
to farming systems depend ultimately on management
decisions of hundreds of millions of farmers and other
entrepreneurs along the ‘U-impact pathways’ or input and
output value chains that connect consumers to farmers
(through agricultural product markets and price signals)
and agricultural researchers (through improved seed,
crop management technologies, information and other
agricultural services). Within the constraints of accessible
resources, institutions and knowledge, farm household
systems co-evolve with the input and output chains in
response to changes in demand (for example, from
population growth, urbanization, shifting consumer pre-
ferences, new processing technologies and improved
value chain efficiency) and changes in factors influencing
supply (for example, access to improved seed, crop
management technologies, inputs and knowledge, as well
as the prevailing institutions and policies) [1, 13]. Fur-
thermore, the decisions of farm women and men to adopt
or adapt improved varieties, breeds and management
practices are conditioned by the characteristics of their
farming systems.
Three cereals – maize, rice and wheat – underpin global
food security and, in most developing countries, national
food security [14]. Table 1 shows the relative contri-
bution of these cereals to global food consumption
(approximately 950 million tonnes) and animal feed
(approximately 515 million tonnes) [15]. Maize, rice and
wheat contribute 19, 46 and 40% to aggregate cereal food
consumption in the developing world. Even though maize
is the most common food staple of Africa, it underpins the
animal feed industry in Asia. Cereals remain the backbone
of many farming systems, albeit complemented by a wide
range of other crops, livestock and off-farm activities. The
productivity of these cereal-based farming systems is a
critical determinant of food security at global, national and
household levels, as well as the household income and
other livelihood improvements for small farm families
[14, 16].
As Table 2 shows, maize, wheat and rice together
accounted for 80% of aggregate cereal area in developing
countries, which was approximately 455 M ha (million
hectares) in 2004–2006, having expanded slowly by 22 M
ha over the period 1990–1992 to 2004–2006. Maize, rice
and wheat account for 22, 33 and 25%, respectively,
of cereal area. The areas of minor cereals are declining,
for example, barley and millet areas contracted over this
period, in contrast to the expansion of rice, wheat and
especially maize. Six major non-cereal food crops account
for a further 335 Mha of cultivated area, of which oil seeds
and pulses account for 43 and 18% of area, respectively.
The areas of oil crops, fruit and vegetables have increased
substantially, albeit from relatively small bases [15].
The central challenge for agricultural research is the
doubling of cereal production to meet expanding food,
fuel and feed needs [17, 18]. Although bioenergy crops
account for less than 1% of global cultivated area, biofuel
already absorbs a major portion of the US maize harvest
(although little cereal grain elsewhere in the world),
thereby reducing grain available for export and animal
feed. The livestock revolution, driven by strong consumer
demand for animal products, is strengthening demand for
feed across most regions. The livestock revolution itself
has transformed farming systems in distinct ways, e.g.,
the intensification of meat and milk production in mixed
crop–livestock farming systems, cf. the emergence of
specialized meat and egg production value chains [19].
In addition, the demand for maize feed grain is inducing
both cereal intensification and diversification, e.g., in rice
farming systems in South and East Asia.
Given the slow expansion of cereal area shown in
Table 2, more than two-thirds of the total increase of
cereal production in the developing regions since 1960
derived from yield growth [20]: from 1961 to 2007
average maize yield in the developing regions increased
from 1.13 to 3.2 t/ha, rice from 1.74 to 4.0 t/ha and wheat
from 0.77 to 2.9 t/ha [15]. Such a notable growth in
productivity stems in substantial measure from invest-
ments in agricultural research, for which high rates of
returns are frequently estimated [21, 22]. Since yields
under favourable commercial production situations in
Table 1 Average annual maize, rice and wheat supply
and consumption, 2001–2003
Developing Developed World
Aggregate
supply (Mt)
Maize 312.4 321.2 633.6
Rice (milled) 369.7 17.8 387.5
Wheat 331.6 250.2 581.8
Aggregate
consumption (Mt)
Maize
Food 100.7 72.4 173.1
Feed 177.4 233.6 411
Other 34.3 15.2 49.5
Rice (milled)
Food 333.2 15.9 349.2
Feed 6.3 0.4 6.6
Other 30.2 1.5 31.7
Wheat
Food 288.7 134.9 423.6
Feed 10.9 86.5 97.4
Other 32 29.8 61.8
Maize, rice and
wheat food
consumption
per capita
Population (b) 5.1 1.3 6.4
Consumption
(kg/cap)
142 172 148
Source: [15].
Note: Following the conventions in FAOSTAT commodity balance
databases, rice is expressed in milled grain equivalents, food
includes the category food manufactures, and other includes seed,
waste, industrial and biofuels uses.
http://www.cabi.org/cabreviews
2 Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources

developed countries are often double or treble the
average yields in developing countries (for example maize
yields in developed countries average 8.3 t/ha cf. 3.2t/ha in
developing regions), agricultural research has an impor-
tant role to play in boosting cereal production in devel-
oping regions. It is therefore important to organize and
target agricultural research in ways which ensure con-
tinuing high returns to research, especially in an era of
contracting public budget allocations for agricultural
research [23, 24].
However, the global trends that can be discerned from
Table 2 obscure an enormous degree of spatial variability
in cropping patterns and productivity (and, as will be
discussed below, institutions and market access) [25]. For
instance, these data portray average conditions across
contrasting farming systems (irrigated and rainfed, wet
and dry, and lowland and mountain) quite apart from the
differences between farmers’ access to resources, inputs
and markets, and management practices that are fre-
quently observed in the field. This review identifies high-
lights of experience with farming systems analyses of such
spatial variability during four decades, and proposes a
systematic way of analysing and classifying such variability
for improved agricultural research and development
targeting. The evolving frameworks for understanding
farming systems are outlined in the next, second, section,
followed by the approaches to classifying and targeting
farming systems in the third section. An illustrative
application of classification is presented in the fourth
section, followed by a general discussion in the fifth sec-
tion. The conclusions are summarized in the sixth section.
Frameworks for Understanding Farming Systems
The implications for development of bio-physical and
socio-economic variability and interdependencies in agri-
cultural production are readily apparent from system
analyses [26], which study the characteristics and func-
tions of farming systems as abstractions from reality
characterized by boundaries, components and their
interactions. Systems analyses underpinned some land-
mark assessments of agriculture during the first half of the
previous century [27, 28]. For the purpose of this review,
a farming system can be defined as a particular type of
mixed natural–human system involving managed plant
and animal production [29–31]. Farming systems can be
differentiated at various levels of aggregation, from com-
munity or village level, for instance, comparing valley-
bottom irrigated rice-based farming systems with hillside
rainfed maize–millet farming systems in Himalayan villages,
to continental level, for instance, comparing regional
maize-mixed farming systems in East and Southern Africa
with pastoral–millet farming systems across the West
African Sahel [32]. It follows that farming systems analysis
is the study of properties, structures and functions of
farming systems – in contrast to a farming systems
approach (FSA) that connotes ways of improving agri-
cultural productivity through a better understanding
of farming system function [33]. In common with the
FSA, many other rural development approaches claim a
systems approach to research and development, for
example, community development [34], farmer-first and
participatory development [35], sustainable livelihoods
[36] and sustainable agriculture and rural development
[37–39].
Norman [40] provides a concise overview of the stages
of growth and evolution of the FSA in a turbulent intel-
lectual and institutional environment. The wellspring of
farming systems was farm management survey research in
the 1960s, which confirmed the rationality of smallholder
farmers [41, 42] and the value of increased understanding
of enterprise and livelihood interactions. Farming systems
applications to agricultural research arose in Latin
America [43, 44], Africa [41, 45–47] and Asia [33] during
the 1970s, focused on the rapid diagnostic and on-farm
Table 2 Major food crop areas in developing regions
Crop
Average
1990–1992
(M ha)
Average
2004–2006
(M ha)
Share of 2004–2006
cereals/other main
food crops (%)
Change
1990–1992 to
2004–2006 (%)
Barley 19 16 4 713
Maize 89 99 22 11
Millet 34 32 7 76
Rice, paddy 142 148 33 5
Sorghum 37 38 8 3
Wheat 105 113 25 8
All cereals 433 455 100 5
Starchy Roots 35 44 13 27
Plantain 5 5 2 8
Pulses 58 62 18 6
Oilseeds 93 145 43 56
Fruit 24 36 11 50
Vegetables 24 43 13 81
Other main food crops 238 335 100 41
Source: [15].
http://www.cabi.org/cabreviews
John Dixon, Jon Hellin, Xiaoyun Li and Glenn Hyman 3

research methods that were codified in several major
texts of the early 1980s (e.g. [31]). Throughout the
emergence and development of farming systems over four
decades, analysis methods have evolved steadily [31, 48–
51] and have been applied for many purposes in all agri-
cultural continents of the world [52]. Two threads of
thinking and practice persisted during these four decades:
firstly, a ‘systems’ thread [e.g., [53–55] reflected in the
title of the 13th International Symposium, viz, Systems-
oriented Research in Agricultural and Rural Development
[56]; secondly, and a ‘participation’ thread (e.g. [35, 43]),
which has spun off many variants of participatory research
and evaluation. More recently, these two core threads
were augmented by a third thread on ‘learning’ [51, 57].
A variety of methodologies have been adapted or devel-
oped for farming systems analysis, including for example
systems diagnoses [31, 58], participatory research [59, 60],
game-playing [61] and systems modelling [62, 63].
The FSA was often viewed as a benchmark against
which newer approaches were compared during the
1980s and 1990s. Since the 1970s, the farming systems
research (FSR) experience has been a rich source of con-
ceptual and methodological ideas, which enriched existing
development paradigms [64–67] through a process of in-
duced evolution of thinking and practice reflecting, in the
judgment of the authors, the dominant issues and chal-
lenges of the period. For instance, the roots of partici-
patory development can be traced to the early ‘sondeos’
or rapid diagnoses of farmer constraints in Central
America [43]; and references to multiple livelihoods and
vulnerability were not uncommon in the farming systems
literature of the 1980s. In fact, livelihoods are often
defined in terms of multi-level systems [68] that highlight
their dynamic character and their positioning in a wider
systems context [69–73].
The International Agricultural Research Centres
(IARCs) and the National Agricultural Research Systems
(NARS) experimented with FSAs during the past four
decades, albeit in very different ways [29, 74] – for
example, the International Center for Research Into the
Semi-Arid Tropics (ICRISAT) utilized detailed household
production and consumption data from village level stu-
dies [29], whereas the International Maize and Wheat
Improvement Center (CIMMYT) placed more emphasis
on informal diagnoses in identifiable recommendation
domains with a predetermined focus on maize or wheat
systems [75]. Both CIMMYT and the International Rice
Research Institute (IRRI) invested substantially in building
NARS’ capacity in farming systems analysis capacity [76].
With the support of IARCS, many NARS established on-
farm commodity research programmes supported by
multidisciplinary research teams [77, 78]. Ample funding
during the 1980s enabled a wave of method innovation
during which the theme of ‘experiential learning’ was
added to farming systems thinking [59, 67] and new var-
iants of participatory diagnostic and evaluation methods
were devised and tested [66, 79, 80]. As FSR, farming
systems and extension (FSRE) and farming systems
development (FSD) methodologies were popularized dur-
ing the 1980s existing capacity was stretched rather thin
and, with the natural tendency to mechanistic applications,
quality suffered in some quarters. Although project fund-
ing tightened up during this period, the systems frame of
reference expanded from the initial focus on the farm to
embrace an hierarchy of systems, for example, from field
systems to cropping systems, farming systems and water-
shed systems [26, 81]. Meanwhile, utilization of farming
systems analysis spread to incorporate extension and
development activities.
A great proportion of research in these areas tended
to be micro- or locally oriented, framed by individual
preferences and choices based on available local assets
and resources [52, 82]. Linking local level knowledge and
learning systems to a broader macro- or regional context
has been one of the additions to development thinking
and practice in the 1990s and 2000s [38]. The challenge
remains a real one because the link between livelihood
assets, livelihoods outcomes and policies and institutions
(within a farming systems context) is not straightforward:
for example, the tendency to focus on livelihoods assets
underplays the policy and institutional drivers and modi-
fiers that are identified in the livelihood framework [57, 83].
In reality, the role of cultural, institutional and political
environments in shaping the economic choices of farm
women and men tends to be underdeveloped in farming
systems and livelihood research in particular and agri-
cultural research in general [65, 84], despite the aspirations
expressed in the 2000 CGIAR vision and strategy [85].
Furthermore, over the last 10 years there has been a
growing recognition that the FSA and livelihoods thinking
have not given sufficient emphasis to market access [83].
The role of markets is important since many rural people
depend for their income on their involvement in a range
of markets as private agents or employees, or through
access to credit. In addition, analysing market relations in
the context of farming and livelihood systems is essential
as market processes are important for pro-poor small
farm development and poverty reduction [86, 87]. As a
result, there has been a growing body of work that elu-
cidates theoretical and practical approaches to enhancing
smallholder farmers’ access to input and produce markets
[83, 84] by investigating the institutions (rules of the game
and organizational structures) that either facilitate or limit
access to markets.
Access to agricultural services (market access) can be a
key variable for priority setting and targeting, for example,
the International Center for Tropical Agriculture (CIAT)
tested the Territorial Approach to Rural Business Develop-
ment. The approach, which shares common precepts and
methods with farming systems analyses, includes the
identification and evaluation of market opportunities
through (participatory) value chain analysis, leading to the
design of research and development action plans [88].
Aside from the FSA, there are relatively few modern
http://www.cabi.org/cabreviews
4 Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources

frameworks that embrace, with relative ease, concurrent
natural resource and institutional variability. In a similar
vein, smallholder access to agricultural services can also
serve as a frame of reference for analyses to identify
impact pathways and target agricultural research [89].
Thus, as discussed in the previous paragraphs, the
conceptualization and scope of farming systems analyses
progressively expanded during four decades of field
experience. As the sophistication of analytical methods
increased and participatory research grew in popularity,
the analytical domain expanded and the ratio of variables
to parameters has increased, i.e. the scope and inclu-
siveness of analyses has expanded [90]. Table 3 sum-
marizes salient shifts in the FSA. One of the earlier shifts
was an expansion from a crops focus (e.g. [43]) to crop-
livestock systems (e.g. [91]) and finally to household
livelihoods [89]; from technology generation to include
technology dissemination, support services, business de-
velopment services, as well as social and human capital
(e.g. [68]). A parallel broadening of the scope of institu-
tions and policies that are addressed has occurred with
growing scale and maturity of farming systems analyses
[86, 92, 93].
Approaches to Farming Systems Classification
and Targeting
Despite the high economic returns to agricultural
research [21, 22] and the impressive reduction of urban
poverty [6], many poor farmers in marginal areas have
failed to benefit from agricultural research [94, 95]. In
part, this unfavourable outcome can be attributed to the
failure to employ a systems approach at relevant stages of
the analysis of issues and identification of interventions,
that is, undue reliance on reductionist and commodity-
orientated perspectives that characterize much traditional
agricultural research, development and policy assessments
(e.g. [96]). Although local, village-level, targeting can be
achieved through conventional methods [97], at higher
scales poverty mapping has emerged as an effective tool
to facilitate the targeting of agricultural research, espe-
cially in marginal areas [98–100]. During the past decade
considerable effort has been devoted to mapping the
distribution of rural poverty [101, 102], and some appli-
cations have related alternative poverty metrics to agri-
cultural resource and risk variables, which appears to be a
promising avenue for further research [103, 104].
Table 3 Shifts of emphasis in farming system approach, 1970–present
1970s 1980s 1990s 2000s
System level
Farmer *** *** *** **
Household * *** ***
Community ** ***
Landscape, watershed ***
National, regional *
System focus
Crops *** ***** *** **
Crop–livestock * * ** ***
Multiple livelihood (include off-farm) * ** *****
Value chain (input–farm–output) * ***
Innovation systems focus
Research *** ***** *** **
Research – extension * ** *** **
Research – extension – support services * *** ***
Market actor (institutional) focus
Public *** *** ** ***
Public – civil society * *** ***
Public – private – civil society * ***
Environment
Productivity **** ***** ** ***
Productivity resource management * *** ****
Agro-diversity climate risk * **
Policy
Agricultural development * ** *
Natural resource management * ***
Market access ***
Other emphases
Gender-sensitivity * **** ****
Impact ** ***
ITK * *** ***
Farmer participation * ** ***** ****
Source: Dixon et al. [89] and authors’ interpretations.
Note: Degree of emphasis scored as * (very low) to ***** (very strong).
http://www.cabi.org/cabreviews
John Dixon, Jon Hellin, Xiaoyun Li and Glenn Hyman 5

Nevertheless, the reality is that most small farms are
complex socio-bioeconomic systems whose classification
requires a systems approach that incorporates inter-dis-
ciplinarity, inter-dependence and feedback loops, in order
to differentiate system performance and, in particular,
responses to new technology and institutional innovations
[89, 95]. The decision-making of farm women and men
which determines system responses to innovations is
shaped by the characteristics of their farming systems,
including resources, crops, livestock and other productive
activities, the family goals and the immediate institutional
environment [35, 83]. Relevant analytical frameworks for
the characterization and classification of farming systems
must provide reasonable representations of the com-
plexity of farming systems function and decision-making,
yet be understandable by researchers and development
practitioners.
There is a long tradition of systems-oriented agri-
cultural classifications, which historically emphasized the
crop components of systems, during the 1930s (e.g.
[105]), the 1970s (e.g. [106, 107]), the 1980s (e.g. [108]),
the 1990s (e.g. [109]) and the present decade (e.g.
[110, 111]). Applications focused on crops and land
use were followed by livestock-oriented classifications
[91, 112, 113]. However, there are relatively few well-
integrated, systems-oriented, crop–livestock or multiple-
livelihood classification systems (e.g. [89, 114]).
The challenge of operational categorization of farming
systems lies in the heterogeneity of agricultural resources,
local institutions, farm and household management
and development drivers [25, 89]. Common criteria
for classification relate to technology, cultivars/breeds,
mechanization, crop management and input use intensity,
which are not intrinsically hierarchical. The commonly
used criteria listed above can, in principle, be organized as
a partial hierarchy [115]. Criteria need not be limited to
currently observable characteristics of existing systems:
classifications may also be based on predicted develop-
ment pathways related to productivity, technology and
poverty reduction [116, 117].
Many of these strands are reflected in the FAO–World
Bank farming systems framework and classification [89],
which blends objective data and expert knowledge con-
cerning farming systems and livelihoods at a meso-scale to
identify and describe populations of farm households with
broadly similar resources, livelihoods and vulnerabilities
in developing regions. The classification was based on
information from spatial databases
1
combined with tacit
knowledge of expert panels personally familiar with local
conditions on the ground. The spatially explicit para-
meters that were used in part to differentiate farming
systems included rainfall, temperature, soils, length of
growing period, altitude, relief, environmental constraints,
land cover, irrigation, permanent crops and arable land,
bovines and small ruminants. Farm size, cropping pattern,
off-farm income, livelihood portfolios, market access,
density of services and poverty escape pathways were
derived from household surveys and expert knowledge of
the area in question. With a primary audience of senior
decision-makers in finance and agricultural policy-makers,
this classification identified eight generic farming systems
categories across the developing world and, in finer detail,
72 major regional farming systems, defined as populations
of individual farm household systems with broadly similar
resources, livelihoods and vulnerabilities, and opportu-
nities and constraints, for whom similar development
strategies and interventions might be appropriate. Based
on expert knowledge, the dominant livelihood patterns
were described for each farming system. Within the
framework of broad projections of agricultural produc-
tion and consumption [20], specific trends, emerging
constraints and strategic development priorities for each
farming system were outlined. In addition, the analysis was
extended in two ways: estimating the relative importance
of household strategies to escape poverty (‘backcasting’
from the target of halving the number of poor people by
2015), and consolidating the findings across all regions to
identify promising areas for research and development
investment.
The eight generic farming system categories described
across the six developing regions of the world are listed
in Table 4. The six irrigated and rice-based wetland
systems
2
contain an agricultural population of nearly
900 million people with some 170 M ha of cultivated
land, of which nearly two-thirds is irrigated. There are
three major categories of smallholder rainfed farming
system (in humid, highland or dry/cold areas), which
together contain an agricultural population of more than
1400 million people who operate approximately 540 M ha
of cultivated land. Dualistic systems comprising farms of
mixed size contain a further 200 million farm people with
a cultivated area of 11 M ha. Finally, two further smaller
categories of smallholder system – four coastal artisanal
fishing mixed and six urban-based systems – contain a
combined total of about 100 million people.
An Illustrative Application: Mapping
Intensification Potential Across Farming Systems
Various aggregations of the 72 major farming systems
characterized in the FAO–World Bank farming systems
1
The overlay maps of thematic layers and the farming systems are
available on the FAO website www.fao.org/farmingsystems/ and the
spatial data bases are available through GeoNetwork.
2
One irrigated farming system in Eastern Europe and Central Asia has
relatively large farms and, for the purpose of the present discussion, is
included in the category of dualistic systems.
http://www.cabi.org/cabreviews
6 Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources

spatial databases [89] have been utilized for specific
research and policy purposes. After dropping urban sys-
tems and some consolidation, a 63-system version was
utilized in an analysis of poverty and drought risk of food
crops [104]. For the purpose of the illustrative application
presented in this section, 28 major farming systems were
selected (see Figure 1), which include 75% of the total
cultivated land and 85% of the agricultural population in
the developing world [15, 89]. As indicated in Table 5,
maize, wheat and millet can be found across more
than half of these farming systems, with rice obviously
contributing an important share of food production
especially in the more productive and populous systems
in Asia.
Given that the rate of agricultural development is
strongly influenced by the quality of the agricultural
resource endowment and access to agricultural services
[1, 118], the 28 selected farming systems can be char-
acterized according to these two dimensions that echoes
some earlier analytical frameworks [106], [116]. This
approach provides a more realistic framework for priority
setting and targeting because it reflects both agronomic
Table 4 Global farming systems categories
Category
characteristic
Small-
holder
irrigated
schemes
Wetland
rice based
Rainfed
humid
Rainfed
highland
Rainfed
dry/cold
Dualistic
(large/ small)
Coastal
artisanal
fishing
Urban
based
Total land
(M ha)
219 330 2013 842 3478 3116 70 n.a.
Cultivated area
(M ha)
15 155 160 150 231 414 11 n.a.
Irrigated area
(M ha)
15 90 17 30 41 36 2 n.a.
Agricultural
population (m)
30 860 400 520 490 190 60 40
Number of major
farming systems
3 3 11 10 19 16 4 6
Note: cultivated area refers to both annual and perennial crops.
Source: [42].
Figure 1 Location of 28 major farming systems of the developing world
http://www.cabi.org/cabreviews
John Dixon, Jon Hellin, Xiaoyun Li and Glenn Hyman 7

potential and socio-economic drivers, and shapes the
opportunity sets for intensification or diversification: for
instance, irrigated systems with good market access gen-
erally have a greater number of alternative development
pathways compared with semi-arid remote systems.
The relative importance of agricultural resource quality
and market access for the 28 selected farming systems is
illustrated in Figure 2, where the size of the bubbles
represents the area of cultivated land. For more than two
decades much attention has been focused on resource
degradation [119, 120]. Based on 20 indicators of soil
quality, a composite index of resource constraints [121]
was constructed for the selected 28 farming systems.
A normalization process was used to create a resource
quality index with the range 0 to 1 (where 0 refers to no
reported constraints, and 1 refers to the highest reported
level of resource constraints). Seven of the farming sys-
tems are characterized by relatively light resource con-
straints, with resource quality indices of less than 0.3 and
good agronomic potential for intensification and diversi-
fication. However, six farming systems suffer from rela-
tively severe production constraints with resource quality
indices greater than 0.6, which may limit the range of
crops or livestock which could be profitably produced.
As indicated earlier, the importance of markets in
farming systems development has been widely recognized
[83]. One indicator of market access is the time required
for a farmer to reach the nearest significant market place.
Based on the location of market places, the coverage and
quality of roads and the local terrain, the time to market
has been estimated for each of the 28 farming systems.
The average travel time to market, whether by foot or
vehicle, for most farming systems varies from 2 to 10 h
(only three farming systems with limited cultivated area
are more than 10 h from a market) (see Figure 2).
Farmers’ demands for inputs such as seed, fertilizer and
knowledge can expected to be correlated to their access
to produce (output) markets. As can be seen from
Figure 2, cultivated land in the 28 selected farming systems
is distributed over a wide range of land potential (repre-
sented by the resource constraint index) and a wide range
of market access (represented by the time to market).
Given the current volatility in food prices and the
strong future demand, noted above, for cereal produc-
tion, the intensification of (cereal-based) farming systems
is essential. Figure 3 presents the intensification potential
of the 28 selected farming systems by market access and
resource quality. In this case, the size of the bubble
represents the importance of intensification, assessed by
expert panels, as a poverty escape livelihood strategy; and
its proximity to the junction of the x- and y-axes suggests
better market access and resource quality. In nearly half
(12) of the selected farming systems, farmers have better
than average resource quality and are located, on average,
Table 5 Prevalence of key food crops in 28 major farming systems in developing regions (% of total crop area)
Farming systems Region Maize Wheat Rice Cassava Potato
Sweet
potato Millet Sorghum
Lowland rice EAP 9.8 12.4 45.1 1.7 1.3 4.1 0.2 0.2
Upland intensive mixed EAP 14.8 9.6 42.1 1.5 2.8 3.4 0.8 0.2
Rice–wheat SA 4.6 34.9 28.6 0.0 1.5 0.1 3.3 1.3
Rainfed mixed SA 4.1 7.3 19.6 0.1 0.2 0.0 9.4 8.7
Temperate mixed EAP 34.3 18.8 8.2 0.2 3.4 1.9 2.4 1.9
Rice SA 0.7 2.7 69.5 0.6 1.3 0.2 2.1 1.0
Maize mixed SSA 37.0 1.5 2.2 7.3 1.8 4.5 4.7 6.0
Cereal–root crop mixed SSA 9.9 0.2 5.5 4.1 0.3 3.4 17.6 20.3
Highland mixed SA 14.1 28.1 29.9 0.4 1.4 0.1 4.9 0.9
Highland extensive mixed EAP 14.2 9.7 45.2 0.7 6.2 0.9 1.3 0.1
Root crop SSA 21.6 0.0 11.8 17.4 0.1 11.9 2.3 5.8
Pastoral EAP 16.6 36.7 0.8 0.2 13.4 0.1 1.5 0.6
Agro-past oral millet/sorghum SSA 5.6 0.4 1.8 1.5 0.1 0.6 32.8 21.1
Highland perennial SSA 12.0 0.7 1.1 6.1 2.8 8.9 1.8 5.3
Dry rainfed SA 2.0 4.1 8.5 0.0 0.1 0.0 20.1 19.4
Forest based SSA 22.3 0.8 17.6 8.5 1.0 0.0 0.0 0.3
Highland temper mixed SSA 23.3 20.0 0.4 0.8 0.7 1.6 5.7 15.8
Highland mixed MNA 2.0 59.4 3.2 0.0 1.1 0.0 0.8 2.4
Mixed ECA 25.7 36.8 0.0 0.0 8.0 0.0 0.0 0.0
Rainfed mixed MNA 0.9 57.6 0.2 0.0 2.1 0.0 0.1 0.4
Large scale cereal-veg ECA 9.8 30.0 0.1 0.0 10.6 0.0 1.9 0.1
Extensive cereal-livestock ECA 0.5 79.3 0.7 0.0 3.1 0.0 0.4 0.0
Dryland mixed MNA 3.2 48.1 2.6 0.0 1.6 0.0 0.1 0.1
Maize–beans LAC 54.2 2.7 1.0 0.2 0.7 0.0 0.0 9.5
Extensive mixed LAC 21.0 0.2 13.7 2.5 0.1 0.1 0.0 3.6
Horticulture mixed ECA 12.5 39.9 0.4 0.0 2.7 0.0 0.2 0.0
Cereal–livestock LAC 20.9 7.4 10.2 3.2 0.5 0.3 0.0 0.8
Temperate mixed LAC 13.1 29.4 0.2 0.1 0.4 0.1 0.1 2.3
Note: EAP, Eastern Asia Pacific; SA, South Asia; SSA, Sub-Saharan Africa; MNA, Middle East North Africa; ECA, Eastern Europe and
Central Asia; LAC, Latin America and Caribbean. Data refer to the year 2000.
http://www.cabi.org/cabreviews
8 Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources

within 6 h of a market (from this perspective, farmers
could visit a market and return home in a single, long,
day). These are systems with good prospects for the
adoption of improved varieties, crop management tech-
nologies and the use of external inputs and services.
From the perspective of rapid intensification and on-farm
diversification, these 12 farming systems should be con-
sidered for early targeting.
Discussion
The responses of farm households to evolving resources,
technologies, markets and other institutions can be
categorized into five poverty escape strategies: (i) inten-
sification of existing patterns of farm production; (ii)
diversification, including market-oriented, value-added
and post-harvest activities; (iii) increased operated farm,
herd or enterprise size, including consolidation of existing
holdings and the expansion of the agricultural frontier;
(iv) increased off-farm income to supplement or replace
on-farm activities; and (v) exit from agriculture, often
involving migration from rural areas [89]. Population-
driven intensification has been well-described in the lit-
erature [105, 116, 122, 123] Enterprise and income
diversification is a common farmer response to changing
resource ratios and market access [8, 124] and many
governments now have programmes to support farm-
level diversification. There is widespread recognition of
the growing importance of off-farm income for small-
holder households [1, 125]. While the strategies of
increasing farm size and exit from agriculture were
common in the evolution of agriculture in many Organi-
sation for Economic Co-operation and Development
(OECD) countries, they have received less attention in
developmental literature. What is missing from knowl-
edge bases supporting development practitioners is the
contextual analysis of the factors influencing the merits
and feasibility of these different strategies for farmers in
the different major farming systems of the world.
One analysis has suggested that in aggregate terms, on-
farm improvements (i.e. intensification, diversification and
increased farm size) would be a greater source of poverty
reduction than off-farm sources (i.e. off-farm income and
exit from agriculture) [89]. Within the category of farm
improvement, diversification is expected to be the key
strategy in a majority of farming systems – benefiting from
the higher income elasticities and expanding local demand
for many non-traditional and processed agricultural pro-
ducts; a demand stimulated by economic growth and
higher incomes levels [19]. The intensification of existing
patterns of production, which include cereals in a majority
of cases, continues to be a crucial source of farm income
growth and poverty reduction in a majority of farming
systems. Finally, a certain proportion of poor farmers
will also benefit by expanding their operational asset
base through increased farm size as land is consolidated,
the agricultural frontier expands in some rainfed humid
farming systems (notably in Latin America and Sub-
Saharan Africa), or land rental markets improve
3
.
Apart from farm improvement options, off-farm
income already contributes a major part of the household
income of many poor farmers, and further increases are
expected to be the second greatest source of aggregate
poverty reduction in future years [87, 126]. In a similar
vein, the exit of farm household members or the entire
family from agriculture within a particular farming system
is expected to be an increasingly common phenomenon,
and is forecast to be of particular importance among
smallholders in rainfed highland and dryland systems.
Globally, diversification (including on-farm processing and
0.0
0.2
0.4
0.6
0.8
1.0
0 5 10 15 20
Access to market (h)
Resource constraints (index)
Figure 2 Cultivated land by farming system potential.
Note: Access to market is estimated by the average min-
utes from all pixels of the farming systems to market;
resource quality is estimated by normalized composite
index of 20 characteristics of land, water and soil. The sizes
of the bubbles correspond to the areas of cultivated land in
each of the 28 farming systems
0.0
0.2
0.4
0.6
0.8
1.0
0 5 10 15 20
Access to market (h)
Resource constra ints (index)
Figure 3 Intensification by farming system potential.
Note: See notes below Figure 2 for explanations of market
access and resource constraints indices. The sizes of the
bubbles represent the relative intensification potential
scored by expert panels from 1 (low potential) to 5 (high
potential) in each of the 28 farming systems
3
In addition, some poor pastoralists may succeed in expanding their herd
size, or poor urban producers may expand their volume of production.
http://www.cabi.org/cabreviews
John Dixon, Jon Hellin, Xiaoyun Li and Glenn Hyman 9

other value-added activities) appears to be an important
household strategy in many situations [8].
However, the great variability in the relative impor-
tance of these strategies across farming systems needs to
be recognized by policy-makers. In some cases, research
managers or policy-makers may need empirically based
indicators for the allocation of resources across the
developing world (which could be based on the above
illustration), at the meso-level, i.e. across farming systems
within a region, or at the micro-level, i.e. within a farming
system. An example of a global resource allocation pro-
blem occurred in the case of the prioritization of food
crop improvement traits by the Generation Challenge
Program. In a collaborative effort across CGIAR Centers,
a multidisciplinary group utilized the above farming system
framework to identify priority farming systems for
drought-breeding targeted to poverty reduction [104].
In other cases, policy-makers may be focused on one
specific farming system, i.e. a micro- as opposed to meso-
level focus [93].
A combination of agricultural resource quality and the
market access can also be used to guide priority setting
and targeting within a specific farming system, such as the
rice–wheat farming system found in the Indo-Gangetic
plains of northern India [100]. In India, there is a high
concentration of rural poor in low agricultural potential
areas compared with high potential areas (e.g. rainfed
versus irrigated [127, 128]). Furthermore, there are data
to suggest that poverty declined more in low potential
areas between 1972 and 1987 than in high potential areas
[129]
4
. This has important implications for where invest-
ments should be targeted in order to achieve further
productivity growth and poverty alleviation [101, 129].
In the rice–wheat farming case, spatial poverty maps were
developed that can serve as a tool to guide priority-setting
and targeting within the Rice–Wheat Consortium [100].
The study used 18 quantitative, spatially explicit variables,
which serve as indicators of poverty based on the five
livelihood capitals of the livelihoods approach: natural,
social, human, physical and financial. The approach albeit
at the level of the farming system is similar to the devel-
opment domain approach pioneered by the International
Food Policy Research Institute (IFPRI) [131].
Conclusions
The substantial increase in food production, especially
from cereal crops, which is required to ensure food
security and improved livelihoods, demands effective
agricultural research that is well targeted to the circum-
stances of different types of farmers. The classification
of agriculture provides a valuable framework for the
targeting of agricultural research and development efforts.
The FAO–World Bank classification of farming systems
outlined in this review blended wide-ranging expert
judgement with secondary data including spatial data on
population, resource use and climate. The classification is
complemented by data of characteristics, trends, emer-
ging constraints and strategic research priorities of some
72 broad farming systems in developing regions, organized
around five leading areas of rural change: natural re-
sources, technologies, markets, policies/institutions and
information/human capital.
Building on the sustainable livelihoods concepts of
multiple assets and vulnerability, the FAO/World Bank
study assessed poverty escape pathways for each broad
farming system, and estimated the relative importance of
intensification, diversification, increased farm or enter-
prise size, increased off-farm income and exit from agri-
culture. By focusing on defined combinations of resource
quality and market access for groups of farming systems,
and on the relative importance of intensification and
diversification pathways out of poverty, the farming
systems framework creates the basis for the systematic
determination of priorities for agricultural research, for
the targeting of the subsequent dissemination of scientific
knowledge produced by research, and for the ultimate
impact assessment of research and development efforts.
References
1. World Bank. World Development Report 2008: Agriculture for
Development. World Bank, Washington, DC; 2007.
2. UN Millennium Project 2005. Halving Hunger: It Can Be
Done. Summary version of the report of the Task Force on
Hunger. The Earth Institute, Columbia University, New York;
2005.
3. FAO. State of Food and Agriculture. FAO, Rome, Italy; 2007.
4. Sen AK. Poverty and Famines: An Essay on Entitlement and
Deprivation. Clarendon Press, Oxford, UK; 1981.
5. Hazell P, Poulton C, Wiggins S, Doward A. The Future of
Small Farms for Poverty Reduction and Growth. 2020
Discussion Paper 42. IFPRI, Washington, DC; 2007.
6. McCalla AF. Agriculture in the 21st Century. Fourth
Distinguished Economist Lecture. CIMMYT, Mexico, DF,
Mexico; 2000.
7. Runge CF, Senauer B, Pardey PG, Rosegrant M. Ending
Hunger in our Lifetime: Food Security and Globalization.
IFPRI-John Hopkins University Press, Baltimore, MD; 2003.
8. Joshi P, Gulati A, Cummings R. Agricultural Diversification
and Smallholders in South Asia. Academic Foundation,
New Delhi, India; 2007.
9. Hellin J. Better Land Husbandry: From Soil Conservation to
Holistic Land Management. Science Publishers, New Delhi,
India; 2006.
10. Lal R. Soil science and the carbon civilization. Soil Science
Society of America Journal 2007;71:1425–37.
4
The greater reduction in low potential versus high potential areas has,
however, been disputed: see, for example, Palmer-Jones [130].
http://www.cabi.org/cabreviews
10 Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources

11. Von Braun J. Rising Food Prices: What Should be Done?
IFPRI Policy Brief. International Food Policy Research
Institute, Washington, DC; 2008.
12. OECD-FAO. Agricultural Outlook 2007–2016. OECD, Paris,
France; 2007.
13. Rabbinge R, van Latesteijn HC. Long-term options for land
use in the European community. Agricultural Systems
1992;40:195–210.
14. Borlaug NE. Sixty-two years of fighting hunger: personal
recollections. Euphytica 2007;157:287–97.
15. FAOSTAT. Available from: URL:http://faostat.fao.org/site/
567/default.aspx (accessed July 2007).
16. Von Braun J. The World Food Situation: New Driving Forces
and Required Actions, IFPRI’s Biannual Overview of the
World Food Situation. CGIAR Annual General Meeting,
Beijing, 4 December 2007.
17. Tilman D, Fargime J, Wolff B, D’Antonio C, Dobson A,
Howarth R, et al. Forecasting agriculturally driven global
environmental change. Science 2001;292(5515):281–4.
18. Pardey PG, Alston JM, Piggott RR. Agricultural R&D in the
Developing World: Too Little, Too Late? IFPRI, Washington,
DC; 2006.
19. Gulati A, Dixon J, editors, Maize in Asia: Changing Markets
and Incentives. CIMMYT- IFPRI-IFAD-Academic Publishers,
Washington, DC, USA; Mexico, DF, Mexico; New Delhi,
India; 2008.
20. Bruinsma J, editor. World Agriculture: Towards 2015/2030.
A FAO Perspective. FAO-Earthscan, London, UK; 2002.
21. Evenson RE, Gollin D. Assessing the impact of the green
revolution, 1960 to 2000. Science 2003;300(5620):758–62.
22. Raitzer D, Kelly T. Benefit-cost meta-analysis of investment
in the International Agricultural Research Centers of the
CGIAR. Agricultural Systems 2008;96:108–23.
23. Anderson JR, Hardaker JB. Efficacy and efficiency in
agricultural research: a systems view. Agricultural Systems
1992;40:105–23.
24. Delmar DP. Agriculture in the developing world: connecting
innovations in plant research to downstream applications.
Proceedings of the National Academy of Sciences of the
United States of America 2005;102(44):15739–46.
25. Weber G. Heterogeneity and complexity in farming systems:
towards an evolutionary perspective. Journal of Farming
Systems Research-Extension 1996;6(2):15–32.
26. Conway G. The Doubly Green Revolution: Food for All in the
21st Century. Penguin, London, UK; 1997.
27. Richards AT. Land Labour and Diet in Northern Rhodesia.
Oxford University Press, Oxford; 1939.
28. de Schlippe B, Batwell BL. Preliminary study of Nyangwara
system of agriculture in the southern Sudan. Africa
1955;XXV(4):321–50.
29. Farrington J, Martin AM. Farmer participation in agricultural
research: a review of concepts and practices. Agricultural
Administration Unit Occasional Paper no 9, Overseas
Development Institute, London, UK.
30. Merrill-Sands D. Farming systems research: clarification
of terms and concepts. Experimental Agriculture
1986;22:87–104.
31. Shaner PF, Philipp, Schmehl WR. Farming Systems
Research and Development: A Guideline for Developing
Countries. Westview Press, Boulder, CO; 1982. ISBN
0-86531-389-X.
32. Garrity DP, Zandstra HG, Harwood RR. A classification of
Philippine upland rice growing environments for use in
cropping systems research. Philippines Journal of Crop
Science 1978;3(1):25–37.
33. Harwood R. Small Farm Development: Understanding and
Improving Farming Systems in the Humid Tropics. Westview
Press, Boulder, CO; 1979.
34. Oakley P. Review Article, Community development in the
third world in the 1990s. Community Development Journal
1998;33:365–76.
35. Chambers R. Rural Development: Putting the Last First.
Longman, London, UK; 1984.
36. Ellis F. Rural Livelihoods and Diversity in Developing
Countries. Oxford University Press, Oxford, UK; 2000.
37. Harris G. Seeking Sustainability in an Era of Complexity.
Cambridge University Press, Cambridge, UK; 2007.
38. Pretty J. The Earth Only Endures: On Reconnecting with
Nature and Our Place in it. Earthscan, London, UK; 2007.
39. Clark WC, Dickson NM. Sustainability science: the emerging
research program. Proceedings of the National Academy of
Sciences of the United States of America 2003;100:8059–61.
40. Norman DW. The farming systems approach: a historical
perspective. In: Proceedings of the 17th Symposium of the
International Farming Systems Association, November 2002,
Lake Buena Vista, Florida; 2002. CD-ROM.
41. Norman DW. Rationalizing mixed cropping under indigenous
conditions: the example of northern Nigeria. Journal of
Development Studies 1974;11(1):3–21.
42. Schultz TW. Transforming Traditional Agriculture. University
of Chicago Press, Chicago, IL; 1964.
43. Hildebrand P. Generating Technology for Traditional
Farmers: A Multidisciplinary Methodology. ICTA, Guatemala
City; 1976.
44. Rhoades R. The Art of the Informal Agricultural Survey.
Training Document 1982–2, Social Science Department.
International Potato Center, Lima, Peru; 1981.
45. Collinson M. Transferring technology to developing
economies: the example of applying farm management
economics in traditional African agriculture. World
Development 1974;2(2):93–7.
46. Dixon J. Farming Systems Research in Ethiopia, Processing
of ORSTOM/CVRS Symposium on Land-use and
Development in Africa South of Sahara-Smallholder’s Logic
and Technical Rationality, April 1978, Ouagadougou, Upper
Volta; 1978.
47. Gilbert EH, Norman DW, Winch EE. Farming Systems
Research: A Critical Appraisal. MSU Rural Development
Paper No. 6. Michigan State University, East Lansing, USA;
1980.
48. Biggs SD. A multiple source of innovation model of
agricultural research and technology promotion. Agricultural
Administration (Research and Extension) Network
Discussion Paper 16, Overseas Development Institute,
London; 1989.
http://www.cabi.org/cabreviews
John Dixon, Jon Hellin, Xiaoyun Li and Glenn Hyman 11

49. Tripp R, editor. Planned Change in Farming Systems:
Progress in On-Farm Research. John Wiley and Sons,
Chichester, UK; 1991.
50. Norman DW, Worman FD, Siebert JD, Modiakgotla E. The
Farming Systems Approach to Development and Appropriate
Technology Generation. FAO, Rome, Italy; 1995.
51. Dixon J, Neely C, Lightfoot C, Avila M, Baker D,
Holding-Anyonge C, et al. editors. Farming systems and
poverty: making a difference. In: Proceedings of the 18th
International Symposium of the International Farming
Systems Association: A Global Learning Opportunity,
31 October–1 November 2005, Rome, Italy. International
Farming Systems Association, Rome, Italy; 2006. Available
from: URL: http://www.fao.org/farmingsystems/
52. Collinson M. A History of Farming Systems Research. CABI
Publishing, Wallingford, UK. Rome, Italy; 2000.
53. Checkland PB. Systems Thinking, Systems Practice. John
Wiley, Chichester, UK; 1981.
54. Bawden RJ, Ison RL, Macadam RD, Packham RG, Valetine I.
A research paradigm for systems agriculture. In: Remenyi JV,
editor. Agricultural Systems Research for Development
Countries. Australian Centre for International Agricultural
Research, Canberra, Australia; 1985. p. 31–45.
55. Ison RL, Maiteny PT, Carr S. Systems methodologies for
sustainable natural resources research and development.
Agricultural Systems 1997;55(2):257–72.
56. Sebillotte M, editor. Systems-Oriented Research in
Agriculture and Rural Development. CIRAD, Montpellier,
France; 1996.
57. Scoones Ian. Livelihoods perspectives and rural
development. Journal of Peasant Studies
2009;36(1):171–96.
58. Tre
´buil G, Kam SP, Turkelboom F, Shinawatra B. Systems
diagnoses at field, farm and watershed levels in diversifying
upland agroecosystems: towards comprehensive solutions
to farmers’ problems. In: Teng PS, et al., editors. Systems
Approaches for Sustainable Agricultural Development:
Applications of Systems Approaches at the Farm and
Regional Levels. Kluwer Academic Publishers and IRRI,
UK; 1997. p 99–114.
59. Lightfoot C, Bimbao MA. Farmer participatory procedures
for managing and monitoring sustainable farming systems.
Journal of the Asian Farming System Association
1993;2:61–87.
60. Chambers R. The origin and practice of participatory rural
appraisal. World Development 1994;22(7):953–69.
61. d’Aquino P, le Page C, Bousquet F, Bah A. Using self-
designed role-playing games and a multi-agent system to
empower a local decision-making process for land use
management: the SelfCormas experiment in Senegal.
Journal of Artificial Societies and Social Simulation
2003;6(3):5. Available from: URL: http://
jasss.soc.surrey.ac.uk/6/3/5.html
62. Clayton E. Economic Planning in Peasant Agriculture.
Department of Economics Monographs, Wye College,
University of London, London; 1963.
63. Gurung TR, Bousquet F, Tre
´buil G. Companion modeling,
conflict resolution, and institution building: sharing irrigation
water in the Lingmuteychu Watershed, Bhutan. Ecology and
Society 2006;11(2):36. Available from: URL: http://
www.ecologyandsociety.org/vol11/iss2/art36/
64. Berdegue JA. Challenges of farming systems research and
extension. Journal of Farming Systems Research-Extension
1993;4:1–10.
65. Butler-Flora C, Flora JL. Rural Communities,
Legacy+Change. 3rd edn. Westview Press, Boulder, CO;
2008.
66. Farrington J. The development of diagnostic methods in FSR.
In: Collinson M, editor. A History of Farming System
Research. CABI Publishing, Wallingford, Oxon, UK; 2000.
p 59–66.
67. Langeveld H, Roling N, editors. Changing European Farming
Systems for a Better Future: New Visions for Rural Areas.
Wageningen Academic Publishers, Wageningen, The
Netherlands; 2006.
68. Ellis F. Livelihood diversification and sustainable rural
livelihoods. In: Carney D, editor. Proceedings of Department
for International Development’s Natural Resources Advisers’
Conference, July 1998. DFID, London; 1998.
69. Allison EH, Horemans B. Putting the principles of the
sustainable livelihoods approach into fisheries development
policy and practice. Marine Policy 2006;30:757–66.
70. Bebbington A. Capitals and capabilities: a framework for
analyzing peasant viability, rural livelihoods and poverty.
World Development 1999;27:2021–44.
71. Bird K, Shepherd A. Livelihoods and chronic poverty in
semi-arid Zimbabwe. World Development 2003;31:591–610.
72. Bond R, Mukherjee N. Livelihood asset status tracking: an
impact monitoring tool? Journal of International Development
2002;14:805–15.
73. Kristjanson P, Radeny M, Baltenweck I, Ogutu J, Notenbaert
A. Livelihood mapping and poverty correlates at a meso-level
in Kenya. Food Policy 2005;30(5–6):568–83.
74. Anderson JR, Herdt RW, Scobie GM. Science and Food: the
CGIAR and its Partners. World Bank, Washington, DC; 1988.
75. Byerlee D, Hesse de Polanco E. Farmers’ stepwise adoption
of technological packages: evidence from the Mexican
Altiplano. American Journal of Agricultural Economics
1986;63:519–27.
76. Anandajayasekram P. The history of FSR training in east,
central and southern Africa. In: Collinson M, editor.
A History of Farming System Research. CABI Publishing,
Wallingford, Oxon, UK; 2000. p 225–233.
77. Benoit-Cattin M. Senegal’s experimental units. In:
Collinson M, editor. A History of Farming System Research.
CABI Publishing, Wallingford, Oxon, UK; 2000. p 178–183.
78. Kean SA, Ndiyoi MC. The institutionalization of FSR
in east and southern Africa: an overview. In: Collinson M,
editor. A History of Farming System Research.
CABI Publishing, Wallingford, Oxon, UK; 2000.
p 208–214.
79. Hellin J, Bellon MR, Badstue L, Dixon J, La Rovere R.
Increasing the impacts of participatory research.
Experimental Agriculture 2008;44:81–95.
80. Lilja N, Dixon J. Responding to the challenge of impact
assessment of participatory research and gender analysis.
Experimental Agriculture 2008;44:3–19.
81. Hart RD. An ecological conceptual framework for agricultural
research and development. In: Shanner WW, Phillipp PE,
Schmehl WR, editors. Readings in Farming Systems
http://www.cabi.org/cabreviews
12 Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources

Research and Development. Westview Press, Boulder, CO;
1980. p. 44–58.
82. FAO. Guidelines for the Conduct of a Training Course in
Farming Systems Development. FAO, Rome, Italy; 1990.
83. Dorward A, Kydd J, Morrison J, Urey I. A policy agenda for
pro-poor agricultural growth. World Development
2003;32(1):73–89.
84. Cernea MM, Kassam AH. Researching the Culture in
Agri-Culture. CABI Publishing, Wallingford, Oxon, UK; 2006.
85. Technical Advisory Committee. A Food Secure World for All:
Towards a New Vision and Strategy for the CGIAR. Technical
Advisory Committee, TAC Secretariat, Rome, Italy; 2000.
86. Kuyvenhoven A. Creating an enabling environment: Policy
conditions for less-favoured areas. Food Policy
2004;29(4):407–30.
87. Janvry DA, Sadoulet E. 2000. Rural poverty in Latin
America: determinants and exit paths. Food Policy
2000;25(4):389–409.
88. Hellin J, Griffith A, Albu M. Mapping the market: market-
literacy for agricultural research and policy to tackle rural
poverty in Africa. In: Almond FR, Hainsworth SD, editors.
Beyond Agriculture – Making Markets Work for The Poor.
Proceedings of an International Seminar February–March
2005, Westminster, London, UK. Crop Post-Harvest
Programme, Natural Resources International Limited,
Aylesford, Kent, UK. Practical Action, Bourton-on-Dunsmore,
Warwickshire, UK, 2005. p. 109–48.
89. Dixon J, Gulliver A, Gibbon D. Farming Systems and Poverty:
Improving Farmers’ Livelihoods in a Changing World. FAO,
Rome, Italy; World Bank, Washington, DC; 2001.
90. Norman D, Matlon P. Systems research and technical
change. In: Crawford J and Collin J, editors. Research on
Agricultural Systems: Accomplishments, Perspectives and
Methods. Nova Science Publishers, Huntington, UK; 2000.
p. 19–47.
91. Sere C, Steinfeld H. World Livestock Production Systems:
Current Status, Issues and Trends. FAO Animal Production
and Health Paper 127. FAO, Rome, Italy; 1996.
92. Dixon J. A farming system contribution to agricultural policy
analysis. In: Collinson M, editor. A History of Farming System
Research. CABI Publishing, Wallingford, Oxon, UK; 2000.
93. Caldwell JS, Akobundu E. Agricultural systems and policy:
systems-oriented approaches to agriculture and rural
development. Journal for Farming Systems Research and
Extension 1997;6(3).
94. Bellon M. Crop research to benefit poor farmers in marginal
areas of the developing world: a review of technical
challenges and tools. CAB Reviews: Perspectives in
Agriculture, Veterinary Science, Nutrition and Natural
Resources 2006;1(70):1–11.
95. Byerlee D. Targeting poverty alleviation in priority setting for
agricultural research. Food Policy 2000;25(4):429–45.
96. Clay JW. World Agriculture and the Environment: A
Commodity-by-Commodity Guide to Impacts and Practices.
Island Press, Washington, DC, USA; 2004.
97. La Rovere R, Aw-Hassan A, Turkelboom F, Thomas R.
Targeting research for poverty reduction in marginal areas of
rural Syria. Development and Change 2006;37(3):627–48.
98. Bellon MR, Hodson D, Bergvinson D, Beck D, Martinez-
Romero E, Montoya J. Targeting agricultural research to
benefit poor farmers: relating poverty mapping to maize
environments in Mexico. Food Policy 2005;30(5/6):476–92.
99. Davis B. Choosing a Method for Poverty Mapping. FAO
Agriculture and Economic Development Analysis Division,
FAO, Rome, Italy; 2003.
100. Erenstein O, Hellin J, Chandna P. Livelihoods, Poverty and
Targeting in the Indo-Gangetic Plains: A Spatial Mapping
Approach. RWC CIMMYT, New Delhi, India; 2007.
101. Bigman D, Fofack H. Geographical targeting for poverty
alleviation: an introduction to the special issue. World Bank
Economic Review 2000;14:129–45.
102. Pachico D, Hertford R, de Janvry A. Assessing the impact of
agricultural research on poverty alleviation: some issues and
priorities. Food Policy 2000;25(4):379–88.
103. Hyman G, Larrea C, Farrow A. Methods, results and policy
implications of poverty and food security mapping
assessments. Food Policy 2005;30(5–6):453–60.
104. Hyman GG, Fujisaka S, Jones PG, Wood S, de Vicente C,
Dixon J. Strategic approaches to targeting technology
generation: assessing the coincidence of poverty and
drought-prone crop production. Agricultural Systems
2008;98(1):50–61.
105. Whittelsey D. Major agricultural regions of the earth.
Annals of the Association of American Geographers
1936;26(4):199–240.
106. Ruthenberg H. Farming Systems in the Tropics. Clarendon
Press, Oxford, UK; 1971.
107. Grigg DB. The Agricultural System of the World: An
Evolutionary Approach. Cambridge University Press,
London, UK; 1974.
108. Bunting AH, editor. Agricultural Environments.
CAB Publishing, Wallingford, UK; 1987.
109. Beets WC. Raising and Sustaining Productivity of
Smallholder Farming Systems in the Tropics. AgBe
Publishing, Alkmaar, Holland; 1990.
110. Wood S, Sebastian K, Scherr SJ. Pilot Analysis of Global
Ecosystems: Agroecosystems. IFPRI-WRI, Washington, DC;
2000. Available from: URL: http://www.ifpri.org.
111. Leff B, Ramankutty N, Foley JA. Geographic distribution of
major crops across the world. Global Biogeochemical Cycles
2004;18:GB1009. doi:10,1029/2003GB002108
112. White D. A Global Agro-Climatic Analysis of the Distribution
and Production of Livestock Commodities. ACIAR, Canberra,
Australia; 1998.
113. Kruska RL, Reid RS, Thornton PK, Henninger N, Kristjanson
PM. Mapping livestock-oriented agricultural production
systems for the developing world. Agricultural systems
2003;77:39–63.
114. Fernandez-Rivera S, Okike I, Manyong V, Williams TO,
Kruska RL, Tarawali S. Classification and description of the
major farming systems incorporating ruminant livestock in
West Africa. In: Williams TO, Tarawali S, Hiernaux P,
Fernandez-Rivera S, editors. Sustainable Crop–Livestock
Production for Improved Livelihoods and Natural Resource
Management in West Africa. Proceedings of an International
Conference Held at International Institute of Tropical
Agriculture (IITA), 19–22 November 2001, Ibadan, Nigeria.
http://www.cabi.org/cabreviews
John Dixon, Jon Hellin, Xiaoyun Li and Glenn Hyman 13

Nairobi, Kenya: ILRI. Technical Centre for Agricultural and
Rural Cooperation, Wageningen, The Netherlands; 2004.
115. Fresco L, Westphal E. Hierarchical classification of farming
systems. Experimental Agriculture 1988;24:399–419.
116. Boserup E. The Conditions of Agricultural Growth: The
Economics of Agrarian Change Under Population Pressure.
Aldine, Chicago, USA; 1965.
117. Pender JL, Jagger P, Nkonya E, Sserunkuuma, D.
Development pathways and land management in Uganda.
World Development 2004;32(5):767–92.
118. Wiggins S. Smallholder farming in Africa. In: Belshaw D,
Livingstone I, editors. Renewing Development in
Sub-Saharan Africa: Policy, Performance and Prospects.
Rutledge, London, UK; 2002.
119. Smaling EMA, Dixon J. Adding a soil fertility dimension to the
global farming systems approach, with cases from Africa
agriculture. Ecosystems and Environment 2006;116:15–26.
120. Shaxson TF. New Concepts and Approaches to Land
Management in the Tropics with Emphasis on Steeplands.
Soils Bulletin 75. FAO, Rome, Italy; 1999.
121. Sanchez P, Palm C, Buol S. Fertility capability soil
classification: a tool to help assess soil quality in the tropics.
Geoderma 2003;114(3–4):157–85.
122. Pingali P, Bigot Y, Binswanger H. Agricultural Mechanisation
and the Evolution of Farming Systems in Sub-Saharan Africa.
Johns Hopkins, Baltimore, MD; 1987.
123. Smith J, Barau AD, Goldman A, Mareck JH. The role of
technology in agricultural intensification: the evolution of
maize production in the Northern Guinea Savannah of
Nigeria. Economic Development and Cultural Change
1993;42(3):537–54.
124. Tiffen M, Motimore M, Gichuki F. More People, Less Erosion:
Environmental Recovery in Kenya. John Wiley and Sons Ltd,
Chichester, UK; 1994.
125. Reardon T, Berdegue J. Rural nonfarm employment and
incomes in Latin America: overview and policy implications.
World Development 2001;29:395–409.
126. Barrett CB, Reardon T, Webb P. Nonfarm income
diversification and household livelihood strategies in rural
Africa: concepts, dynamics, and policy implications. Food
Policy 2001;26(4):315–31.
127. Fan S, Hazell P. Should India Invest More in Less-Favored
Areas? IFPRI, Washington, DC; 1997.
128. Byerlee D, Morris M. Research for marginal environments:
are we underinvested? Food Policy 1993;5:381–93.
129. Fan SG, Hazell P, Haque T. Targeting public investments
by agro-ecological zone to achieve growth and poverty
alleviation goals in rural India. Food Policy 2000;25:411–28.
130. Palmer-Jones R. Agricultural growth, poverty reduction
and agro-ecological zones in India: an ecological fallacy?
Food Policy 2003;28(5–6):423–31.
131. Pender JL, Place F, Ehui S. Strategies for Sustainable
Agricultural Development in the East African Highlands.
EPTD Discussion Paper 41. IFPRI, Washington, DC; 1999.
http://www.cabi.org/cabreviews
14 Perspectives in Agriculture, Veterinary Science, Nutrition and Natural Resources