1
Exploring future changes in smallholder farming systems by linking socio-economic
1
scenarios with regional and household models
2
3
Mario Herrero
1*
, Philip K. Thornton
2,3
, Alberto Bernués
4
, Isabelle Baltenweck
3
, Joost
4
Vervoort
2,5
, Jeannette van de Steeg
3
, Stella Makokha
6
, Mark T. van Wijk
3
, Stanley
5
Karanja
3
, Mariana C. Rufino
3
and Steven J. Staal
3
6
7
1
Commonwealth Scientific and Industrial Research Organisation, 306 Carmody Road, St
8
Lucia, 4067 QLD, Australia
9
2
CGIAR Research Programme on Climate Change, Agriculture and Food Security
10
(CCAFS), PO Box 30709-00100, Nairobi, Kenya
11
3
International Livestock Research Institute (ILRI), PO Box 30709-00100, Nairobi, Kenya
12
4
AgriFood Research and Technology Centre of Aragon, Avda. Montaña 930, 50059 –
13
Zaragoza, Spain
14
5
Environmental Change Institute, University of Oxford, OX1 3QY, UK
15
6
Kenyan Agricultural Research Institute (KARI), Nairobi, Kenya
16
17
18
*Corresponding author: Dr Mario Herrero, CSIRO, 306 Carmody Road, St Lucia,
19
4067, QLD, Australia, email: Mario.herrero@csiro.au; tel +61 7 3214 2538.
20
21
22
23
24
25
26
27
28
29
30
31
2
Abstract
1
2
We explore how smallholder agricultural systems in the Kenyan highlands might
3
intensify and/or diversify in the future under a range of socio-economic scenarios. Data
4
from approximately 3000 households were analysed and farming systems characterized.
5
Plausible socio-economic scenarios of how Kenya might evolve, and their potential
6
impacts on the agricultural sector, were developed with a range of stakeholders. We study
7
how different types of farming systems might increase or diminish in importance under
8
different scenarios using a land-use model sensitive to prices, opportunity cost of land
9
and labour, and other variables. We then use a household model to determine the types of
10
enterprises in which different types of households might engage under different socio-
11
economic conditions. Trajectories of intensification, diversification, and stagnation for
12
different farming systems are identified. Diversification with cash crops is found to be a
13
key intensification strategy as farm size decreases and labour costs increase. Dairy
14
expansion, while important for some trajectories, is mostly viable when land available is
15
not a constraint, mainly due to the need for planting fodders at the expense of cropland
16
areas. We discuss the results in relation to induced innovation theories of intensification.
17
We outline how the methodology employed could be used for integrating global and
18
regional change assessments with local-level studies on farming options, adaptation to
19
global change, and upscaling of social, environmental and economic impacts of
20
agricultural development investments and interventions.
21
22
23
Keywords
24
25
Scenarios, smallholders, household modeling, land use modeling, sustainable
26
intensification, agriculture, dairy
27
28
29
30
1. Introduction
31
3
1
The role that smallholder agricultural producers are likely to play in global food
2
production and food security in the coming decades is highly uncertain. In many parts of
3
the tropics, particularly sub-Saharan Africa, smallholder production is critical to the food
4
security of the poor. Industrialisation of agricultural production is occurring in many
5
places, largely in response to burgeoning demand for food. Some smallholders may be
6
able to seize the opportunities that exist and develop, and operate as sustainable and
7
profitable smallholder agricultural production systems (Herrero et al., 2010; Thornton,
8
2010). Whether large numbers of smallholders will be able to do this in a carbon-
9
constrained global economy and in an environment characterised by a changing climate
10
and by increased climatic variability, will depend on many things such as increasing
11
regulation, building social protection and strengthening links to urban areas, and
12
substantial investment in agriculture (Wiggins, 2009; World Bank, 2009). Understanding
13
how smallholder systems may evolve in the future is critical if poverty alleviation and
14
food security goals are to be achieved.
15
In many parts of the tropics, particularly Africa and Asia, smallholders operate
16
mixed crop-livestock systems, which integrate different enterprises on the farm; crops
17
provide food for consumption and for cash sales, as well as residues to feed livestock, and
18
livestock provide draft power to cultivate the land and manure to fertilise the soil. These
19
systems are often highly diversified, and the synergies between cropping and livestock
20
keeping offer real opportunities for raising productivity and increasing resource use
21
efficiency (Herrero et al., 2010). Whether these systems can increase household incomes
22
and enhance the availability of and access to food for rapidly increasing urban
23
populations in the coming years, while at the same time maintaining environmental
24
services, is a question of considerable importance.
25
Studying this question requires some consideration of theories of change. A
26
general model of agricultural intensification originated with Boserup (1965), who
27
described it as an endogenous process responding to increased population pressure. As
28
the ratio of land to population decreases, farmers are induced to adopt technologies that
29
raise returns to land at the expense of a higher input of labour. The direct causal factor is
30
relative factor price changes, in accordance with the theory of induced innovation. At
31
4
low human population densities, production systems are extensive, with high availability
1
of land and few direct crop-livestock interactions. Population increases lead to increases
2
in demand for crop and livestock products, which in turn increases the value of manure
3
and feed resources and other inputs, leading to increased crop and livestock productivity.
4
As population increases yet further, systems intensify through specialisation or
5
diversification in production as relative values of land, labour and capital continue to
6
change: fertilizer replaces manure, tractors replace draft animals, concentrate feeds
7
replace crop residues, and cash crops replace food crops (Baltenweck et al., 2003).
8
Other factors can also play a significant role in determining the nature and
9
evolution of crop-livestock systems (McIntire et al., 1992). In humid areas with a high
10
disease challenge for large ruminants, crop-livestock interactions are likely to be limited
11
owing to lower livestock densities. Other factors include economic opportunities, cultural
12
preferences, climatic variability (e.g., droughts that lead to livestock losses), lack of
13
capital to purchase animals, and labour bottlenecks at some periods of the year that may
14
prevent farmers from adopting technologies such as draft power (Powell and Williams,
15
1993). Nevertheless, common patterns of both the drivers and the outcomes of
16
intensification of tropical crop-livestock systems can be identified. Choice of crops and
17
livestock interventions have been shown to be at least partly dependent on relative labour
18
and land costs and on market access, at a wide range of sites throughout the tropics
19
(Baltenweck et al., 2003). Furthermore, in the same study education level, market access
20
and human population densities were shown to be major drivers of crop-livestock
21
systems intensification (Baltenweck et al., 2003).
22
At the same time, alongside these larger scale drivers of farm development, the
23
ability of smallholders to implement new practices is further determined by intrinsic
24
system properties that may act as modifiers to their adoption. Farmers’ objectives and the
25
rules governing labour allocation and gender differentiation in the household are
26
examples (Thornton and Herrero, 2001; Waithaka et al., 2006). Such factors are not
27
necessarily related to spatial or macro-economic drivers; they therefore need to be studied
28
at the farm household level.
29
30
To understand the evolution of smallholder crop-livestock systems, we propose that.
31
5
these systems should be examined at multiple levels by analysing and linking macro-level
1
socio-economic drivers, regional-level land-use patterns, and micro-level household
2
dynamics and strategies. Complementary methods should be used that appropriately
3
reflect the key dynamics of each of these levels (Cash et al., 2006). The significant
4
complexity and uncertainty associated with the interacting biophysical and socio-
5
economic dimensions of agricultural systems should be taken into account by using a
6
multiple scenarios approach, informed by relevant stakeholder perspectives (Biggs et al.,
7
2007). Interactions of smallholder systems with changing contexts should be simulated
8
and discussed iteratively with key stakeholders to explore longer-term evolutionary
9
pathways (Kinzig, 2006).
10
11
In this paper we provide an example of this multi-level, multi-scenario, evolutionary
12
framework for the analysis of smallholder systems, using complementary modeling
13
approaches and harnessing relevant stakeholder perspectives. We build on the work of
14
Baltenweck et al. (2003) and Herrero et al. (2007a) by studying the potential household-
15
level impacts of crop-livestock intensification using crop-dairy systems data obtained
16
from longitudinal monitoring of representative case studies and key informants
17
(extension officers and policy makers) from Kenya. The objective of the study was to
18
generate socio-economic development scenarios as to how crop-livestock systems in the
19
highlands of Kenya might evolve in the next two decades and evaluate these plausible,
20
alternative futures through a multi-level modeling framework that includes a) the
21
development of scenarios providing different socio-economic conditions at the country
22
level and above; b) a regional land-use change analysis projecting the spatial distribution
23
of farming systems into the future; and c) the use of a household model to evaluate the
24
results of the spatial analysis at the farm level, allowing for a deeper understanding of
25
internal farm dynamics. We conclude with a discussion of the value of multi-scale,
26
stakeholder-generated, iterative analyses in evaluating synergies and trade-offs in farming
27
systems, particularly related to the dynamics of global change in tropical smallholder
28
systems.
29
30
