Working Papers

Year 2009 Paper 12

The Political Economy of Migration Processes:

An Agenda for Migration Research and
Analysis

Dr Sarah Collinson

IMI does not have a view as an Institute and does not aim to present one. The views
expressed in this document are only those of its independent author.

1
Abstract

This Working Paper proposes the development of an approach to researching and

analysing migration processes that combines a livelihoods approach to exploring local-
level migration dynamics with a relational political economy perspective that deepens
understanding of the broader social, economic and political processes and interests
interacting with migration at different levels (e.g. causing patterns of vulnerability or
creating opportunities that encourage migration). This approach aims to address a
number of important weaknesses in contemporary migration research, including a
failure to analyse migration as an integral part of social and economic transformation
processes (such as development and globalisation), inadequate attention to the precise
causes, circumstances and consequences of people’s movement, continuing reliance on
overly-simplistic migration categories and dichotomies, and poor appreciation or
understanding of the changing dynamics of migration processes over time (IMI, 2006).
Patterns and dynamics of migration are highly varied, complex and context-
specific. Understanding these dynamics in specific contexts might be improved by
exploring the interaction of local-level factors immediately influencing people’s
migration decisions and strategies (linked to livelihoods) with interacting political,
economic and social factors and processes at different levels which affect the agency,
vulnerability and/or opportunities of migrants (and associated non-migrants) during the
migration process, and that shape migration outcomes. Differential and shifting power
relations are crucial to understanding the interaction of migration strategies at the local
level with political and economic processes and relationships affecting these strategies,
since these are fundamental to understanding the interests and relative agency of
individuals, households and other actors involved. The concept of process is also
critical to the analysis, since the factors that affect people’s migration options,
strategies and outcomes are played out over time and can only be properly examined
within a historical perspective.

Keywords: Migration theory, migration dynamics, livelihoods, political economy,

power and agency.

Author: Dr Sarah Collinson is a Research Fellow and Programme Leader of the

Overseas Development Institute (Humanitarian Policy Group) and a Research
Associate of the Department of International Development, University of Oxford.
Email – [email protected]

2
1. Introduction

As outlined by the International Migration Institute (IMI) in setting out its current
research agenda, 1 the multi-layered, dynamic and complex nature of contemporary
migration processes present a number of significant challenges for migration research.
These include:
• The relationship between patterns and processes of international migration and
globalisation;
• The role of individual states and regions in migration processes;
• The nature of migratory behaviour at the micro-level and associated dynamics
of community transformation, trans-national identity formation, and livelihoods;
• Complex and politicised policy processes.

Migration pathways and motivations are highly varied and dynamic, and thus highly
resistant to generalisation. Simple categorisations and clear-cut dichotomies are
inadequate or misleading.

Shortcomings in contemporary understanding of migration highlighted by IMI include:

• A common failure to analyse migration as an integral part of development and
global transformation processes.
• Lack of attention to the process of migration itself – i.e. how people move and
their experience of travelling.
• Lack of attention to the causes and consequences of migration in places and
countries of origin.
• Continuing reliance on overly-simplistic categories and dichotomies to
characterise migrants.
• Lack of attention to the changing dynamics of migration processes over time.

Improved understanding of migration will depend on the development of more

comprehensive and fine-tuned theoretical and analytical approaches, suitable for
investigating the evolution of entire migration processes across space and time, and
suitable for developing a deeper understanding of the interaction of migration dynamics

1
www.imi.ox.ac.uk

3
with broader global transformation processes, and linking micro-level understanding of
migration to macro-level trends (IMI, 2006).

This paper proposes the development of an approach to investigating migration

processes that combines a livelihoods approach to analysing migration dynamics at the
local level with a relational political economy perspective. This political economy
perspective deepens understanding of the broader social, economic and political
processes and interests interacting with migration and livelihoods at different levels
(e.g. producing vulnerability or creating opportunities that encourage migration), and
focuses attention on structures and relationships of relative agency and power that
affect the dynamics of specific migration (and associated livelihood) processes. It is the
interaction of local-level factors immediately influencing people’s migration decisions
and strategies (linked to livelihoods) with a range of political, economic and social
factors and processes affecting the agency of migrants (and non-migrants) that
ultimately shape migration outcomes within specific contexts. Differential and shifting
power relations are therefore crucial to understanding the interaction of migration
strategies at the local level with political and economic processes and interests affecting
these strategies.

2. Background: approaches and challenges relevant to a

micro political economy approach in migration research

2.1. Migration theories

Economic theories of migration have reflected dominant models and theories of

economic development and under-development (de Haas, 2008; Waddington and
Sabates-Wheeler, 2003). Theories based on neo-classical economics treated migration
as a component of rational markets tending towards equilibrium, driven by the
(rational) economic decision-making of individual migrants and/or their households.
The Harris-Todaro model, for example, explains rural-to-urban migration in terms of
expected wage differentials between rural and urban areas (Harris and Todaro, 1970);
‘push-pull’ theories of international migration explain migration flows in terms of
supply and demand for labour between sending and receiving areas; the New

4
Economics of Labour Migration literature focuses on migration decision-making at the
micro level in terms of households seeking to maximise or protect household income
and/or reduce consumption (e.g. Stark and Bloom, 1985). By contrast, neo-Marxist
‘historical-structural’ approaches developed within political economy (e.g. Frank,
1966; Baran, 1973) have viewed migrants not as free and rational economic actors
operating within a benign and equalising market, but as workers whose movement is
determined by the exploitative structures and processes of capitalist development and
accumulation.

Critiques of economics-based theories have focused on the homogenising and

reductionist view that they encourage: of migrants and their households as atomised
economic decision-makers in more or less rational (labour) markets (over-emphasis on
migrants’ economic agency); or of migrants as unwitting pawns in globalised systems
of capital penetration, capitalist production and commercial extraction (over-emphasis
on the structural economic determinants of migration). Either way, in the search for
general theories to explain migration and its consequences, there has been a tendency to
abstract “economic migration” from the highly diverse and dynamic political, social,
cultural, institutional, historical and (indeed) economic contexts in which people move
(Gold, 2005). As a consequence, these approaches have failed to provide generally
applicable, predictive theories of migration that hold water in the real, complex world
of human mobility, or that provide an adequate explanation of who is moving where
and why. For example, push-pull theory ‘fails to explain why only some move, it
doesn’t explain who those people are, why they choose to move when others remain,
when they moved, to where and for which job’; micro neoclassical theories,
meanwhile, ‘are weakened by the assumption that decision-makers have perfect
knowledge about the costs and benefits of migration’ (McDowell and de Haan,
1997:9).

New approaches have challenged any attempt to understand migration on the basis of a
single level of analysis or discrete factor such as income differentials or labour demand.
They emphasise instead the inter-linkages between different migration streams; the
importance of agency, autonomy, perceptions, cultural and historical factors and
institutional constraints; the complex multi-level and trans-national nature of migration;
and the importance of social groups and relationships – including migration networks –

5
for shaping migration dynamics and migration experiences, straddling migration
‘sending’, ‘receiving’ and ‘transit’ locations, and a range of actors within them (Gold,
2005). As noted by de Haas (2008), this shift reflected a broader rejection within the
social sciences of ‘grand’ structuralist or functionalist theories, and a move towards
more pluralist or hybrid approaches that attempt to bridge the divide between structure
and agency in explanations of social processes.

Livelihoods studies of migration among poor populations have revealed the extent of
migration as a crucial, long-standing and highly varied and dynamic livelihood strategy
among many, if not most, rural communities in different parts of the world. Within this
literature, migration is not examined in isolation, but as one of a range of livelihood
strategies that might be available to households. While household agency and decision-
making is central to the analysis, institutions, structures and processes are also seen as
important in shaping livelihood opportunities, strategies and outcomes (cf., e.g., David,
1995; Deshingkar and Start, 2003; de Haan et al., 2000).

Migration network theories emphasise that migration is embedded in political, ethnic,

familial and communal relationships, including complex social networks and
relationships, that strengthen collective agency among migrants and their communities,
and that influence and mediate the dynamic interaction between structure and agency in
migration processes (Gold, 2005:4; Castles, 2008; de Haas, 2008).

Migration systems theories emphasise the dynamic social, cultural, economic and
institutional impacts of migration at both the ‘sending’ and ‘receiving’ ends, with
analyses seeking to incorporate both the causes and consequences of migration across
entire migration processes and involving geographically dispersed ‘transnational
communities’. Migration systems are conceptualised as linking people over space and
time in geographically dispersed (often trans-national) communities. Migrants exercise
agency within these systems, but migration systems have the effect of ‘structuring and
clustering’ migration geographically ‘by encouraging migration along certain pathways
and discouraging it along others’ (de Haas, 2008:21, citing Mabogunje, 1970:12).

All of this represents a significant advance on the earlier reductionist, abstracted and
deterministic economics-based models of migration. Migration network and systems

6
approaches have been particularly important for depicting the social and spatio-
temporal complexity and dynamism of migration flows, and livelihoods-based studies
have been very important for depicting the complex interaction between migration and
other livelihood strategies at the local level, particularly within poor rural communities
involved in seasonal, temporary and/or rural-urban migration.

However, it is arguable that in recent years, and as a consequence of the retreat from
flawed explanatory models, a great deal more energy and resources have gone into
describing and documenting contemporary migration and the varied institutions and
networks associated with it than has gone into analysing and explaining the deeper
causal and consequential dynamics of migration in different contexts and at different
levels.

Moreover, this descriptive endeavour has often remained relatively narrow in focus. In
the case of network studies, for example, there has been a tendency to research and
illustrate empirically the importance of migrant networks without investigating their
relative significance in relation to other factors affecting migration (de Haas, 2008: 20).
Although livelihood frameworks generally include ‘transforming structures’,
‘mediating processes’, ‘institutions’ and ‘organisations’, there has been a tendency for
migration livelihoods studies (like livelihoods studies more generally) to downplay
these structural features and to focus on households’ livelihood assets and activities (de
Haan and Zoomers, 2005:32-33).

What this large and growing volume of migration literature reveals, perhaps more than
anything, is the extreme diversity and complexity in patterns of migration, as viewed
both from places of origin and destination and within migration networks. As noted in
the introduction, this diversity underlines the need for analytical approaches that can
capture and cope with high degrees of variation at all levels. It adds further ammunition
to critiques of any attempts to posit general, universalising theories of what drives
migration and what its consequences are for the ‘receiving’ or ‘sending’ society /
community, or for migrants themselves. The dynamics of migration are not determined
simply by the ‘push’ of economic need and ‘pull’ of economic opportunity between
‘sending’ and ‘receiving’ locations or labour markets.

7
2.2. The rationale for a political economy approach to migration analysis

People’s livelihood strategies and ways of coping with economic, political, social or
environmental change – positive or negative (or both) – depend upon a broad range of
factors, including location, relative wealth, security regimes, kinship structures and
other informal institutions, the nature of local governance and social networks, and
access to land, food, roads, markets, water and other resources. Changes and
transformations at the macro and meso levels – such as those associated with structural
adjustment or development, political crisis, conflict and/or environmental or health
shocks and hazards – can transform local political economies, and cause communities
to constantly adapt to the opportunities and constraints that these transformations bring.
People adapt their behaviour and their livelihoods – including migration strategies – in
order to survive or minimise risk (survival or coping strategies), or to capitalise on the
opportunities to increase their economic position or welfare (accumulative strategies)
(Bhatia and Goodhand et al., 2003; Collinson, 2003:5; Deshingkar and Start, 2003: 2).

Thus, specific patterns of migration are determined to a large extent by the particular
interaction of individual or household livelihood strategies with a range of social,
economic and political relationships, processes, institutions and structures that make up
the (historical) context in which these strategies are pursued. Migration itself represents
a key process through which many communities are connected to the wider political
economy, and the nature of migration experiences and relationships for different
households and communities may be crucial for determining whether this connection
is, on balance, adverse or positive for their overall livelihoods and wellbeing at a
particular point in time

As suggested by Kothari, the complex interaction of individual agency and structures

within an historical context provides a useful overarching framework for understanding
migration processes, i.e. why people migrate and the consequences of this movement.
For example, how are households’ experiences shaped by micro conditions in particular
places and simultaneously by wider political and economic processes, such as land (re)-
distribution and government policies? (Kothari, 2002:10). A range of motivational
factors commonly reflect wider processes of economic, political and social change.
Thus:

8
 ‘decisions influencing ‘voluntary’ migration are not made in an economic,
political or social vacuum. Natural disasters, development initiatives, such as
the building of dams and roads, and conflict and war also displace people and
particularly affect those who are poor and tend to have minimal control over, or
access to, the political and economic capital necessary to affect the decisions
which impact on their lives and livelihoods. National and international
emigration and immigration policies further constrain or encourage people’s
decisions to move or stay. Decision-making does not only involve the migrants
but also many others with whom they are connected and thus has wider
implications and consequences than on the migrant alone’ (Kothari, 2002:9).

A particular migration stream is likely to include various migrants and/or groups of

migrants moving with varying degrees of agency in different circumstances, in
response to different causal dynamics, and often with very disparate outcomes in terms
of their own welfare and the wider implications of the movement. In Afghanistan,
Bhatia and Goodhand et al. report that even a single village commonly includes
households profiting from migration, other households coping through migration, and
some more vulnerable households surviving through migration but suffering a
consequential depletion of their asset base (Bhatia et al., 2003). In a study of migration
in rural West Bengal, Rafique, Massey and Rogaly found similarly that migration held
very different meanings for different households, with landless households most likely
to be engaging in migration to meet daily needs in the off season, whereas slightly
better-off households were able to undertake migration in order to save or invest
(Rafique, Massey and Rogaly, 2006:7). In Andhra Pradesh and Madhya Pradesh,
Deshingkar and Start report that, contrary to dominant perceptions of migration
patterns, people migrating from poorer areas can be on positive ‘accumulation’
migration pathways while some migrating from better-endowed areas might be on
coping pathways (e.g. due to local over-supply of labour due to land polarisation and
mechanisation in wealthier areas); indeed, one person’s ‘coping’ strategy can be
another person’s ‘accumulation’ strategy (Deshingkar and Start, 2003:vi,7). In
discussing migration from Burkina Faso to other West African countries and to Europe,
Black et al. (2006) note that while intercontinental migration is more likely to lead to
accumulation by migrant households than continental migration, this is not necessarily
the case: unsuccessful international migration can have very negative economic
consequences for the migrants themselves and others in their households or community
who have invested in migration in the hope of receiving remittances (Black et al.,
2005).

9
Understanding how and why particular people or households opt for migration, or not,
or are forced to move, and the implications of their movement at different levels and
locations, depends on a fine-grained analysis of the political, economic and social
structures and change processes operating or located at various levels (micro, meso,
macro) that differentially affect, and result from, people’s livelihoods and the
distribution of power, vulnerability and opportunity within and between households,
communities and groups over time. It also depends on analysis of the structure and
dynamics of relative power and agency within specific migration or livelihood systems
involved.

In order to capture the dynamic and transformative nature of migration and associated
livelihood processes, it is essential to pay attention to both power structures and
relations and actors’ agency, and the interaction between them. Migrants’ or
households’ ability to exercise power to pursue particular migration or other strategies
can be understood in relation to the differential constraints and opportunities created by
the power that people and groups exert over each other. These are shaped and mediated
by a variety of social institutions, such as gender, class or ethnic identity. As argued by
David Mosse, tracing the connections of power from broad political systems to
individual subjectivities depends on considering the systemic processes, structural
relationships and agency of different actors involved – and hence on examining ‘the
relationship between structural and voluntaristic expressions of power’ (Mosse,
2007:8). The concept of process is also central: the relationships and interactions that
cause people to migrate (or not) are played out over time, and can only be properly
appreciated and analysed within a historical perspective.

2.3. Related theoretical and analytical challenges within poverty research

and livelihoods approaches

The analytical challenges facing migration studies mirror, in many respects, similar
challenges within mainstream poverty research, which, as John Harriss argues, ‘has
generally failed to address the dynamic, structural and relational factors that give rise to
poverty’ (Harriss, 2007:1). Thus, while there has been a huge amount of research into
profiles of poverty and into ‘poverty dynamics’, looking, for example, at the
implications of access to different capitals for individuals’, households’ or

10
communities’ movements in and out of poverty, there has been comparatively little
research into ‘how and why it is that the distribution of factors measured and analysed
in these profiles and dynamics of poverty are distributed in the way that they are
through society’ (ibid.:1).

Some international political economy analysis, such as world systems theory, is very
much concerned with processes of capitalist exploitation, inequalities in international
economic power relations, etc., yet has not sought to link directly down to the local
level. Moreover, by restricting explanations of poverty to the structure and process of
the economic system, these models fail to take account of the social mechanisms that
perpetuate inequality and support relations of exploitation (Mosse, 2007:18). These
approaches therefore provide limited insight into how the macro dynamics of the
international capitalist system determine or affect, in complex ways, the real-life
experiences of poor individuals, households and communities at the local level. In the
Indian context, for example, poor people’s immediate experience is likely to be of
power exercised through informal systems of caste dominance, patronage or brokerage
(ibid.:21).

The dominant liberal economic view of international development, meanwhile, views

poverty as resulting primarily from poor people’s marginalisation or exclusion from the
benefits of the mainstream economy. Consequently, ‘poverty is conceived in terms of
marginality and exception rather than as a consequence of normal economic and
political relations’ (ibid.:2), or as ‘a kind of social aberration rather than an aspect of
the ways in which modern state and a market society function’ (Harriss, 2007:6). It
follows from this approach that strengthening poor communities’ incorporation into the
wider economy should help to overcome their poverty (through better access to roads,
markets, micro-credit and other support for local enterprise, etc.). This, indeed, is the
implicit logic that has underpinned much international development programming in
recent decades. Andries du Toit contends, however, that ‘often the problem is not that
poor people have simply been excluded from particular institutions, resources or larger
processes, but that they have been included on inequitable or invidious terms’ (du Toit,
2005:16).

11
Partly as a reaction against Structural Adjustment Programmes and their emphasis on
macroeconomic ‘solutions’ to underdevelopment and poverty, more ‘people-centred’
approaches to combating poverty emerged during the 1980s and 90s which took issue
with unitary macroeconomic approaches and which stressed, by contrast, the diversity
of poverty situations and the multidimensionality of poverty as a problem (Kaag et al.,
2003:3). The focus shifted onto the characteristics, welfare and capabilities of the poor
themselves – on features of poor households and their livelihoods, correlations of
factors associated with their poverty, and the significance of crises and shocks to their
welfare (ibid.:9). Less attention was given to the deeper structures and relationships
that give rise to this poverty. Consequently, Harriss argues, ‘causes and effects are
muddled up, and the characteristics of individuals or households that are associated /
correlated with poverty [e.g. landlessness] are represented as causal’ (Harriss, 2007:5).

Livelihoods approaches, for example, concerned very much with the agency and
capabilities of poor people, have focused attention on the ‘micro world’ of ‘lived
experience’ – of families, networks and communities – and on households and other
local actors (de Haan, 2005:9). Livelihoods frameworks (such as the ‘Sustainable Rural
Livelihoods Framework developed by the UK’s Department for International
Development) encourage analysis that is highly dynamic in many respects. This
framework can reveal, for instance, the conversion of different livelihood ‘assets’ (such
as land, waged income, education, social networks, etc.) into different livelihood
outcomes, the inter-relationships between different assets and the multiple benefits
generated by particular assets, or the transition of households in and out of poverty.
However, with the emphasis on these systemic aspects of livelihoods, and in the
absence of an explicit theory to analyse broader structural and transformational
elements within the frameworks, a liberal presumption of benign markets and other
positive institutions has tended to prevail, with consequently little attention paid to the
relationships, structures and processes creating, reinforcing and/or resulting from
poverty, inequality and vulnerability (de Haan and Zoomers, 2005:32-33; du Toit,
2005: 23; Collinson, 2003:13).

Maia Green criticises the depoliticised nature of most contemporary poverty studies,
which, she argues:

12
 ‘tell us that people are hungry because of lack of access to food or that infant
mortality is high because of poor health services …. [but] do not … tell us why
food cannot be accessed or why health services are inadequate …. The
emphasis on poverty as the problem and the locus of analysis diverts attention
from the social relations, local, national and international, which produce
poverty as an attribute of people. Very often it is not among the poor that we
should be looking for those relations which have contributed most to the
poverty of others …. The poor are poor not because of ‘poverty’, but are poor
because of other people’ (Green, 2005:38).

A better understanding of poverty, according to Mosse, would result from a ‘relational’

approach to analysing the problem, which views persistent poverty as the result of
historically developed economic and political relations, and which emphasises the
importance of social processes and relations of power (Mosse, 2007:1). Leo de Haan
argues along similar lines, stressing that:
‘life is a power play, livelihood is political and poverty is not only the result of
bad luck, ignorance or hazard but primarily the result of exclusion by others. …
‘Structures of domination’ and ‘social relations and flows of surplus’ shape the
way livelihood opportunities are perceived … and can be successfully accessed
or not’ (de Haan, 2005:8).

As discussed in the following section, migration analysis could gain a great deal from
poverty and development-related research that is focused more explicitly on the
analysis of the structures, social, political and economic processes and relationships and
activities that give rise to and perpetuate poverty, vulnerability (and wealth, security
and opportunity) and inequality. It would depend on close attention to actors’ agency
and to changing social, political and economic relationships between actors, as well as
to structural factors (economic and political) that can be demonstrated to produce
inequality and poverty for some, and economic and other benefits for others.

3. Investigating the micro and macro political economy of

migration: possible analytical approaches

As noted in the introduction, the purpose of this paper is to propose an approach to

investigating and understanding complex, changing and multi-level migration dynamics
in particular migration contexts that combines a livelihoods approach to analysing
migration dynamics at the local level with a political economy perspective concerned
with the broader economic and political processes and interests interacting with

13
migration and livelihoods at different levels, and with relationships of relative agency
and power within specific migration (and associated livelihood) processes. The aim is
to begin developing an approach to analysing migration which is:

• Contextually-specific.
• Explanatory as well as descriptive in orientation.
• Dynamic – both in terms of time (diachronic) and in terms of interaction
between actors, institutions and structures (synchronic).
• Historically grounded.
• Actor-oriented.
• Concerned with agency as well as structure within social, economic and
political processes.
• Explicitly concerned with linking between micro, meso and macro levels.

3.1. A ‘political economy of livelihoods’ approach

By focusing attention on individuals and households and their actions and strategies at
the local level, livelihoods approaches are extremely valuable for capturing the agency
of (poor) people and for exploring the ways in which this agency plays out in livelihood
processes as people use their assets to pursue diverse and complex livelihood strategies
(including migration) to achieve a variety of livelihood outcomes. There is already a
significant literature on migration and livelihoods, reflecting a growing recognition of
the scale and significance of both internal and international migration for people’s
livelihoods, particularly among poor communities (Hammond et al., 2005:10). Within
this literature, migration is viewed predominantly as contributing positively to the
achievement of secure livelihoods, and creating opportunities for poor people to escape
poverty (Ellis, 2003).

However, as noted above, the emphasis within livelihoods research on agency at the
household level has come at the price of much explicit engagement with issues of
structure and process, or with power and wealth relationships at the micro and macro
levels that interact with and impact upon people’s (individual and/or household)
agency. Without this engagement, livelihoods studies have been weak in capturing
(historical) processes of social and economic change or transformation – such as

14
accumulation, domination, exclusion, marginalisation, impoverishment or
disempowerment – that play such a central part in determining or shaping the context
and dynamics of people’s livelihoods, including their migration strategies.

Even at the local level, livelihood activities can engender processes of inclusion and
exclusion in arenas of conflicting or co-operating actors (de Haan and Zoomers,
2005:34). Particularly in poor communities, economic vulnerability contributes to, and
is exacerbated by ‘stressed, power-laden and conflictual’ social networks (du Toit,
2005:12). Livelihoods studies have tended not to explore how the agency of certain
actors affects, directly or indirectly, the agency of others, or how particular ‘assets’ or
‘capitals’ can have different meanings or significance for different members of a
household or community. There is a need to consider potential, sometimes negative,
links or implications for certain groups, households or individuals resulting from
others’ migration or other livelihood strategies.

For example, in situations of protracted conflict, crisis and displacement, such as Sri
Lanka, Afghanistan and Somalia, geographical dispersal or fragmentation of
households and associated migration often represents a coping or survival strategy for
the household as a whole, but can sometimes result in heightened vulnerability of
certain household members. For example, elderly, sick or other non-migrant members
may become more economically and/or socially vulnerable when depletion of
household assets (e.g. loss of sons’ or daughters’ labour or care) is not compensated by
reliable inflows of financial remittances. Those who have left may be exposed to
violence, exploitation or other hardships (e.g. separation from children) during the
process of migration or at the point of destination. In Darfur, following the closure of
the national border between Sudan and Libya, many groups of prospective young
labour migrants were absorbed instead into tribal militia and warring factions;
according to one young Darfurian in Benghazi, ‘In Darfur we had three options: join
the rebels, go to the camps or get out’ (Young et al., 2005:87). The consequence of
more young men joining militias as a livelihood strategy is that other people’s lives and
livelihoods are further threatened or destroyed, which in turn increases the risk of
distress migration or forced displacement. In South Africa, Du Toit observes that the
‘social capital’ embodied in ‘care chains’ among poor communities in the Western
Cape play an important role in sustaining many households and create benefits for

15
some household members, but the social roles and (sometimes violent) sanctions that
enforce them are also part of what keeps many women poor (du Toit, 2005:13).

Understanding the competitive and sometimes destructive or impoverishing

relationships between the livelihoods of different individuals, households and groups,
and how these play out over time is essential for understanding how livelihood assets
and strategies reflect and translate into power and wealth relations and, potentially, the
systemic marginalisation and exclusion of particular people or groups. Livelihoods are
not always ‘sustainable’ for all those affected, and one actors’ agency is often the cause
of another’s subjugation. (Collinson, 2003:13). Livelihoods frameworks therefore need
to be adapted and strengthened by integrating additional factors and questions that can
provide a more systematic basis for relating the micro-dynamics of livelihood activities
and outcomes (individual and household agency) with wider relationships, processes,
structures and institutions that constitute the social, economic, political, and historical,
context in which these livelihoods are pursued (cf. Figure 1). These wider factors
interact with individuals’ and households’ agency at the local level (Collinson,
2003:13-14; Lautze and Raven-Roberts, 2006; de Haan and Zoomers, 2005),
constraining and/or expanding it in different ways. Understanding this interaction
requires critical analysis and explanation of the nature of the key relationships,
processes, structures and institutions that are significant in a particular livelihoods
context and that affect particular livelihoods processes.

Within the migration and livelihoods literature, Kothari (2002), and Deshingkar and
Start (2003) propose ‘social exclusion and livelihoods’ approach ‘to show why some
groups of people have succeeded in entering accumulative migration pathways while
others have been excluded’ (Deshingkar and Start, 2003:vi). Uma Kothari suggests
that:
‘various forms and processes of exclusion produce different groups amongst the
excluded. These groups are differentially compelled or excluded from adopting
migration as a livelihood strategy. … [T]he presence or absence of different
forms of capital are both the cause and consequence of processes of exclusion
and discrimination which limit or enable migration. It is the particular package
of vulnerabilities which shape the extent to which people can or cannot move.
However, it is also clear that a lack of capitals can both require and limit
movements and that by acquiring capital, an individual can be in a position to
stay put profitably. … The chronically poor are often those who stay put or are
left behind in an environment where others are migrating.’

16
In the Indian context, Deshingkar and Start observe that caste is an important
determinant of who is excluded from positive migration streams: the disadvantage and
social exclusion suffered by certain (lower caste) groups results in migration remaining
a low-return coping activity rather than a high-return activity, as enjoyed by many
higher-caste groups (Deshingkar and Start, 2002:5). By drawing attention to the
differential assets, opportunities and potentially conflicting interests and diverse
outcomes of different groups, this social exclusion and livelihoods approach goes part
of the way towards addressing the key limitations of mainstream livelihoods analysis
highlighted above.

However, ‘social exclusion’ is itself an imprecise and potentially oversimplifying

concept, and could distract attention away from other important power and welfare
dynamics affecting migration and other livelihood processes. Crucially, what the
‘social exclusion’ approach does not adequately consider is what deeper processes or
structures account for the social exclusion that it is concerned with. Nor does it provide
a basis for examining how certain forms of (adverse) inclusion in particular (possibly
exploitative) migration or other livelihood processes or particular institutions might be
a primary cause of negative (e.g. impoverishing, hazardous or marginalising) livelihood
strategies or outcomes (including involuntary migration) (du Toit, 2005; Mosse, 2007;
Green, 2005). The concept also implicitly encourages the view that inclusion in
migration streams and associated processes and institutions is likely to be positive for
the welfare of migrants and their households. There is a danger here of jumping to
unwarranted conclusions, both about the significance or potential of migration in
overcoming exclusion and disadvantage or reducing poverty among the migrants and
households involved, and about the underlying causes of migration in different
contexts.

Building directly on the ‘relational’ approach to chronic poverty research advocated by

David Mosse (Mosse, 2007) and Andries du Toit and colleagues (du Toit, 2005; du
Toit et al., 2005), a more powerful approach to analysing migration dynamics could be
derived from combining livelihoods research and analysis with a ‘relational’ political
economy approach that focuses attention directly on the dynamic relationships,
processes and structures of relative power between different actors involved in
migration or other livelihood processes. This would aim to trace and explain the

17
complex interaction of exclusionary and inclusionary processes and relationships
between the many different actors, institutions and structures involved over time and at
different levels – micro, meso and macro. This approach is not concerned so much with
whether people are included or excluded in particular migration streams or other
livelihood processes, but how they are included or excluded, and with what the
implications of these terms of inclusion or exclusion are for their and others’ welfare,
for the dynamics of migration processes, and for wider social, economic and political
structures and change processes.

Boxes 1 and 2 outline how this type of approach has illuminated understanding of the
dynamics of migration, livelihoods and the local and wider political economy in the
case of adivasi cultivators and seasonal labour migrants in India (Mosse, 2007) and
poor rural communities in South Africa’s Eastern Cape (du Toit et al., 2005). In the
case of adivasi communities, labour migration often allows survival at the margins for
the poorest, and has become the only means by which many agrarian livelihoods can be
maintained (Mosse, 2007:17). Yet the terms of involvement in migration processes are
far more advantageous for some than for others, and this is explained in part by pre-
existing structural inequalities that migration processes often play a part in supporting
and reproducing. Vulnerable migrants and successful recruiters and gang leaders,
Mosse notes, ‘are part of a chain of self-interest that in aggregate gives stability to a
highly exploitative system generating mass chronic poverty’. It is not migration that
causes chronic poverty, but rather the social relationships of exploitation involved in
migration processes (ibid:29). Personal histories of suffering, survival, social mobility
and migration within rural communities of the Eastern Cape in South Africa show how
the poor households are adversely incorporated into the broader macro political
economy (including through migration) and within a set of highly localised and
unequal socio-economic relationships that provide both marginal opportunities for
survival for the poorest and opportunities for better-off households to expand their
economic base. Power relations are rooted in access to and claims over key resources
(land, patronage, labour), and these, in turn, lead to particular households and groups
being marginalised, disempowered and trapped in long term poverty (du Toit et al.,
2005).

18
BOX 1

An analysis of the local political economy, livelihoods and personal histories by du Toit, Skuse and
Cousins (2005) in the Mount Frere area of South Africa’s Eastern Cape region encompasses
individual migration strategies, depicting these and other aspects of people’s livelihoods as entirely
embedded in the dynamics of relative advantage, disadvantage, marginalisation, coping and power
within the poor communities concerned.

Although remote from the centres of South Africa’s industrial and commercial economy, the
deprivation and poverty of the area is explained to a large extent by a history of systematic
underdevelopment of the black rural economy. This has left poor rural land-based livelihoods highly
fragile and vulnerable and failed to provide adequate alternative forms of employment and
livelihoods. In this context, migration to Cape Town and other distant urban locations represents a
key link with the urban political economy, with many poor households heavily dependent on migrant
members’ remittances to cope and survive. Widespread job losses caused by declines in key
sectors of the urban economy have negatively affected migrants from areas such as Mount Frere
and their ability to remit to their relatives. The local political economy is also linked into the wider
South African and global economies through the expansion of corporate capital into the area, such
as through the Spar Supermarket which markets cheap bulk foodstuffs produced by the white
commercial farming sector. The availability of cheap food in the supermarkets supports the survival
of many poor households, but also undermines the viability of local agricultural livelihoods and,
through the profits extracted by the supermarkets, represents a key process through which meagre
economic resources of the local economy are transferred out to South Africa’s economic centre.

Relationships of relative power and poverty among members of these poor communities mean that
that there is a local political and economic ‘elite’ that often guards and controls access to local
resources, including the few local jobs that exist in the public or service sectors. The poorest
households are adversely incorporated not only within the broader political economy but also within
unequal socio-economic relationships at the local level. The local political economy simultaneously
provides survival opportunities for the poorest whilst also providing opportunities for better-off
households to maintain or improve their relative economic advantage. The significance of migration
and other livelihood strategies for different households is determined to a great extent by the local
processes and dynamics that benefit and empower some but trap others in chronic poverty.

Individual life histories and livelihood trajectories depicted in the study reveal these varied
experiences and the complex social and economic dynamics and processes behind them. Du Toit et
al. contrast the experiences of two households in the community, one of which has been able to
enter into the local economic and political elite through economic migration and capital
accumulation, while the migration and other livelihood strategies of the other (more typical) poor
household have failed to provide a route out of chronic poverty and vulnerability. They observe that
life histories of this kind highlight the importance of an ethnographic understanding of agency and
changing power relations, which is reflected in the social mobility and class repositioning
experienced by some, and the immobility and persistence of poverty experienced by others.

Elsie, the female head of the first household, had grown up in a very poor family. She had been
cared for by her grandmother in her natal village after her father died and her mother had had to
migrate to Port Elizabeth to find employment. When her grandmother died, Elsie’s mother returned
to the village to care for Elsie, where she managed to find a domestic job with a local shopkeeper.
Although the family’s continuing poverty led to Elsie’s mother migrating once again (to
Johannesburg), financial help from a local chief enabled Elsie to attend school and this
subsequently enabled Elsie to undertake teacher-training and secure a relatively secure and
lucrative teaching job in a nearby village. Elsie’s story illustrates how intergenerational social and
economic mobility can be possible for those able to access stable employment.

By contrast, the (more typical) experiences of a much poorer household reflects various processes

19
of impoverishment, exclusion and disempowering incorporation into the local and wider political
economy and society, including failed migration strategies. This life history focuses on Patricia, the
wife of a former migrant labourer. For twenty years, her husband had been working in Johannesburg
and supported the household through regular remittances. He lost his job in the early 1990s as a
result of the contraction of mining and other key economic sectors in South Africa, and he now
works locally looking after livestock. Patricia has been unable to migrate to seek urban employment
as a domestic worker because of she herself lacks help with child and other care needs within the
household. With the household unable to rely on self-grown food, their survival depends on a
complex web of local social, political and economic relationships and survivalist modes of reciprocity
such as working for food and seeking gifts from neighbours. The household’s chronic poverty and
vulnerability is reinforced by their inability to pay school fees, which has left Patricia’s children with
very few options, given the lack of employment opportunities within the village or beyond. Patricia’s
daughter disappeared after migrating to Cape Town three years ago and Patricia fears that she is
dead, representing a failure to effect a successful migration and improve a rural livelihood.

(du Toit, Skuse and Cousins, 2005)

There is no single all-encompassing approach to exploring and analysing the micro-

and macro political economy of a particular context. It is necessary to draw flexibly on
a number of frames of reference to help build up comprehensive picture of the political
economy over time. International political economy provides theoretical and
conceptual frameworks for analysing global or macro level economic processes, power
relations, processes of exploitation and accumulation, and structures of inequality.
Gender studies provide a rich source of theoretical work on social power relations
which could substantially strengthen understanding of the significance of power
relations in livelihood processes. This includes processes of empowerment and
disempowerment, and of ‘wielding and yielding’ negotiation processes between
different actors (de Haan, 2005; de Haan and Zoomers, 2005). Maia Green notes that
anthropology has long perceived poverty as a consequence of relations between people,
i.e. as a social relation, and that ethnographic studies can provide numerous insights
into the local level dynamics of poverty and inequality and the significance or
implications of caste, social exclusion, structural transformations, social constructions
of property relations, and trans-national or global processes affecting these relations
(Green, 2005:18). Andries du Toit argues similarly that anthropology and critical
sociology provide the tools for investigating complex issues connected with the
interactions between poverty, power relations, vulnerability and agency – i.e. for
investigating the local political economy of poverty. Collinson’s adaptation of a
sustainable livelihoods framework for application in conflict situations highlights
dynamic interactions between micro-, meso- and macro-level political economy
dynamics (see Figure 1) (Collinson, 2003; see also Lautze & Raven-Roberts, 2006).

20
 Figure 1: Adapted Livelihoods Framework for Situations of Conflict and Political Instability (Collinson, ed., 2003)

VULNERABILITY /CONTEXT
environmental/political/economic/climatic/military shocks and trends

affects

affects
affects

affects

affects
TRANSFORMING
STRUCTURES AND
PROCESSES
LIVELIHOOD
LIVELIHOOD STRATEGIES OF
• infrastructure PARTICULAR SOCIAL
ASSETS OF A • state/government institution
PARTICULAR ACTORS
• kinship networks
HOUSEHOLD / • markets influencing
GROUP/ • civic institutions • agriculture
COMMUNITY/ • traditional authority • labour
POPULATION • private sector • trade
• ethnic institutions • migration
• religious institutions • smuggling
• predation and asset
• laws stripping
RELATIVE •
POWER/WEALTH/ affects
policies • external aid
H • culture
VULNERABILITY/ engagement
S N with • ethnic & religious identity
affects POVERTY OF • conflict and violence
PARTICULAR Determining/
• war economy achieving
HOUSEHOLD / • displacement
and access to
GROUP/ • environmental degradation
P F COMMUNITY/ • asset transfer LIVELIHOOD OUTCOMES FOR
Pol POPULATION • aid inputs PARTICULAR SOCIAL ACTORS
• foreign investment • income
F= financial • militarisation • food security
and impacts
assets • foreign intervention
H= human assets of/significance of • health and education
N= natural assets
determines
• trading
P= physical • economic vulnerability
assets • political vulnerability
S= social assets
Pol= political • vulnerability to violence
assets • use of Natural Resources base21
The type of ‘fine-grained, critical sociology’ advocated by du Toit depends on
crossing intellectual and disciplinary boundaries between sociology, political
economy, social history, political geography and anthropology (du Toit, 2005:4 &
11). The challenge, he argues, ‘is not only to develop modes of analysis that link
quantitative and qualitative, or to bridge the sterile opposition between macro-scale
‘structural’ analyses and nuanced exploration of agency on the ‘micro-scale’, but also
to create space for debates that allow a qualitative sociology informed by critical
theory and political economy, and an awareness of the dynamics of conflict,
inequality and social processes within otherwise depoliticised and technicist
discourses’ (ibid., 2005:26). Within the migration literature, the call for inter-
disciplinarity is echoed by Stephen Castles, who, in the context of migration /
development relationships, argues for a much broader inter-disciplinary analysis of
the development of social structures and relationships in the context of globalisation,
and emphasises the need to improve understanding of the relationships between
macro-, meso- and micro factors of change (Castles, 2008:9).

Whatever analytical lens is used, there are a number of significant methodological

challenges and constraints associated with researching complex social, political and
economic processes in any local or wider context. Particularly in poor and/or unstable
or contested political environments, research and analysis may be hindered by a lack
of reliable data and lack of good quality background research material to draw upon,
particularly at the micro- and meso levels. The sensitivity of exploring particular
social, political and economic issues will vary a great deal from place to place and
over time. There is never a single ‘right’ understanding of any complex situation, and
what information and data is available will vary a great deal in terms of scope,
reliability and detail. What is clear, however, is that qualitative research will need to
take priority, and that a variety of methods will need to be employed, ideally over a
period of time, in order to build up a detailed picture of the dynamics and significance
of migration for a particular community, group, or within a particular economic or
other sector or process. These might include personal histories, informal observation,
key informant and focus-group interviews, participatory resource mapping and
timeline activities, community-level surveys, and analysis of a wide variety of
background and ‘grey’ literature, particularly relating to the meso-, national and
international-level political economy and history.

22
BOX 2

David Mosse (2007) examines the local political economy of poor adivasi (tribal) cultivator and seasonal
labour migrant livelihoods in India’s border districts of Rajasthan, Madhya Pradesh and Gujurat. The
high level of chronic poverty in tribal forested (/ deforested) regions has its roots in the long-term
dispossession, exploitation and erosion of livelihoods among local adivasi groups by colonial forest
regimes. Although usually explained in terms of the immediate factors and constraints associated with
poverty, including land pressure and declining subsistence agriculture, remoteness from markets, poor
education, unemployment, indebtedness and out-migration, Mosse points to continuing relationships
and processes of dispossession, exploitation and inequality as the deeper causes of persistent poverty
today.

This is reflected in the livelihoods of very poor, indebted subsistence farmers in the Bhil tribal region,
where half of the adult population is typically absent for the half of the year, usually working as casual
and seasonal labourers in urban construction sites. The context of this labour migration is one of deep
structural inequality, casualised markets, collapsing rural livelihoods and absent formal labour welfare
and social safety nets, all of which are factors associated with broader (including global) processes of
development and economic and social change that are mediated by complex social, economic and
political relationships and institutions at the local level. Continuing development in India’s urban centres
has been facilitated by these large scale flows of easily exploitable labour. The larger established
builders and contractors, Mosse observes, tend to favour casual migrant labour recruited through
brokerage and debt-dependence over the more independent labour available through urban daily labour
markets.

The highly segmented nature of urban labour markets means that Bhil migrants are typically excluded
from skilled work and employed instead in the lowest-paid and least secure positions as construction
labourers. Their recruitment takes place through a multi-tier system of labour gang leaders and brokers
(mukkadams), recruiter-supervisors and labour contractors that reproduces this segmentation and
ensures that Bhil migrants follow well-defined and repeated routes from particular villages to particular
urban work sites. The poorest migrants are acutely vulnerable to unemployment and are therefore highly
dependent on these recruiters and gang leaders, who also arrange cash advances and shelter and
other protection at work sites. The livelihoods of mukkadams – themselves usually successful migrants -
depend, in turn, on their loyalty to employers and contractors.

The impoverishing aspects of the system are ill-perceived, Mosse argues, because they exist alongside
aspects of the same system that, because of different structural conditions (landholding, education,
influence, etc.) allows accumulation and upward mobility for some – such as from labourer to recruiter,
skilled worker or contractor. Many better-off adivasi farmers are able to use migrant incomes to improve
agriculture, and to invest in essential social networks. Moreover, for poorer farmers, migration often
provides a crucial survival or coping strategy – indeed, for many, it is the only means of maintaining
agrarian livelihoods. In this context, migrants are likely to prioritise securing work over protection from
exploitation. Yet their adverse incorporation into the labour market means that they have little prospect
of escaping from poverty and debt through labour migration. Most casual labour migrants experience
very poor working conditions, social isolation, high levels of insecurity at work sites, little upward
mobility, and chronic debt, including debt associated with cash advances from mukaddams. Those who
are most exploited and have least power to protect their interests are families for whom migration is a
defensive survival strategy – those trading their labour for cash to meet urgent food needs in the lean
season, often migrating the furthest, for longest, under the worst conditions of deprivation and insecurity,
with least reward, and who are most fully tied into relations of dependence and exploitation. In their
home villages, long absences and dependence on distant patrons tends to erode migrants’ local social
capital, through, for example, their marginalisation from social networks through which credit, marriages
or benefits from development projects are obtained.

(Mosse, 2007)

23
3.2. ‘Value chains’ and analysis of the political economy of migration
processes

A major challenge for migration studies is to link what is observed at the local
household and community livelihoods level in migrants’ origin and destination
locations to wider migration networks and processes.

One potentially useful way of linking micro- to macro factors and processes is to
investigate how a variety of capitals important for people’s local and/or migrant
livelihoods and well-being are transmitted or transferred in different directions
through specific migration relationships and related processes. Financial and other
remittance flows, of course, are of key importance and are already usually
incorporated into migration analysis at all levels. Other asset flows or resource
transfers may be just as important for households (migrants’ and non-migrants) and
communities, such as:
• Transfer of human assets out of local agricultural production.
• Transfer of human assets into households or community through improved
access to education.
• Transfer of social assets away from elderly and other vulnerable non-migrants
within sending communities.
• Transfer of political assets into a community through links with a powerful
and influential emigrant diaspora, or the possible inward transfer of education
and skills (human assets) associated with return migration.

Patterns and flows in these transfers are likely to be very different for different
households, communities and groups involved or affected.

It is important to look beyond asset flows associated directly with migrants and
migrant networks or systems to the transfer of migration-related assets in the broader
political economy. Money flowing into a particular community as remittances, for
example, might subsequently flow out if the local food and other commercial markets
are dominated by external firms. Transforming power relations among (migrant and
non-migrant) individuals, households and groups in the context of specific migration
experiences, processes or networks can be analysed in terms of changes and transfers

24
of political assets, and in terms of the dynamic relationships between political and
other assets.

Value chain analysis, a framework developed originally within world systems theory,
provides a useful approach to exploring the links and interaction between local
livelihoods and specific migration networks, systems or processes, and associated
asset flows and transfers (such as labour, land, money and ‘social’ capitals), and,
importantly, for investigating relative power relations within these (cf. Gereffi and
Korzeniewicz, 1994; Gereffi, 1995; Raikes et al., 2000). Value chain analysis
identifies power relations, governance structures and exchange relationships within
commercial networks, from primary production / local level, through to consumption /
international level. Of particular interest to political economy analysis is the
identification of who controls flows (e.g. of particular commodities) and exchanges at
particular levels. In a commercial context, a value chain may be predominantly
controlled and driven by producers or by commercial intermediaries, or by buyers at
the consumer end of the chain. Gereffi identifies four dimensions of global
‘commodity’ / value chains: the input-output structure of the chain, the territory it
covers, its governance structure and the institutional framework. The governance
structure includes notion of barriers to entry and coordination within the commodity
chain, and distinguishes producer-driven from buyer-driven (and intermediary-driven)
structures and processes. The institutional framework determines how key or ‘lead’
actors within the chain involve less powerful or subordinate actors through their
control of market access or information, for example by monopolising export and
marketing networks (Gereffi, 1995).

With some adaptation, all of these concepts are potentially applicable to the analysis
of migration processes – particularly for analysing how migration strategies at the
household or local level are linked into and structured by broader migration networks,
systems or processes, and for identifying important power relations, governance
structures and institutions within them:

• Like producer-driven commercial commodity chains, migrants themselves

may exercise considerable agency and control within some migration

25
 processes (e.g. international migration of rich entrepreneurs or certain highly
skilled migrants).
• In others, individual migrants have little or no control and remain highly
vulnerable to the exploitative interests of other actors involved (e.g. ‘survival’
or ‘coping’ migration of poor or destitute landless labourers, and various
forms of ‘involuntary’ or ‘forced’ migration, such as that of child soldiers,
trafficked sex workers, development-induced migration, flight from violence
or persecution, slum clearance).
• In some migration processes, considerable power and control may be
exercised by intermediaries – such as labour recruiters, gang leaders, brokers,
powerful actors within trans-national migration networks, smugglers &
traffickers – who influence very directly how and where and the circumstances
of their migration.
• Some migration processes are clearly structured by a powerful institutional
framework (e.g. strict and effective immigration controls, intra-community or
intra-ethnic affiliations and networks, formal or informal recruitment
networks); others may appear more fragmented (e.g. ‘mixed migration’ flows
in West and North Africa).
• Some migration processes are controlled to a large extent by economic or
other actors at the ‘receiving’ end (e.g. targeted recruitment of health
personnel by public health services), others more by actors or constraints at the
‘sending’ end (e.g. Eastern Bloc emigration controls during the Cold War).

State actors, such as immigration control authorities, may be considered as particular

actors among many others in the chain, exerting considerable control in some
situations and less in others. Not all institutions of the same government – such as
home affairs, foreign relations, development, trade and industry and health agencies
or ministries, and local authorities – will necessarily be operating with the same set of
interests or the same degree of control in any particular context or point in time.

As with certain commercial networks, many people will face barriers to entry or will
be entirely excluded from particular migration processes, for example, because entry
into the migration stream is highly segmented according to structural income, skill or

26
other differentials, and/or selective or restricted according to gender, age, ethnic
group, nationality, skills, financial resources, etc. Because migration strategies at
household and community level are often a matter for those who stay as much as for
those who leave, non-migrants within communities affected by migration must be
considered as key actors within migration streams, and thus included in the chain
analysis.

4. Pointers for research and analysis arising from a political

economy approach to migration processes

There are a number of potentially important and theoretically challenging insights that
are generated by adopting a relational political economy perspective on migration
processes.

1) Detailed qualitative research with a historical perspective, focused on specific

households and local communities and discrete migration networks, systems or
processes is likely to reveal more about the real (highly complex and varied) causes
and consequences of migration than (present-focused) quantitative surveys and
analyses.

2) Rather than viewing migrants (or non-migrants) simply as participating (or not) in
particular migration streams and/or labour markets or other commercial or social
networks, there is a need to consider the qualitative aspects of their incorporation in
the local, national and international political economy and in specific migration-
related markets and networks, and to explore this incorporation as a process with a
dynamic time dimension:
• What are the patterns and dynamics of poverty, vulnerability, wealth and
power within and across communities involved in migration, and what are the
underlying causes of these (e.g. marginalisation from national development
assistance, penetration of international commercial interests, discrimination,
environmental factors, patterns of land ownership and/or competition for land
and other resources, presence or absence of aid agencies, remittance flows,
etc.)? How does this affect the relative agency and vulnerability of different

27
 migrants (c.f. migration as positive choice / necessity for coping or survival /
coerced, etc.)?
• How are migrants (and associated non-migrants) incorporated into migration
streams or networks, labour markets, etc., and on what terms (exploitative or
not / contributing positively to social capital or not, etc.)? How has this
changed over time and in relation to other groups or factors?
• Who are the winners and losers in this incorporation? What power dynamics
are involved?
• How are different communities’ welfare and livelihoods and other actors in
society (e.g. public institutions, corporations, etc.) affected in particular
‘sending’ and ‘destination’ locations by migrants’ involvement in particular
labour markets or commercial or other networks?
• How do migrants’ (/households’ / community’s) involvement or incorporation
in particular migration-related (labour/) markets and other networks reflect
and/or interact with their broader relationships and interaction with the local,
national and international political economy?

3) “Positive” or “negative” impacts of migration cannot be generalised in any

meaningful way. As in all complex social processes, there is likely to be a dynamic
distribution of different advantages and disadvantages accruing directly and indirectly
to different actors in the communities and societies involved; these are likely to
change over time, and will often be qualitative and subjective in nature and highly
resistant to ‘measurement’. In the case of the very poor adivasi migrants in India
studied by Mosse, for example, migration has enabled vulnerable households to cope
or survive in a context of extremely limited and/or dwindling livelihood options, and
yet the dynamics of these migration processes play an important part in perpetuating
or even deepening the poverty and vulnerability of many of those involved.
Immigration of unskilled workers into rich countries may directly benefit particular
employers but impose strains on (particular parts of) local public services at certain
points in time or certain places.

As regards the relationship between migration and development, dominant opinion

has shifted in recent years from a previously pessimistic stance to a more positive

28
view of the potential benefits of migration for development (Castles, 2008; DAC,
2006:53). A relational political economy approach would challenge both “negative”
and “positive” camps, and encourage instead more nuanced appraisals reserved to
particular migration processes or contexts. It would almost certainly reveal a very
mixed, complex and changing picture of the benefits and costs of migration for
different (groups of) migrants, non-migrants and other actors, structures and
institutions involved or affected. It could therefore sensitise analysts and policy-
makers to the potentially complex or counter-intuitive impacts or implications of
particular policy measures (such as the indirect negative impacts that assisting
‘accumulating’ migrants could have on poorer migrant and non-migrant households or
groups). It should direct analysts (and policy-makers) to ask more probing questions
about precisely how certain migration processes might benefit whom in a particular
migration and development context.

For example, much temporary, seasonal or circular ‘coping’ or ‘survival’ migration

among the poorest of the rural poor is indicative of fragile or unsustainable rural
livelihoods. By providing a basis for many poor households to survive or cope,
temporary, seasonal and ‘circular’ migration may prevent or reduce more permanent
forms of ‘rural-urban’ migration. It is not necessarily a route out of poverty, however,
due to often exploitative labour and other economic and social relations involved and
strained or limited social networks and capital that these migrants are able to draw on
(in context of surplus rural labour, casualised urban labour markets, etc.). The
principal impact on ‘development’ of this migration could be in providing a cheap
source of exploitable labour for national & global capital based predominantly in
urban centres. Other poor communities may benefit from the associated economic
growth, making this migration-supported development ‘pro-poor’ for some, but not
for many migrants involved, who provide unskilled labour for low wages in poor
conditions and are often tied into debt and other relationships of dependency from
which they can’t escape. These migrants have limited agency: migration for many is a
necessity, rather than a choice, and often an impoverishing experience. Yet although it
may play a role in reinforcing their relative poverty, migration may nevertheless help
to prevent or delay the complete collapse of marginal rural livelihoods and reduce
vulnerability to shocks.

29
4) In analysing the impacts and implications of migration involving a particular group
or locality(s), it is likely to be instructive to explore processes of asset transfer
between different actors involved, including migrants, their households, actors within
wider migration networks, employers, etc. For example:

• Human assets – e.g. labour transferred from household to employer, or

extracted through trafficking.
• Natural assets – e.g. land or other natural resources transferred to migrants in
destination locations.
• Financial assets – e.g. remittances; debts paid through migrants’ wages to
smugglers or traffickers.
• Political assets – e.g. linked with migrant diaspora’s engagement in national
politics.
• Physical assets – e.g. remittance of construction materials, cars, consumer
goods, etc.
• Social assets – e.g. care relationships and care chains disrupted by migration;
migrant networks enabling access to employment.

Improved understanding of the significance of particular asset flows and transfers and
the interactions and relationships between them should support improved
understanding of the relative significance of migration and associated ‘capital’ flows
for reducing (or increasing) migrants’ and households’ poverty, vulnerability or
dependency / power or wealth. In the analysis of remittances, for example, analysis
could consider whether money from migration tends to be recycled within the local
‘sending’ community (within / beyond households), or whether it is extracted out
again through the penetration of local food and other markets by national or global
companies / capital.

5) Different migration ‘types’ or categories could be distinguished at a general level

according to dynamic and relational political economy concepts that capture the
relative or constrained agency of different migrants and the qualitative nature of their
involvement in migration processes. Possible categories could include, for example:

30
 • ‘Survival’ / ‘coping’ migrants from chronically poor households or
communities experiencing asset depletion / precarious asset maintenance and
adversely incorporated into labour and other markets in which they have little
agency.
• Unskilled / skilled migrants from poor rural or urban households or
communities accumulating at low or modest levels and participating in labour
and other markets on broadly disadvantaged terms but with some degree of
agency and bargaining power.
• Moderately (or highly) skilled / moderately (or highly) endowed internal /
international migrants experiencing asset depletion and adversely incorporated
into labour and other markets in which they have little agency.
• Moderately (or highly) skilled / moderately (or highly) endowed internal /
international migrants from rural / urban households or communities
accumulating at low or modest levels and participating in labour and other
markets on disadvantaged / advantaged terms with some degree of agency and
bargaining power.
• Highly skilled / highly endowed internal / international migrants accumulating
at high rates and incorporated into labour and other markets on highly
advantageous terms with a high degree of agency.

Categories of this kind could be combined with more conventional migrant categories,
such as seasonal / temporary / circular / rural-urban / internal / international /
unskilled / skilled / legal / illegal / worker / family / forced / voluntary / involuntary,
etc.

Distinctions based on political economy criteria may nuance conventional distinctions

between different types of migrants in ways that have significance for how we
understand, conceptualise and analyse the dynamics of migration. For example, many
households and communities in a variety of migration contexts are engaged in both
‘internal’ and ‘international’ migration. The significance of different migration
strategies and outcomes for the migrants, households or communities concerned is
affected not only by whether migration is international or not, but also by the nature
and circumstances of the migration experience – for example, whether it represents a

31
defensive survival strategy or a positive strategy for asset accumulation, whether
migrants exercise individual agency in their migration decisions, and how people’s
strategies are affected by particular social relations, networks and institutions. Issues
of distance, communication links, migration policies, border controls and migrant
status are likely to feature as central factors in most situations of international
migration, but they are not necessarily the most important factors affecting the actual
causes, patterns and implications of people’s mobility. Analysis of “worker”
migration focused on role of labour markets or state policies on labour migration may
not pay adequate attention to other crucially important institutions, such as family
structure, land distribution, property relations, rural food markets and local brokerage.

6) The complex, dynamic and multi-layered nature of migration streams that these
categories highlight suggests that only a comparatively complex and multi-layered
mix of policy measures could hope to achieve desired impacts on any migration
process. For example, where migration includes large numbers of ‘survival’ or
‘coping’ migrants, or where it can be demonstrated that migrants’ movement is
determined and controlled to a significant extent by identifiable and exploitative
social and economic structures, actors and relationships – and, crucially, where policy
is focused on the well-being of migrants and their communities, or on the protection
of rural livelihoods and/or slowing permanent rural-urban migration – the mix of
relevant policy might include measures to address the underlying vulnerability of the
poorest households and/or the regulatory environment in which exploitative economic
and other actors operate. These same migration streams might also include large
numbers of ‘accumulating’ migrants whose livelihoods or strategies might either be
jeopardised or supported by the same policies.

What is deemed ‘appropriate’ policy action is, of course, a highly political and
contested question, reflecting differing views as to what policy should ultimately be
seeking to achieve: improved well-being of vulnerable migrants and their households
and communities? Reduced rates of rural-urban or international migration? Improved
“development outcomes” (however defined) from migration streams? Maximisation
of remittance flows? Different policy and other actors, of course, will be seeking
varied, and sometimes conflicting, policy outcomes at different levels and in different
contexts. Consequently, any migration stream is likely to be influenced (whether

32
intentionally or not) by highly dynamic interactions of policy objectives, measures
and impacts that will vary considerably from one context to another and over time. By
conceptualising or categorising migration streams on the basis of (micro-) political
economy criteria, there is improved scope for analysing the differential and dynamic
impacts of various or specific policies on particular migration streams or on specific
sub-groups of migrants and non-migrants involved.

7) A political economy approach to categorising migration could provide a useful way

into analysing so-called ‘mixed migration’ flows. One of the reasons that the
phenomenon of mixed migration is so problematic for policy-makers and researchers
is that the migration movements involved and associated networks are not clearly
segmented according to conventional migration ‘categories’. This reflects the
complex institutional and political and economic dynamics at play, including, for
example, the degree of power and control over people’s movement that is exerted by
non-state actors, including smugglers and traffickers who (unlike states’ migration
control institutions) are not concerned with which categories particular migrants
belong to. The focus on migrants’ relative agency in migration processes and the
structure and dynamics of migration processes (mediated by social institutions and
power relations) might help to shed light on commonalities and differences in the
experiences, problems and outcomes for different migrants caught up in ‘mixed
flows’, despite diversity in the original causes and initial dynamics of their migration.

As highlighted by Anthony Richmond’s concept of ‘reactive’ migration, the agency of

migrants whose initial movement was compelled for whatever reason – by violence
and insecurity, fear of persecution, development projects, and/or as a survival strategy
in response to an economic, environmental or other shock, or due to collapsed or
destroyed livelihoods – is likely to be significantly constrained in various ways
(Richmond, 1994). However, some refugees and internally displaced persons have a
lot more control over their own destiny than others, and some certainly enjoy greater
agency following their initial flight than many ‘economic’ migrants. Whether in the
context of complex ‘mixed flows’ or in the context of more discrete migration
processes or systems, what determines migrants’ relative agency, and hence much
about the circumstances, dynamics, experience and outcomes of their migration, is
how their movement and welfare is affected by the particular social, economic,

33
political relational and institutional context in which it takes place, both at the micro-
and macro levels. For example, where movement and welfare is supported or
facilitated by humanitarian agencies and permissive state policies and institutions that
protect certain rights (e.g. leave to enter, free movement, right to work), ‘compelled’
migrants are likely to enjoy greater degrees of agency and possibly better welfare than
migrants moving in circumstances where fundamental rights are denied, and where
exploitative relationships (e.g. with traffickers, labour recruiters, employers) dominate
the dynamics of movement. As a consequence of the specific circumstances of their
movement, migrants who initially lack agency may gain more power and control over
time, while others may find themselves increasingly at the mercy of other actors and
other interests.

8) Migration is highly embedded in a variety economic, social and political processes,

structures and relations at micro (household and community) and wider levels. Its
effects – for example on poverty – are therefore extremely difficult to isolate or
measure (Deshingkar and Start, 2003:2). The embedded nature of migration also
means that migration itself may not be the most important or significant issue from
the point of view of the people involved. As reflected in much of the work on
migration within livelihoods research, migration or mobility may be only one of a
variety of strategies that people use to enhance their welfare or to cope or survive, and
its relative importance compared to other strategies and factors affecting people’s
lives may vary over time and may often be overshadowed by other factors affecting
their lives. An actor-oriented approach may encourage a more holistic view of
communities’ experiences and predicament which does not prioritise migration over
other issues. The significance of migration for different migrants and their families /
households may vary considerably even within a specific location.

9) Too narrow a focus on migration without closer investigation of its precise and
relative significance for particular households / communities / sectors might lead to
flawed assumptions about commonalities and similarities between different migration
experiences or phenomena. Although they may look similar, we may not always be
treating like with like when comparing different migration streams in terms of their
causes or significance for the political economy in the sending and destination
locations and at trans-national level.

34
For instance, while migration resulting from poverty is common to different areas and
contexts, the underlying causes of poverty – and hence of the migration associated
with it – can differ significantly from one situation to another (cf. Green, 2005:36-37).
Du Toit et al. contrast the very different dynamics of poverty between three areas of
South Africa’s Cape, each of which has different implications for local migration
dynamics. In one area (Mount Frere), severe biophysical constraints combine with
lack of infrastructure development, poor returns on agricultural labour and a long
history of capital penetration to create widespread and chronic poverty and
vulnerability. In another area (Ceres), the agrarian economy is highly productive, but
ownership of the economy is concentrated in the hands of a small, landed white elite;
integration of this economy into global markets and increasing buyer control of
agricultural commodity chains have resulted in extreme casualisation and insecurity in
the agricultural labour market on which many of the poorest households depend.
Meanwhile, in Cape Town’s African suburbs, poverty is shaped at the local level by
the ‘racial and spatial geo-politics of the post-industrial metropolitan labour market’
which emphasises high-skilled manufacturing and services and relegates the chronic
poor, including recent migrants, to ‘unsafe and crime-ridden peri-urban racial
ghettoes’ (du Toit et al., 2005:11). Where poverty appears to play a major role in
generating migration, it is only by explaining the dynamics and causes of poverty
(rather than simply describing this poverty) that the causes of associated migration
can be properly appreciated or tackled.

10) Examining migration and related processes and their causes and consequences
within the context of local, national and international political economy may reveal
dynamics and implications associated with migration that might otherwise be missed
or downplayed. Richard Black and Mohamed Sessay’s 1997 study of Liberian and
Sierra Leonean refugees and their impact on land use and deforestation in three
villages in Guinea, for instance, challenged the widely held view that rapid rates of
deforestation had simply resulted from the increased demand for land caused by the
arrival of large numbers of refugees. Historical examination of the development of the
agricultural economy of the region, and of socio-political change in local communities
revealed a more complex relationship between refugee arrivals and deforestation.
Local farmers – keen to establish a claim over forest land in a context where
indigenous rights of access appeared under threat, and keen to invest in production of

35
coffee and other cash crops (and encouraged to do so by government policy) –
allowed refugees to clear forest land which they themselves later planted with coffee
or cocoa or other cash crops. This study, they argue, ‘highlights the potential role of
refugee farmers as a resource that is capable of being mobilized by both the state and
local communities in a developing struggle for control over the land resources of the
forest region’. Hence ‘the presence of refugees in an area is best viewed as a catalyst
for more dynamic social relations, rather than as necessarily having one or other
‘environmental’ impact’ (Black and Sessay, 1997:605).

11) A relational political economy approach encourages an expanded range of issues

to be investigated in the context of globalisation and its implications for migration. In
addition to exploring the trans-national nature of migration networks, and the
international economic, commercial and technological transformations that are
facilitating particular forms of international migration, it encourages attention to how
globalised capital and markets have penetrated and affected the livelihoods and
communities of people involved in particular migration movements, and how this has
created or influenced the local dynamics and significance of migration and other
related survival, risk-reduction and welfare maximisation strategies. The relaxation of
labour controls and standards associated with economic liberalisation – driven by the
dynamics of global capitalism – has led to the progressive casualisation of labour
markets, particularly in poorer countries. Start and Johnson observe that while
freedom and flexibility in labour market institutions is arguably essential to the
efficiency and flexibility of enterprise in a global economy, ‘the result has left many
livelihoods increasingly divided, multiple and diverse, not to mention precarious and
pauperised’ (Start and Johnson, 2004:9).

5. Conclusion

Migration is a complex phenomenon that is deeply embedded in the structures,

relationships and processes of micro- and macro political economies and in the
dynamics of people’s livelihoods at the local level. Specific patterns of migration are
therefore determined to a great extent by the dynamic interaction of individual or
household livelihood strategies with a broad range of micro-, meso- and macro level

36
relationships, processes, institutions and structures that make up the social, economic,
political and historical contexts in which migration takes place – such as wealth
distribution and access to land and other resources, kinship structures, local and wider
governance institutions, labour markets, social and commercial networks, state
welfare and security regimes. The particular opportunities and constraints affecting
people’s migration options, strategies, experiences and outcomes depend to a great
extent on the differential distribution and changing dynamics of power, vulnerability,
agency and opportunity within households, communities and the wider population.

This Working Paper proposes the development of an approach to researching and

analysing migration processes that combines a livelihoods approach to analysing
local-level migration dynamics with a relational political economy perspective that
deepens understanding of the social, economic and political processes, interests and
power dynamics interacting with migration at different levels (e.g. causing patterns of
vulnerability or creating opportunities that encourage migration). Livelihoods
approaches are extremely useful for capturing the agency and capabilities of migrants
and their households at the local level, and for exploring the multi-dimensional
dynamics of migration and related livelihood opportunities, strategies and outcomes
within particular communities. However, migrants’ (or their households’) ability to
exercise agency or ‘power to’ pursue particular migration strategies can only be
properly understood in relation to the varying constraints and opportunities created by
dynamic (and often competitive) social, economic and political relations and
institutions through which people and groups exert ‘power over’ each other (Mosse,
2007). There is a need to consider the potential for competition, exploitation and
unequal power dynamics between key actors in migration processes, and to explore
how the relationships between these actors are shaped and mediated by a variety of
social, economic and political structures, processes and institutions. ‘Relational’
political economy analysis centres directly on the relationships, processes and
structures of relative power and agency between different actors involved. It also
helps to trace and explain the interaction and implications of exclusionary and
inclusionary processes and relationships over time and at different levels – micro,
meso and macro.

37
There is no single approach to exploring and analysing the micro- and macro political
economy of a particular context. It is necessary to adopt a historical and cross-
disciplinary perspective, drawing flexibly on a number of frames of reference,
including, potentially, international political economy, gender studies, anthropology,
sociology, social history and political geography. ‘Value chain analysis’ provides a
potentially useful framework to begin exploring the links and interaction between
local livelihoods and specific migration networks, systems or processes, and,
importantly, for investigating relative power relations and actors’ agency within these.
Migrants themselves may exercise considerable agency and control within some
migration processes, whereas in others, individual migrants may be highly vulnerable
to the exploitative interests of other actors. Intermediary actors (e.g. labour recruiters,
brokers, smugglers) may have considerable power in some migration streams; others
might be controlled more by state or other actors at the ‘receiving’ end. Any
migration stream is likely to include various migrants or groups moving with varying
degrees of agency in different circumstances and in response to different causal
dynamics. Different actors within the migration ‘chain’ are likely to exert different
forms of control at different levels and points in time.

A key objective of a livelihoods and relational political economy approach is to move

beyond overly-deterministic economics-based models and theories of migration, and
beyond ever-more detailed descriptions of migration processes, to seek improved
evidence-based explanations for the causal and consequential dynamics of migration
in different contexts and at different levels. While these dynamics are highly varied,
complex and context-specific, variability and complexity need not lead to resignation
as regards the prospects of developing useful theory (Castles, 2008:14). A livelihoods
and political economy approach, incorporating close attention to the dynamics both of
local livelihoods and of specific migration processes not only provides a robust
theoretical and conceptual framework through which to address this complexity. Over
time, it also has the potential to reveal common themes, problems and challenges in
the midst of all the complexity. Common themes might emerge, for example, around
issues such as the expansion of commercial markets into poor rural areas, the
casualisation of urban labour markets, access to land and other resources, local and
national governance, or environmental change. Rather than being imposed from the
‘top-down’ through pre-ordained and reductionist models of what drives migration,

38
these insights would emerge incrementally from the ‘bottom-up’ and from detailed
research material relating to a wide range of migration processes and contexts.

References
Bhatia, M. and J. Goodhand et al. (2003) ‘Profits and Poverty: aid, livelihoods and
conflict in Afghanistan’, in S. Collinson (ed.), Power, livelihoods and conflict:
case studies in political economy analysis for humanitarian action. HPG
Report 13. London: Overseas Development Institute (Humanitarian Policy
Group).
R. Black and M. Sessay (1997) ‘Forced Migration, Land-Use Change and Political
Economy in the Forest Region of Guinea’, African Affairs, 96, 587 – 605.
Black, R., C. Natali and J. Skinner (2006) ‘Migration and Inequality’, draft
background paper for Equity and Development. World Development Report
2006. Sussex: Research Centre on Migration, Globalisation and Poverty.
Bracking, S. (2003) The Political Economy of Chronic Poverty. CPRC Working Paper
23. Manchester: Institute for Development and Policy Management (Chronic
Poverty Research Centre).
Carney, D. (ed.) (1998) Sustainable Rural Livelihoods. What Contribution Can We
Make? Nottingham: Russell Press Ltd., for DFID.
Castles, S. (2008) ‘Development and Migration – Migration and Development: What
Comes First?’. Paper prepared for the Social Science Research Council
Conference: Migration and Development: Future Directions for Research and
Policy, New York, 28th February – 1st March 2008. Oxford: International
Migration Institute.
Collinson, S. et al. (2002) Politically Informed Humanitarian Programming: Using a
Political Economy Approach. Network Paper 41. London: ODI.
Collinson, S. (ed.) (2003) Power, livelihoods and conflict: case studies in political
economy analysis for humanitarian action. HPG Report 13. London: Overseas
Development Institute (Humanitarian Policy Group).
DAC (2006) Promoting Pro-Poor Growth: Agriculture. DAC Guidelines and
Reference Series; a DAC Reference Document (extract from Promoting Pro-
Poor Growth: Policy Guidance for Donors. Paris: OECD (Development
Assistance Committee).
David, R. (1995) Changing Places: Women, Resource Management and Migration in
the Sahel. London: SOS Sahel.
de Haan, A. with K. Brock, G. Carswell, N. Coulibaly, H. Seba and K. Toufique
(2000) Migration and Livelihoods: Case Studies in Bangladesh, Ethiopia and
Mali. IDS Research Report 46. Sussex: Institute of Development Studies.
de Haan, A., K. Brock and N. Coulibaly (2002) ‘Migration, Livelihoods and
Institutions: Contrasting Patterns of Migration in Mali’, Journal of
Development Studies, Special Issue.
de Haan, L. (2005) The Livelihood Approach and African Livelihoods. Leiden:
African Studies Centre.
de Haan, L. and A. Zoomers (2005) ‘Exploring the Frontier of Livelihoods Research’,
Development and Change 36(1).
de Haas, H. (2008) Migration and Development: A Theoretical Perspective. Working
Paper 9. Oxford: International Migration Institute.

39
de Waal, A. (1997) ‘A Re-assessment of Entitlement Theory in the Light of Recent
Famines in Africa’, Development and Change 21(3).
du Toit, A. (2005) Chronic and Structural Poverty in South Africa: Challenges for
Action and Research. CPRC Working Paper 56 & PLAAS Chronic Poverty
and Development Policy Series No.6. Manchester & Cape Town: Chronic
Poverty Research Centre & Programme for Land and Agrarian Studies.
du Toit, A., A. Skuse and T. Cousins (2005) The Political Economy of Social Capital:
Chronic Poverty, Remoteness and Gender in the Rural Eastern Cape. PLAAS
and CPRC Working Paper. Manchester & Cape Town: Chronic Poverty
Research Centre & Programme for Land and Agrarian Studies.
Deshingkar, P. and D. Start (2003) Seasonal Migration for Livelihoods in India:
Coping, Accumulation and Exclusion. Working Paper 220. London: Overseas
Development Institute.
Duffield, M. (1994) ‘The Political Economy of Internal War: Asset Transfer,
Complex Emergencies, and International Aid’, in J. Macrae and A. Zwi (eds.)
War and Hunger: Rethinking International Responses to Complex
Emergencies. London: Zed Books.
Ellis, F. (2000) Rural Livelihoods and Diversity in Developing Countries. Oxford:
Oxford University Press.
Ellis, F. (2003) A Livelihoods Approach to Migration and Poverty Reduction. Paper
commissioned by the Department for International Development. London:
DFID.
Everett, B., R. Kuhn and R. Silvey (200*) ‘Gendered Migrants Networks and the
Health of the Left Behind: Evidence from Indonesia’. [incomplete reference].
Farrington, J.D., D. Carney, C. Ashley, and C. Turton (1999) Sustainable Livelihoods
in Practice. Early Application of Concepts in Rural Areas. Natural Resources
Perspectives 42. London: Overseas Development Institute.
Gereffi, G. (1995) ‘Global Production Systems and Third World Development’, in B.
Stallings (ed.) Regional Response: The New International Context of
Development. Cambridge: Cambridge University Press.
Gereffi, G. and M. Korzeniewicz (eds.) (1994) Commodity Chains and Global
Capitalism. Westport, CT: Greenwood Press.
Gold, S. J. (2005) ‘Migrant Networks: A Summary and Critique of Relational
Approaches to International Migration’, in Mary Romero and Eric Margolis
(eds.) The Blackwell Companion to Social Inequalities, Malden MA:
Blackwell.
Green, M. (2005) Representing Poverty and Attacking Representations: Some
Anthropological Perspectives on Poverty in Development. Global Poverty
Research Group Working Paper 9. Manchester and Oxford: GPRG.
Hammond, L., J. Bush, K. Savage and P. Harvey (2005) The effects of food aid on
household migration patterns and implications for emergency food
assessments. Desk Review for the “Strengthening Emergency Needs
Assessment Capacity” (SENAC) project. Rome: World Food Programme,
with the Overseas Development Institute.
Harriss, J. (2007) Bringing Politics Back In To Poverty Analysis: Why Understanding
of Social Relations Matters More for Policy on Chronic Poverty than
Measurement. Q-Squared Working Paper 34 (April 2007); paper
commissioned by the Chronic Poverty Research Centre and presented at the
CPRC ‘Workshop on Concepts and Methods for Analysing Poverty Dynamics

40
 and Chronic Poverty’, Manchester 2006. Toronto: Centre for International
Studies, University of Toronto.
Harris, J. and M. Todaro (1970) ‘Migration, Unemployment and Development: A
Two Sector Analysis’, American Economic Review, Volume 60.
IMI (2006) Towards a New Agenda for International Migration Research. Oxford:
International Migration Institute.
Kaag, M. and R. van Berkel, J. Brons, M. de Bruijn, H. van Dijk, L. de Haan, G.
Nooteboom and A. Zoomers (2003) Poverty is Bad: Ways Forward in
Livelihood Research. CERES ‘Pathways of Development’ project. Leiden:
Research School for Resource Studies for Development (CERES)
http://www.livelihoods.org/info/docs/Leiden5.pdf
Keen, D. (1994) The Benefits of Famine: A Political Economy of Famine and Relief in
Southwestern Sudan, 1983-1989. Princetown, NJ: Princetown University
Press.
Kothari, U. (2002) Migration and Chronic Poverty. Working Paper No.16.
Manchester: Institute for Development Policy and Management.
Lautze, S. and A. Raven-Roberts (2006) ‘Violence and Complex Humanitarian
Emergencies: Implications for Livelihoods Models’, Disasters 30(4).
Le Billon, P. et al. (2000) The Political Economy of War: What Relief Agencies Need
to Know. Network Paper 33. London: ODI.
Mabogunje, A.L. (1970) ‘Systems Approach to a Theory of Rural-Urban Migration’,
Geographical Analysis, Volume 2.
Massey, D. (1993) ‘Power-geometry and a Progressive Sense of Place’, in Bird, J. et
al. Mapping the Futures. London: Routledge.
McDowell, C. and A. de Haan (1997) Migration and Sustainable Livelihoods: A
Critical Review of the Literature. IDS Working Paper 65. Sussex: Institute of
Development Studies.
Mosse, D. (2007) Power and the Durability of Poverty: A Critical Exploration of the
Links Between Culture, Marginality and Chronic Poverty. Working Paper 107.
Manchester: Chronic Poverty Research Centre.
Murray, C. (2001) Livelihoods Research: Some Conceptual and Methodological
Issues. Background Paper 5. Manchester: Chronic Poverty Research Centre.
North, D. (1990) Institutions, Institutional Change and Economic Performance.
Cambridge: Cambridge University Press.
Pain, A. and S. Lautze (2002) Addressing Livelihoods in Afghanistan. Kabul:
Afghanistan Research and Evaluation Unit.
Rafique, A., D. Massey and B. Rogaly (2006) Migration for Hard Work: A Reluctant
Livelihood Strategy for Poor Households in West Bengal, India. Working
Paper T17. Sussex: Development Research Centre on Migration, Globalisation
and Poverty.
Raikes, P., M.F. Jensen and S. Ponte (2000) Global Commodity Chain Analysis and
The French Filiere Approach: Comparison and Critique. Copenhagen: Centre
for Development Research.
Rao, G.B. (2001) Household Coping / Survival Strategies in Drought-Prone Regions:
A Case Study of Anantapur District, Andhra Pradesh, India. SPWD-
Hyderabad Centre.
Richmond, A. (1994) Global Apartheid. Oxford: Oxford University Press.
Sen, A. (1981) Poverty and Famines. Oxford: Clarendon Press.

41
Skeldon, R. (2003) ‘Migration and Poverty’. Paper presented at the conference on
“African Migration and Urbanization in Comparative Perspective”,
Johannesburg, South Africa, June 4-7, 2003.
Stark, O. and D. Bloom (1985) ‘The New Economics of Labour Migration’, American
Economic Review 75(2).
Start, D. and C. Johnson (2004) Livelihood Options? The Political Economy of
Access, Opportunity and Diversification. Working Paper 233. London:
Overseas Development Institute.
Tamas, K. (2006) Towards a Migration for Development Strategy. Stockholm:
Ministry for Foreign Affairs, Sweden.
Vertovec, S. (2002) ‘Transnational Networks and Skilled Labour Migration’. Paper
presented at the conference: Ladenburger Diskurs “Migration” Gottlieb
Daimler- und Karl Benz-Stiftung, Ladenburg, 14-15 February 2002.
Waddington, H. and R. Sabates-Wheeler (2003) How Does Poverty Affect Migration
Choice? A Review of Literature. Working Paper T3. Brighton: Development
Research Centre on Migration, Globalisation and Poverty.
Young, H., A. Monium Khider Osman, Y. Aklilu, R. Dale, B. Badri and A. Jabbar
Abdullah Fuddle (2005) Darfur – Livelihoods Under Siege. Medford MA:
Feinstein International Famine Center (Tufts University).

42