4.2.1 The household
The household (usually defined as a group of people who ‘eat from the same pot’ and live in the same residential unit) has been the commonest unit of analysis for studies of acute poverty to date. It is increasingly recognised that well-being is stratified within the household, especially along lines of gender, age, and health status (Haddad et. al., 1997; Miller, 1997). In many regions of Africa, girls in poor households are less likely than boys to receive adequate education and health care. Throughout their lives, poor women perform a triple role – reproductive work (including frequent childbearing and responsibility for the care of the household), productive work (often highly physically burdensome), and community work – placing additional obstacles in the way of escaping poverty. Thus, some women in poor, but not acutely poor, households may be acutely poor. Their poverty is lasting and difficult to escape even when other members of the household improve their situation. The process through which changes in household composition – through marriage, divorce, birth, death and migration – affect mobility also is differentiated along lines of gender, age and health status. In addition (i) assumptions about the durability of households are often wrong (making it difficult to delineate ‘who’ is in the household) and (ii) many of the acute poor may not be ‘in a household’ e.g. street children and the destitute.

4.2.2 Individuals
Another level of analysis is the individual. Its actualisation requires data about incomes, consumption, social conditions and relations, personal attributes and motivations, and so on, that relates to specific individuals. Necessary data will be those at individual and household levels important to understand intra-household processes. The great advantage of the analysis of individuals is the ease of definition and identification. A disadvantage is that most development and poverty reduction interventions are aimed at ‘the household’ and policy-makers steer clear of the complexity of intra-household relations.

4.2.3 Social groups
Acute poverty focuses on specific groups of people. Sometimes these are ‘real’ groups and have a common social identity and associational forms (e.g. members of pastoralist communities). At other times, these are groupings of people who have common characteristics but who do not share a social identity and/or practice association (e.g. physically challenged people in rural areas, daughters-in-law). Others vulnerable sub-groups include poor people living in remote rural areas such as marginalised tribes, ethnic groups and nationalities, immigrants and pastoralists,  older people, people living with HIV/AIDS, and women (particularly widows and female-headed households). It has often been the case that a significant proportion of the acutely poor are those who experience multiple and overlapping vulnerabilities (Abul Naga, 1994). It has also been observed that the most ‘endangered’ sub-groups are the elderly poor, who regularly become very poor at the end of their life due to marginalisation and poor health, and die without escaping poverty; and the physically challenged poor, whose experience of social-economic discrimination and physical impairment may be sufficient not only to make them poor but also severely to reduce the life chances of their children.

4.2.4 Geographical Areas
The ‘geographical location’ of the poor in remote rural areas, urban slums, and ‘interstitial’ communities usually make them vulnerable to natural hazards, pollution, agro-climatic shocks, conflict and instability. Their isolation creates infrastructural and social remoteness, particularly from markets, health centres and centres of political decision-making, increasing the susceptibility of the entire population to acute poverty. More often than not, acutely poor areas are likely to contain acutely poor people.  Within these spatial poverty traps, it is unlikely that individuals or households can escape from, or reduce their poverty. If they do, it is likely going to be at the expense of others.  

5. What are the Causes of Chronic Poverty?
5.1. Little pictures and big pictures
Why does acute poverty occur? There is no single answer to that very important question. A vast range of reasons could cause acute poverty (Table 2). Using analytical frameworks, this article will focus on specific sets of these factors and examine the ways in which they interact to explain the incidence and nature of acute poverty. These frameworks can range from the simple, such as environmental determinism which argues that poverty is the result of too many people living on poor lands that are unhealthy for humans, to the highly complex, such as theories of globalisation (Castells, 2000) that attempt to weave all of the factors in Table 2, and more, into an analysis that goes from the micro to the macro level. The frameworks suggested here will have both an analytical perspective seeking to know why things are the way they are, and a normative perspective seeking to identify how things could be done better so as to alleviate acute poverty. From the broadest sense, research on acute poverty can be viewed as the resultant effect of development failure incapable of delivering improved livelihoods. Thomas (2000:43), albeit with many caveats, has attempted to summarise the debates about development into four main positions:

Table 2: Causes of acute poverty

5.2   Analytical frameworks
5.2.1 Quantitative panel data set analysis
Analysis of quantitative panel data sets has so far been the main means by which researchers have investigated the existence, extent and causes of acute poverty. Two major forms are identified: panel data where some or all of those surveyed in the first round are surveyed again in subsequent rounds; and repeated cross sections, where a new sample is selected each time. The former provides a means of understanding poverty dynamics, down to the level of the household or individual, whereas the latter provides a means of trend analysis to the level of geographic area, social group or 'community'. Due to the practical difficulties associated with undertaking repeated panel data collection over long periods of time, most panel data sets either span a short period of time, or have few data points. Therefore, 'most panel data is not suitable for addressing life cycle aspects of poverty dynamics at the individual household level. Information on life cycle experiences of individual households will generally be more effectively captured using retrospective techniques with a qualitative focus' (McKay, 2001). Quantitative analysis of acute poverty generally drawing upon income/consumption data. It can also draw upon other measures of poverty and deprivation. Child anthropometric data, generally collected in national longitudinal household surveys, may be of particular use, as it can provide both an accurate representation of present household well-being as well as a proxy for retrospective data. Most researchers who have been engaged in the analyses of longitudinal quantitative data have been able to cross-tabulate the vicissitudes of acute poverty with various other poverty indicators and personal characteristics, including income and expenditure (Chaudhuri and Ravallion 1994, Gaiha 1989); ethnicity (Devine et al., 1992; Rodgers and Rodgers 1993; Blee 1996);  access to land, labour, and other assets (Gaiha 1989; Chaudhuri and Ravallion 1994; McCulloch and Baulch 2000); levels of education and skills (Gaiha 1989; Jalan and Ravallion 1998; McCulloch and Baulch 2000; Rodgers and Rodgers 1993); health (Jalan and Ravallion 1998); household structure and life cycle effects (Devine et al., 1992; Jalan and Ravallion 1998; McCulloch and Baulch 2000; Rodgers and Rodgers 1993); food security (Braun 1995; Chaudhuri and Ravallion 1994; Jalan and Ravallion 1998); and geography, climate and ecology (Gaiha 1989; Jalan and Ravallion 1998). Though not clearly, cross-tabulate can indicate correlates, showing causes or consequences of acute poverty. 

5.2.2 Livelihoods analysis
This is a common approach, both for academic and policy-related work. A number of summaries are available but the approach introduced by Ellis (1998, 2000a) is widely utilised (Table 3). Livelihoods analysis proposes that the way in which a household meets its present and future needs, and pursues its aspirations, must be seen holistically and dynamically (Lipton ad Maxwell, 1992; Adi, 2003). By examining the full set of ‘assets’ at a households disposal the factors that shape the well-being or ill-being of its members can be understood at the micro-level in great detail (Berry, 1998, 1993; Murray 2000) or at a meso-level through sample surveys of settlements and comparative aggregate analysis (Hart, 1995; Ellis 2000b). This can shed light on the ways in which household members, businesses, ‘civil society’, and state action interact to create, maintain or reduce poverty and vulnerability (Brycesson, 1996). According to Murray (2000:118), as household livelihoods often transcend both sectoral and geographical boundaries, it is important that livelihoods research “transcend local ‘communities’ in order to comprehend both intra-household relationships and significant inter-household social relationships”. Factors such as migration, absentee landlordism and other livelihood diversification strategies often require an understanding of the multiple livelihoods of people living in different locations from various socio-economic strata. Particularly through an understanding of historical context and the interactions between macro and micro levels, an awareness of the livelihoods of the urban not-so-poor, for instance, can contribute significantly to an understanding of both the political-economy and social context of poverty in rural areas (Ellis and Freeman, 2004; Ellis and Seeley, 2005). Assets such as natural, physical, human, financial, and social capital can be modified by the researcher so that alternative or additional concepts can be included. In particular, many critics have argued that the orthodox approach fails to incorporate power relations and that ‘political capital’ must be added. Participatory research has highlighted the importance of physical security. Table 4 illustrates the way in which ‘assets’ might be elaborated and of their significance for chronic poverty (Brock and Coulibay, 1999).

5.2.3 Freedoms
During the 1980s and 1990s, the seminal work by Sen (1993) has spurred researchers to examine the entitlements and capabilities of those experiencing deprivation. His recent work (Sen 1999, 2003) takes these ideas forward and proposes that poverty (and overcoming poverty) can be understood in terms of five freedoms (Table 4). Arguably, these permit a holistic analysis of all of the relevant social, economic, political and environmental factors that deepen the appreciation of the mutually reinforcing ways in which private, civil and state action can improve individual, group and social welfare. Operationalising this framework is a contemporary task: Table 4 provides a starting point for relating freedoms to chronic poverty. An advantage of this framework as a starting point is that politics and governance receive adequate and explicit attention. Some would no doubt criticise the strong commitment to market economics implicit in the framework.

5.2.4 Social exclusion
In Europe, the concept of poverty has recently been supplanted by ‘social exclusion’. As this concept has gained its currency within a welfare state context, and as exclusion presupposes inclusion, debates continue regarding the applicability of social exclusion approaches to poverty in developing countries which have never had a functioning comprehensive social welfare system. But according to de Haan (1999), the concept of social exclusion should be viewed as elastic that is, encompassing the multidimensional, relational aspects of deprivation and poverty. For example, people are excluded not only from publicly provided services but from a broad range of social, political and economic institutions, and as a result experience lack of capabilities and entitlements. Social exclusion may therefore be a useful starting point from which to understand the politics surrounding chronic poverty, perhaps especially where chronic poverty is based on the discrimination of a particular social group.
A distinction between socio-political and socio-cultural exclusion may help to steer analysis towards two important sets of issues: the politics of persistent poverty (by comparison with the politics of poverty as a whole) (William, 1970; Jones, 1986), and the renewed debates around cultures of poverty and related concepts of coping strategies. Both analyses would pick out key institutions in the sense of ‘rules in use’. Socio-political analysis would focus on the institutional mechanisms and processes which keep poor people either excluded or adversely included in subordinate, constraining positions. It would also seek out institutional developments (like the institutionalisation of political parties and inter-party competition) which might open windows of inclusion (Narayan, Chambers, Shah and Petesch, 2000). Socio-cultural analysis would look at the more micro level of social relationships and socialisation which transmit values, attitudes and behaviour within and between generations, with impacts on potential for exit from poverty, and which help to direct individuals towards or away from the ‘mainstream’ of society and resource them (or provide entitlements to them) in it. This level of analysis is always nested within wider societal and global processes which would also need analysis (Messil, 1957; Lewis, 1965).

Table 4: Achieving the five freedoms for the acutely poor: implications for public policy.

Source: adapted from Sen (2000).

<<Previous Page| Next Page>>

COPYRIGHT ©2009. Designed and hosted by Sean Cussons Ltd