The literature on poverty in general and poverty and children in particular is reviewed in this chapter to understand how poverty affects the well being of children. The definition and measurement of poverty has been the subject of several studies undertaken. However, it was not until recently that child poverty became the focus of research as separate from poverty. One approach to defining poverty is in terms of some kind of deprivation and lacking 'resources to obtain the types of diet, participate in the activities and have the living conditions and amenities which are customary, or at least widely encouraged or approved' in a particular society (Townsend, 1979, p 31).
It must be recognized at the outset that defining, conceptualizing and measuring poverty is a challenge since recognizing the different facets of poverty reflect varying value judgments related to political ideologies. 'Discourses of poverty and exclusion are culturally and historically specific; they change over time and reflect different political and social conditions and dispositions.' (Ridge and Wright, 2008, p 3)
Poverty has been explained in terms of structural and social and cultural factors. Theories of poverty due to structural factors attribute the cause of poverty to factors beyond the control of the individual. Culture based explanations are more individualistic and hold the individual responsible for their state. The argument is that people are poor for reasons of their own making. Structural causes are those that are beyond the control of the individuals living in poverty. (Rupasingha and Goetz, 2003)
There is a need to adopt a multi-disciplinary approach to poverty that encompasses the human rights based approach; the basic needs approach and the capability approach. The monetary approach is considered inadequate since it fails to recognise the multi-faceted nature of poverty. A purely monetary approach is inappropriate in measuring poverty since it fails to take into consideration household structure, gender and age. It is possible that within a household there could be variation in access to resources. Members of a household could be discriminated against on the basis of age and gender. In this case an increase in income will not result in a proportionate increase in the share of resources for all household members (Minujin et al, 2006).
There is no general consensus on the definition and measurement of poverty. An overarching similarity in the various definitions of poverty is the underlying notion that poverty implies being without key indicators of acceptable social functioning (Gordon 2006 as cited in Ridge and Wright, 2008, p 39). Sen described the core of poverty on the basis of the human need to 'live without shame' (1983, p163).
Absolute versus relative poverty has been the subject of much discussion amongst academics. The notion of absolute poverty suggests that the person or household experiencing it does not have some minimum amount of consumption to survive. Hence it is also referred to as subsistence poverty. The concept of absolute poverty is associated with the early works of Rowntree.
A family living on the scale allowed for must never spend a penny on railway fare or omnibus. They must never go into the country unless they walk. They must never purchase a halfpenny newspaper or spend a penny to buy a ticket for a popular concert â€¦ and what is bought must be of the plainest and most economical description. (Rowntree, 1901, p167)
The concept of individuals being denied access to transport, to the print media and entertainment is portrayed in the above description of a poor person. This is relevant in the case of Pakistan where absolute poverty would mean denial of access to adequate food, clothing and shelter and can be extended to the insufficient access to health and education and leisure facilities.
The relative approach to poverty conceptualizes poverty in relational terms and can differ according to social circumstances. Townsend is largely attributed with being the proponent of the relative approach to poverty.
Individuals, families and groups in the population can be said to be in poverty when they lack the resources to obtain the types of diet, participate in the activities and have the living conditions and amenities which are customary â€¦ in the societies to which they belong. (Townsend, 1979, p31)
However, it has been observed that researchers combine the two approaches when they operationalise research on poverty. (Ridge and Wright, 2008, p 40). The first step in a research design aimed at studying poverty would be to identify the essential elements of a decent standard of living in a particular society. The next step would be to measure the number of people actually deprived due to inadequate incomes or resources. There is no doubt that any study has to be contextualized and the indicators of poverty are in both absolute and relative terms.
Sen contends that the conceptualization of poverty in absolute versus relative terms depends on what is being measured namely commodities, incomes or capabilities. (Sen, 1983). The concept of capabilities is introduced by Sen which he considers the "closest to the notion of standard of living".
Perceptions of poverty and definitions and measures are influenced by the social circumstances in which poverty arises. The perception of poverty in a developed nation will be different from the one in a developing nation. For instance the notion that a child will not attend primary school due to poverty is out of the question in a developed nation whereas in a developing country such as Pakistan it is a reality. Consequently view of poverty and the concept of deprivation due to insufficiency of resources will vary across countries and can change over time.
The focus of academic research has been on objective indicators of poverty such as income, expenditure and other indicators of standard of living. The rationale for adopting this approach is that objective indicators are measurable and free from subjective bias. This renders such indicators a more convenient tool for comparison of trends both spatially and over time. However, there is a growing concern that those experiencing poverty themselves may have a different perception about their own condition. At times they may not define themselves as poor. It has bee suggested by Lister (2004) that it is important to extend our understanding beyond academic definition and measurement to the experiences and the agency of those who are in poverty.
The use of different indicators to measure poverty will give different results with respect to numbers and those experiencing poverty. The first question is whether to use income or consumption as a measure of poverty. Income is easier to measure but expenditure is more relevant. The welfare of a household ultimately depends on the utility gained from the consumption of goods and services. At this point it is worth noting that expenditure on members within the household will be more relevant since intra-household allocation of resources can differ. Although higher income may result in higher consumption expenditure but this may not always be the case. For instance those with savings and assets may be able to enjoy a higher level of expenditure than indicated by their income. On the other hand those repaying debts will have lower levels of expenditures and consequently a lower standard of living. The composition of expenditure will also be an important indicator of how resources are allocated to alternative uses affecting the welfare of members of a household both in the current time period and in the future. If the head of the household suffers from either drug or alcohol addiction the household's resources will be diverted towards the procurement of such substances thereby reducing the resources available for fulfilling basic needs and the household will fall below the poverty line. It has been argued that expenditure should be adopted as a measure of poverty. (Ringen 1988)
A related question that arises is whether to measure resources available or deprivation experienced. Income measures do not reflect the standard of living of households. This partly due to a distinction between what Rowntree (1901) referred to as primary poverty and secondary poverty. Primary poverty occurs when people do not have sufficient resources to fulfill their needs. Secondary poverty arises when people have the resources but do not spend on necessities. Income may be spent on non-essential activities. The distinction between primary and secondary poverty is relevant in the case of Pakistan and the current study. If enrolment in primary school is one type of deprivation and the lack of it a measure of poverty, then it is plausible that a family may have the resources but still not send their children to school. Consequently a purely monetary measure of poverty will not capture such households. It is also possible that a household characterized as income poor may still send children to school.
The literature reveals poverty as one of the determinants of child labour. The luxury axiom states that if the wages of parents are high enough there may not be any need for sending children to work. Basu identifies "multiple equilibria - one in which wages are low and children work and another in which wages are high and children do not work" (Basu and Van 1998). This result arises because child labour is not because of parental selfishness but because of the parents' concern for the household's survival. The evidence seems to confirm the axiom that parents only send their children to work when compelled by circumstances. (Basu and Tzannatos 2003). However, the link between poverty and child labour is a tenuous one and has been questioned in some empirical studies. Ray (2000a, 2000b), Bhalotra and Heady (2003)
Another aspect of child labour that has been explored is the trade off between child labour and schooling. Psacharopoulos (1997), Patrinos and Psacharopoulos (1997), Akabayashi and Psacharopoulos (1999), Ravillion and Wodon (2000), Ray (2000a, 2000b, 2000c), Ray and Lancaster (2005), Ray (2006), Rosati and Rossi (2006), Gunnarsson, et al. (2006) are part of the literature that provides evidence of the adverse impact of child labour on the child's learning opportunities which in turn negatively impacts human capital formation.
A number of empirical studies have compared the determinants of child labour in the context of different countries or within regions of the country. The data used in these studies is usually from household surveys. Bhalotra and Heady (2003) have compared Ghana and Pakistan; Ray (2000a and 2000b) analyses the evidence from Peru and Pakistan; Patrinos and Psacharopoulos (1997) carries out an empirical analysis of family size, schooling and child labour in Peru; Akabayashi and Psacharopoulos (1999) consider the trade off between child labour and human capital formation in the context of Tanzania; Rosati and Rossi (2003) study the determinants of children's working hours and school enrollment based on the evidence from Pakistan and Nicaragua; Basu et al. (2003) analyse birth order, gender and wealth as determinants of child labour based on the Indian experience. This paper contributes to the present literature by considering the determinants of child labour and school enrolment at the community level among migrant communities in Karachi. A disaggregated approach highlights the heterogeneity of the population in a large city. The aggregate picture fails to reveal differences within households belonging to diverse ethnic and religious backgrounds.