Content uploaded by Keetie Roelen
Author content
All content in this area was uploaded by Keetie Roelen
Content may be subject to copyright.
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
291
article
© The Policy Press • 2012 • ISSN 1759-8273
Key words social protection • child-sensitive • vulnerability • social differentiation • cash transfers
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
A child-sensitive approach to social
protection: serving practical and
strategic needs
Keetie Roelen and Rachel Sabates-Wheeler
Social protection is increasingly considered a development success story. At the same time, it
still does too little to account for social differentiation and to address vulnerability, as opposed
to poverty. Child sensitive social protection has gained considerable momentum, particularly in
a developing country context, raising questions about its concept and practical implications. We
argue that three types of vulnerabilities call for more tailored thinking about social protection for
children and discuss implications for social protection interventions on the basis of case studies.
Child sensitive social protection requires a critical perspective and for context to guide its design
and delivery.
Introduction
The last two decades have seen increased acknowledgement of the importance of
issues surrounding child poverty, vulnerability and well-being and the need for a
special focus within the development and poverty reduction debate to address these
issues (Roelen, 2010; Ben-Arieh, 2000). Child poverty is widely recognised to have
far-reaching short-term and long-term adverse impacts on income, education, health
and other areas of well-being (see, for example, Haveman and Wolfe, 1995). With
social protection having established itself as a core function of development policy
(Devereux et al, 2011), it is increasingly being considered as an integral part of the
response to child poverty and vulnerability in low- and middle-income countries.
Child-sensitive social protection (CSSP) is the term used to summarise a wide range
of policies and programmes. What, however, does CSSP actually encapsulate? What
does the concept mean with respect to instruments and interventions? The interest
in CSSP and the related body of research on social protection and its potential
benets has grown substantially in recent years but lacks a shared understanding of
the meaning of the term. Guiding principles have been formulated by UNICEF
and other development partners (UNICEF, 2009) to respond to the great interest in
CSSP; but these err on the general side, and fail to provide detailed insight into what
CSSP should or could look like.
The purpose of this paper is, rst, to take stock of and evaluate the current agenda
on CSSP. We recognise the importance of bringing a child-centred focus to the
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
292 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
established social protection eld of policy and research. However, we argue that the
idea of CSSP is not to propose a new set of measures or interventions but rather to
act as a tool to assess such interventions against the extent to which they respond to
children’s practical and strategic needs. Second, we identify three types of vulnerability
that are pertinent to, or exacerbated for, children and require particular consideration
within the remit of social protection. These are: 1) physical/biological vulnerabilities;
2) dependence-related vulnerabilities; and 3) institutionalised disadvantage (see also
Sabates-Wheeler and Roelen, 2011). CSSP means that social protection must respond
to these vulnerabilities by building in safeguards and instruments that minimise and
counteract their impacts on children’s lives. Building on these elements of a child-
sensitive approach to social protection, we consider the degree of ‘child-sensitivity’
of current social protection measures, thus necessarily thinking beyond commonly
held assertions about what is good or bad policy for children. In this way, this paper
challenges some of the policy interventions and rhetoric that are widely acclaimed to
benet children and also considers those that have not been intentionally designed to
benet children but can actually have a positive impact on their lives.
Making the case for being child-sensitive
Recent decades have seen an increased recognition that children deserve a special
focus within the debate on poverty reduction, both in developed and developing
countries. There are a number of grounds on which one can argue for a special child
focus within policies and programming, which hold across time and space.
First, it is now widely acknowledged that children have dierent basic needs from
adults and are harder hit, both in the short and long term, when their basic needs
are not met. Jones and Sumner (2011) point towards the ‘dierential experience’ of
poverty in childhood, setting their situation apart from adults (as well as from other
children, depending on their life-stage). Children growing up in a poor or low-
income family are more likely to receive poorer healthcare, to have lower educational
outcomes and to reach lower levels of attainment in the labour market (Haveman
and Wolfe, 1995; Esping-Andersen and Sarasa, 2002). Children living in poverty are
also more likely to grow up to become poor adults (Corak, 2006). Furthermore,
eects are more pronounced for those children who experience persistent poverty,
that is, living in poor and vulnerable conditions for a number of consecutive years
(Brooks-Gunn and Duncan, 1997; Duncan and Brooks-Gunn, 1997).
The case for child-sensitive policies can also be made by appealing to rights- and
eciency-based arguments (Blank et al, 2011). Rights-based approaches have been
at the heart of policy discourse around child poverty and have informed the response
to it by organizations such as UNICEF and many NGOs (Jones and Sumner, 2011).
Rights-based approaches resonate with the widely ratied Convention on the
Rights of the Child (CRC) and build on the premise that poverty in itself is a
violation of human rights. In addition, Blank et al (2011) argue that social protection
in itself is a basic human right, as the 1948 Universal Declaration of Human Rights
(UDHR) as well as the CRC refer to social protection as an entitlement rather
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
293
Keetie Roelen and Rachel Sabates-Wheeler
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
than charity. Although these rights-based arguments are not particular to children
and hold for all humans, the argument is reinforced by moral obligations related to
children’s ‘innocence’ and dependence on others for the provision of basic needs.
Eciency grounds for social protection emphasise its potential for the stimulation of
productivity and economic growth (Blank et al, 2011), considering social protection
as an investment rather than a mere welfare or protective measure. Alderman and
Behrman (2006) estimate the economic gain at about $510 for each infant that
is moved out of low birth weight status, largely stemming from increased labour
productivity and reduction of the costs incurred by infant illness and death. There
is ample research pointing to the high returns from investing in children with
broad social and economic gains, underlining the need to give children special
consideration in policy processes to respond to their unique position in terms of
needs and vulnerabilities as well as opportunities.
The remit of child-sensitive social protection
The rise of social protection as a central element in development policy has been
accompanied by deserved criticism that popular instruments and programmes have
not paid adequate attention to social dierentiation and the dierential opportunities
that social or age-specic groups have for access to the variety of forms of social
protection (Sabates-Wheeler and Roelen, 2011). Sabates-Wheeler and Feldman
(2011) explore the contribution of a consideration of migration and migrants to
social protection. Molyneux (2009) and Jones and Holmes (2010) bring a gender
critique to social protection. Similarly, organisations and researchers with a mandate
to advance children’s wellbeing, particularly in developing countries, have increasingly
advocated for child-sensitive social protection (CSSP). UNICEF has pushed the
global agenda on CSSP, in terms of both the denition of the term and advocacy
for its goals. It has initiated the Joint Statement on CSSP in 2009, which calls for
consensus on a stronger focus on children in the social protection agenda, to which
many international organisations and NGOs (non-governmental organisations) have
signed up. So what is CSSP, and what does it add to a mainstream social protection
agenda?
Denitions of the concept of CSSP are thin on the ground, but primarily emphasise
the need for interventions addressing the particular vulnerabilities that children
face, as well as human capital investment, and make reference to target groups and
mechanisms. ‘Child-sensitive social protection focuses specically on addressing
the patterns of children’s poverty and vulnerability and recognizing the long-term
developmental benets of investing in children… In addition, interventions do not
have to target children directly to be child-sensitive’ (Yates et al, 2010: 210).
Temin (2008) refers to child-sensitive social protection as ‘the range of economic
and noneconomic social protection interventions that need to be strengthened if the
most vulnerable children and [their] families are to benet. These include (but are
not limited to) cash transfers, social work, early childhood development centres and
alternative care.’
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
294 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
In addition, Jones and Holmes (2010: 1) suggest that CSSP needs to be informed
by ‘an understanding of the multiple and often intersecting vulnerabilities and risks
that children and their care-givers face’; and that children’s experiences of such
vulnerabilities and risks dier depending on their life-stage. Despite the various
pointers provided in these denitions and guidelines, such as the complex and
interlinked set of risks, long-term adverse consequences of poverty and potential
interventions, we believe that they fail to provide a tangible idea of CSSP and its
practical implications. UNICEF’s guiding principles (UNICEF, 2009) go some
way in providing guidance for the meaning of child-sensitive social protection, but
fall short in terms of translating those conceptual guidelines into practical policy
implications. Although the document lists a number of steps towards making social
protection child-sensitive, these present widespread consensus rather than specic
pointers. Examples of such steps include an increase in available resources, an increase
in capacity and coordination and building the evidence base on CSSP. We do not
dispute the relevance of these steps, but consider them imperative for achieving any
kind of policy. Here we seek to elaborate a fuller understanding of the constituent
elements of child-sensitive social protection, evaluating current policy interventions
and their degree of ‘child sensitivity’. In other words, we consider the remit of
child-sensitive social protection to refer to outcomes rather than a set of inputs or
instruments.
Elements of a child-sensitive approach to social protection
We argue that there are three distinct sets of vulnerability and asymmetry pertinent
to children and call for more tailored thinking about social protection taking these
into account. These are: 1) physical/biological vulnerabilities; 2) dependence-related
vulnerabilities; and 3) institutionalised disadvantage (see Sabates-Wheeler and
Roelen, 2011 for a more detailed discussion).
The rst vulnerability refers to the fact that children at dierent ages are more
susceptible to the negative impacts of malnutrition or disease by virtue of their
immature immune systems and under-development. There is sound evidence that
malnutrition, lack of healthcare and low levels of education during infancy and
childhood have far-reaching and long-lasting detrimental consequences (Haveman
and Wolfe, 1995; Brooks-Gunn and Duncan 1997), which aect not only the child
as an individual but also society as a whole (Esping-Andersen and Sarasa, 2002).
Clearly this vulnerability changes as children grow but, in general, the physiological
immaturity of children makes them more vulnerable than adults to negative
outcomes of shocks (other things being equal). On a positive note, childhood can be
considered as a unique window of opportunity in life for physical, cognitive and social
development, with strong returns to investment in terms of nutrition, healthcare and
education. Investments in micro-nutrients and food fortication are found to have
high returns, especially for children (Horton et al, 2008). Research in Guatemala
shows that higher pre-school cognitive ability is associated with higher secondary
school enrolment and achievement scores. Similarly, research in Zimbabwe nds that
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
295
Keetie Roelen and Rachel Sabates-Wheeler
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
improved nutritional status of pre-schoolers is associated with school entry at an early
age and higher numbers of completed grades (Alderman et al, 2006). These ndings
are underlined by studies in other contexts, all pointing towards the positive eect
of early cognitive development on schooling outcomes in later years (Grantham-
McGregor et al, 2007). Finally, early childhood development programmes can also be
considered a unique opportunity for investing in children’s cognitive development
with immediate as well as long-term cognitive eects (Behrman et al, 2000).
The second type of vulnerability refers to the fact that children are by necessity
dependent on adults for their wellbeing. Children are not supposed to be economic
agents in their own right and are therefore highly dependent on adult members of the
household, family or community for the distribution of resources in order to meet
their physical, emotional and social requirements. Again, a caveat is in order here, as
older children may be more economically and socially independent. Nevertheless,
on the whole children are subject to asymmetrical relationships of dependence and
have limited autonomy in their wellbeing decisions. As a result, relationships are
open to misuse and abuse, thereby reinforcing children’s vulnerability. Alternatively
(or in addition), a child’s carers may be mis-informed or may lack the resources
for his or her adequate care, which increases his or her vulnerability. To ensure that
children’s needs are met within that wider structure, policies and programmes should
keep a close eye on issues such as the intra-household distribution and allocation of
resources (White et al, 2003). As Sabates-Wheeler et al (2009) argue, the risk of intra-
household discrimination might be particularly pertinent when women have limited
access to independent livelihoods or when the distribution of power and resources
is highly inequitable. As a result, policies and programmes cannot simply rest on the
assumption that resources will be equally distributed or prioritised towards those
who are most in need (see UNICEF, 2009). Finally, it also has to be noted that some
of the most vulnerable and marginalised groups of children live completely outside
family and community structures and are self-reliant by necessity rather than by
choice. A failure to recognise and acknowledge these groups would lead to their
exclusion from policy eorts and exacerbate their disadvantaged position, pushing
them into further isolation.
Lastly, institutionalised disadvantage, or what some sociologists refer to as ‘cultural
devaluation disadvantage’ (Kabeer, 2005), refers to the devaluation of certain groups
in society based on perceptions of who they are. As Lister (2004) pointed out in
reference to the agency of people living in poverty, this is not only about their
behaviour but also about how those in power act in relation to them. So, for instance, if
society at large places little value on women or children, the vulnerabilities associated
with this type of disadvantage present themselves in the form of voicelessness of
these populations, lack of recognition, lack of representation and often entrenched
inequalities that can provide fertile grounds for the deliberate abuse and exclusion
of these groups. Group characteristics that commonly underlie cultural devaluation
include gender, ethnicity, religion and age, as they are thought to denote persons
of lesser worth following the dominant beliefs, perceptions and attitudes in a given
society (Kabeer, 2005). Rather than such an asymmetrical relationship with the rest
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
296 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
of society being a natural or inherent result of physiological conditions leading to a
lack of agency or limited autonomy, this type of disadvantage is a social and cultural
construct that is maintained and reinforced by society. Children can be considered
one of the groups in society which are culturally devalued in this way (Harper and
Jones, 2008), thereby reinforcing and perpetuating their already weakened socio-
cultural position resulting from the rst two vulnerabilities.
Although children have become increasingly visible on the development
agenda and in debates around poverty reduction, they remain a highly invisible
population with a lack of voice who are hard to reach. It is important to note
here that the degree of invisibility cannot be considered as homogenous across all
children, and that particular characteristics or conditions, including disability, HIV
infection, orphanhood or living on the streets, might add to and aggravate children’s
voicelessness. The widespread lack of institutional visibility is further underlined by
the fact that children’s issues are by and large represented by the weakest ministries in
government. As Jones and Sumner (2011) point out, ‘Many developing countries lack
a dedicated children’s ministry’; and, where they do exist, ‘they are typically among
the least inuential and under-resourced’ (p. 67). A response to the strategic need for
children to be more visible thus does not only pertain to a stronger involvement of
children themselves in policy processes but also a strengthening of the institutions
that speak on their behalf.
The implications of these three vulnerabilities suggest that an appropriate social
protection response for children requires particular elements that constitute the
degree of child-sensitivity of the intervention. Child-sensitive social protection
interventions should cater for both the practical and the strategic needs of children,
their carers and the community. Moser (1989) describes practical [gender] needs as
‘those needs which are formulated from the concrete conditions women experience’,
while strategic gender needs are ‘those needs that are formulated from the analysis
of their subordination to men’ (p. 1803). We apply these concepts to the needs of
children, leading to the interpretation that children’s practical needs are apparent from
the concrete conditions they experience given their stage in life, while children’s
strategic needs are observed from an understanding of their limited autonomy
and their relative invisibility within the population at large. Both needs should be
addressed by social protection to ensure that policies and programming are indeed
child-sensitive. For example, while a conditional cash or food transfer may ensure
that particular nutritional and health (that is, practical) needs of children are met, it
may also require birth registration or proof of identication (that is, strategic needs)
for children to actually access that transfer. As such, the rst vulnerability referring to
children’s particular biological and physical conditions can be largely translated into
practical needs that need to be met, while the (natural and constructed) asymmetrical
relationships that come into play in the second and third areas of child vulnerability
translate instead into strategic needs.
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
297
Keetie Roelen and Rachel Sabates-Wheeler
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
Child-sensitive social protection in practice
Although it is useful to consider the extent to which theory can guide and support
the notion of CSSP, it is also important to review how the notion of CSSP can be
applied to social protection programming in practice. In this section, we discuss a
number of social protection interventions and consider the extent to which they can
be considered child-sensitive when assessed against children’s practical and strategic
needs. This illustrates the use of the notion of CSSP as an assessment tool rather than
a programmatic response per se. Interventions discussed here include those that have
been claimed to benet children, including conditional cash transfers and targeted
orphan benets, as well as those that are not widely considered to be particularly
child-sensitive, such as old age pensions.
Conditional cash transfers (CCTs)
The role of cash transfers in development and the recognition of their potential for
stabilising income, reducing income poverty and food insecurity and increasing levels
of well-being have gained unprecedented momentum in recent years (Adato and
Bassett, 2008; Hanlon et al, 2010). Especially in relation to children, CCTs are now
considered an important programme for their role in human capital development
by linking the receipt of transfers to school enrolment and/or health check-ups
(Fiszbein and Schady, 2009). This type of social protection intervention can thus
be considered to meet the practical needs of children by responding to children’s
multiple physiological and biological requirements. Where services are available,
conditional transfers are considered a particularly appropriate tool for reaching
the most vulnerable children and increasing their level of well-being and access to
services (de Janvry and Sadoulet, 2006; Fiszbein and Schady, 2009). An increasing
body of evidence from CCTs in such contexts is now available, primarily in Latin
America, and conrms those positive impacts (for example, Das et al, 2005; Fiszbein
and Schady, 2009).
Despite the overwhelming positive evidence of the impact of CCTs, however,
there is also evidence of less preferential outcomes for children. Evidence suggests
that CCTs do not unequivocally respond to children’s physical and biological
situations and meet their practical needs. A study of Brazil’s Bolsa Alimentação
nds that children receiving benets conditional on regular contact with the health
system have a lower weight than those receiving unconditional benets. Instead of
promoting improved nutrition practices for children, these ndings suggest that the
programme provides an incentive for keeping children at low weight to ensure the
receipt of benets (Morris et al, 2004). In Colombia, Ospina (2010) nds that CCTs
can act as insurance for the schooling of poor children but are not able to prevent
child labour as a coping strategy. A study by Barrera-Osorio et al (2008) suggests that
CCTs may even increase labour market work, while decreasing school attendance
for siblings or peers living in the same household but not enrolled in the programme.
It shows that although the peer eects of CCTs at school or community level are
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
298 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
largely positive (see also Lalive and Cattaneo, 2009), such results cannot be used to
draw inferences about spillover eects at the household level. Children’s voicelessness
and lack of autonomy limit their ability to counteract decisions made on their
behalf, underlining their strategic needs for more equitable power relationships.
These examples show that, rather than meeting children’s strategic needs, CCTs
can perpetuate and reinforce children’s dependent and marginalised position.
More generally, the imposition of conditionalities raises concerns with respect to
paternalism and people’s freedom to choose appropriate services (Adato and Bassett,
2008; Standing, 2011) and various scholars have argued that poor people know best
how to allocate their resources (Hanlon et al, 2010).
In sum, the claim that CCTs are a particularly child-sensitive instrument is far
from evident. Although the theory of CCTs has the potential to respond to those
vulnerabilities deemed crucial for children and thereby to address their practical and
strategic needs, their potential to positively aect children’s lives is highly context-
specic. In itself, the intervention is well-equipped to respond to the multidimensional
nature and age- (and gender-) specicity of child poverty. Nevertheless, the degree
of child-sensitivity should be considered on a case-by-case basis and issues related
to the transfer of power and creation of perverse incentives should be monitored
closely. This should include a concern for the eect of CCTs on children who are
not enrolled in the programme, comparing their situation at community, group and
household level. Finally, as Lund (2011) argues with respect to the Child Support
Grant (CSG) in South Africa, the design of social protection interventions in favour
of conditionality should not be misdirected by shifts towards more conservative
approaches and the rising popularity of the notion of ‘co-responsibilities’.
Child-targeted programmes
Categorical targeting of social protection might be a preferred option in resource-
constrained contexts (Coady et al, 2003). Findings by Coady et al (2004) suggest that
the targeting performance of programmes targeted towards children is better in terms
of transferring benets to those in poverty when compared to targeting towards
elderly people. Child-targeted programmes are also found to lead to improved
outcomes for children (see Samson et al, 2011). In a context of growing numbers of
orphans due to HIV, and against a backdrop of eroding traditional care and support
structures, cash transfers are increasingly being considered as an intervention to help
families to cope with increased dependence rates and safeguard the development
of orphans (Stewart and Handa, 2011). In addition, they are increasingly being
considered as a mechanism to incentivise foster care, particularly in Sub-Saharan
Africa (Roelen and Delap, 2012). The use of cash transfers for such purposes would
imply explicit targeting of children whose parent(s) have died or of other vulnerable
children.
In the light of children’s specic vulnerabilities, the availability of cash or in-kind
transfers targeted to orphans and/or other vulnerable children has the potential to
meet both practical and strategic needs. The rst objective of such transfer schemes
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
299
Keetie Roelen and Rachel Sabates-Wheeler
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
is to provide for children’s basic and immediate requirements such as nutrition and
shelter, thereby fullling physical and biological requirements. Findings from Kenya’s
OVC Cash Transfer Programme indeed suggest that orphan-targeted transfers lead
to increased expenditures on food and health (Kenya OVC-CT Evaluation Team,
2012a). Impact evaluations of the same programme also show that the transfers
improve school enrolment, particularly in secondary school (Kenya OVC-CT
Evaluation Team, 2012b). Although narrowly targeted cash transfer programmes
may thus meet short- and long-term practical needs, they are likely to undermine
strategic needs, however. Strategically, the incentivisation of foster care strongly
recognises children’s dependence on adults to full those basic needs. Nevertheless,
in consideration of the distinct disadvantaged position of children whose parent(s)
have died, particularly orphans, the question arises as to whether interventions
should be particularly targeted towards these groups or rely on broader targeting
of poor and vulnerable children (Subbarao et al, 2001). With reference to children’s
dependent positions and asymmetric power relationships in relation to adults, the
implementation of a transfer payment to accompany care for orphans might lead to
‘commodication’ of children and undermine rather than meet this strategic need.
Such an eect is likely to be stronger if other transfer programmes are not available,
more dicult to access, or oer lower benets. Furthermore, the narrow targeting
of orphans might give rise to issues of stigmatization and equity in terms of the
exclusion of other groups of poor and vulnerable children and perpetuate their lack
of voice in society by further ‘cultural devaluation’.
The Orphan Care Programme (OCP) in Botswana presents an illustration of
the way in which a targeted programme fails to meet its objectives, despite good
intentions. Botswana’s current orphan care programme is narrowly targeted towards
children who have lost one or both parents and provides a response to their material
needs in terms of food vouchers and provision of clothing (Ellis et al, 2010). Eligibility
for the programme is exclusively based on whether or not the child is registered as
an orphan, or with one parent who has died, regardless of living conditions or other
criteria. Other social protection interventions in Botswana include programmes
for the destitute and for those who live in remote areas and are generally a lot
more stringent in terms of eligibility, applying means testing or proxy means testing.
Roelen et al (2011) nd that these two very dierent approaches towards targeting
of benets have led to a number of perverse incentives, exacerbating orphans’
vulnerable positions rather than alleviating them. The discrepancy in targeting
mechanisms between the orphan care and other programmes has led some orphans
to be considered ‘assets’ and a gateway for foster families to access food vouchers
and other benets. In other words, rather than the programme giving children more
voice and a degree of autonomy, it builds on and reinforces asymmetrical power
relationships. Although clearly recognizing the dependent position of orphans, the
orphan programme seems to provide perverse incentives and in some cases even to
‘commodify’ children.
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
300 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
Old age pensions
Old age or social pensions have been introduced in a number of Sub-Saharan
African countries and they are under consideration in other developing countries
(Holmqvist, 2010). Despite elderly people being the prime target group for old
age pensions, the potential positive eects of such pensions for other demographic
groups are widely recognized. Especially in contexts where there are extended
family and multi-generational households, old age pensions do have the potential
to reach not only the direct recipients but also working-age adults and children
(Bertrand et al, 2003; Ardington et al, 2009). As Kakwani et al (2006) point out, the
proportions of children living in poverty are much higher among those children
living in households headed by elderly people. Old age pensions are thus likely
to reach children in poverty and thereby make a considerable contribution to the
reduction of child poverty. For instance, in Namibia, an estimated 28% of social
pension money is spent by pensioners on themselves, but 78% is spent on other
individuals or on the household (for example, on buying food and groceries for the
whole household) (Devereux, 2000). It has to be noted, however, that the extent to
which intra-household distribution is to the benet of children cannot simply be
assumed and needs to be assessed on a case-by-case basis.
Considering the particular case of old age pensions in South Africa, the largest
social pension scheme in Sub-Saharan Africa, a number of studies have pointed to
their benecial side-eects for children. South Africa’s old age pension scheme was
put in place to support those who were out of the labour force due to age and
could not access a private pension (Bertrand et al, 2003). Edmonds (2005) nds that
the mere anticipation of a substantial and regular cash transfer to elderly people is
associated with higher school enrolment rates and reductions in the hours worked
by children. Actual receipt of an old age pension augments these eects and goes
hand-in-hand with an increase in educational attainment and primary school
completion rates. Duo (2003) nds similar positive associations between old age
pension and child outcomes, albeit with a considerable dierential in terms of their
gendered eect. Old age pensions in South Africa improved health and nutritional
outcomes for girls living in households with pension recipients. Pensions received
by women were found to particularly improve nutritional outcomes for young
girls. In this way, old age pensions respond to children’s practical needs in terms of
their physiological and biological requirements and the fullment of those by other
members in their household. Credit constraints are lifted and allow for access to and
provision of nutrition, healthcare and education. Furthermore, the intra-household
distribution of resources appears to benet children, with a substantial allocation
of resources directed to the improvement of children’s outcomes. Strategic needs,
however, are hardly addressed, as the old age pension policy measure does nothing to
lessen children’s institutional disadvantage by either increasing their own visibility or
strengthening the institutions that speak on their behalf.
Obviously, the positive outcomes of pensions for children have to be considered
within a larger context, bearing in mind that the pension benets also have an impact
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
301
Keetie Roelen and Rachel Sabates-Wheeler
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
on labour supply and the migration decisions of prime-age household members
(often the children’s parents), as well as fertility decisions. Nevertheless, evidence
does suggest that programmes not designed specically to meet the needs and
vulnerabilities of children do have the potential for a degree of child-sensitivity.
Discussion
This review of practical examples of social protection interventions shows that
context is of the utmost importance in terms of the claim that social protection can
alleviate child poverty and improve the lives of children. Although the three sets of
interventions described above all have the potential to meet children’s practical and/
or strategic needs, their particular context, design features and delivery methods can
undermine and perpetuate rather than alleviate those needs. Even the most obvious
example of an intervention that claims to be child-sensitive – child-targeted transfers
– does not unequivocally benet children. By the same token, social protection
measures designed to protect a dierent group in society, such as old age pensions,
have the potential to address practical needs of children and carry a degree of child-
sensitivity. As Duo (2003) pointed out, with respect to the old age pension scheme
in South Africa, it is dicult to generalize ndings across Sub-Saharan Africa or the
developing world as a whole, as the South African scheme is of such a nature and
scale that this makes it unique and dicult to replicate. Hence, for social protection
to address children’s physical and biological requirements and to be child-sensitive
(thereby addressing their dependent position and institutional voicelessness and
invisibility), it should be noted that context matters and that the degree of child-
sensitivity of social protection interventions cannot be generalised across place and
time.
Conclusion
In this paper, we argue that there is a need for a more concrete and clearer
understanding of child-sensitive social protection in order for social protection to
meet both the practical and the strategic needs of children. Such an understanding
is currently not provided by existing denitions and guidelines, such as UNICEF’s
Joint Statement, given their generalist and all-purpose nature. In this paper, we
elaborate on three particular elements of child vulnerability that social protection
should address in order to be labelled child-sensitive. These three elements pertain
to children’s practical needs, being their biological and physical requirements, as
well as their strategic needs, referring to children’s limited levels of autonomy and
their dependence on adults for receiving care and support, as well as to children’s
institutional invisibility and lack of voice in the larger policy agenda.
The practical examples, and their degree of ‘child-sensitivity’ in terms of their
response to children’s practical and strategic needs, show that no set of interventions
can be considered child-sensitive across the board. Claims about what makes social
protection child-sensitive and concurrent claims of child-sensitivity are often made
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
302 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
on (widely agreed) assumptions rather than sound and rm evidence about what
works for children in a particular situation. There are no universal truths about how
to design and deliver child-sensitive social protection and the degree to which a
policy, instrument or intervention meets children’s practical and strategic needs is
highly context-, design- and delivery-specic. Despite overwhelming evidence for
the positive eects of certain instruments on child well-being, we should not be
tempted to generalise and turn a blind eye to those contexts or groups of children
for which those ndings do not hold true. The term ‘child-sensitive’ should be used
with caution in order to avoid it becoming a misnomer and to ensure that it denotes
policy initiatives that truly aim to improve children’s lives. Although the examples
used in this paper are primarily from a developing country context, this message also
holds with respect to developed countries. The suggestion by Standing (2011) that
the expanding popularity of CCTs from developing countries to countries such as
the US and UK hinges on ‘libertarian paternalism’ rather than evidence of improved
outcomes makes a deeper and evidence-based understanding of CSSP very pertinent.
On a more positive note, children’s unique risks also make for unique opportunities.
The practical examples in this paper suggest that CSSP need not be a separate
stream or form of social protection. All types of interventions have the potential
to carry a degree of child-sensitivity, responding to one or more of children’s
practical and strategic needs. Further research is required to investigate the extent to
which measures and interventions can respond to all needs simultaneously without
undermining their initial objectives or exacerbating the needs and vulnerabilities of
other social and demographic groups.
References
Adato, M. and Bassett, L. (2008) ‘What is the potential of cash transfers to strengthen
families aected by HIV and AIDS? A review of the evidence on impacts and key
policy debates’, Joint Learning Initiative on Children and HIV/AIDS (JLICA).
Alderman, H. and Behrman, J. (2006) ‘Reducing the incidence of low birth weight in
low-income countries has substantial economic benets’, The World Bank Research
Observer, 21(1): 25–48.
Alderman, H., Hoddinott, J. and Kinsey, B. (2006) ‘Long term consequences of early
childhood malnutrition’, Oxford Economic Papers, (58)3: 450–74.
Ardington, C., Case, A. and Hosegood, V. (2009) ‘Labor supply responses to large
social transfers: longitudinal evidence from South Africa’, American Economic
Journal: Applied Economics, 1(1): 22–48.
Barrera-Osorio, F., Bertrand, M., Linden, L. and Perez-Calle, F. (2008) ‘CCT in
education’, Policy Research Working Paper 4580, Impact Evaluation Series No. 20,
New York: World Bank.
Behrman, J.R., Cheng, Y. and Todd, P. (2000) The impact of the Bolivian integrated
‘PIDI’ preschool program, Philadelphia: University of Pennsylvania.
Ben-Arieh, A. (2000) ‘Beyond welfare: measuring and monitoring the state of
children – new trends and domains’, Social Indicators Research, 52: 235–57.
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
303
Keetie Roelen and Rachel Sabates-Wheeler
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
Bertrand, M., Mullainathan, S. and Miller, D. (2003) ‘Public policy and extended
families: evidence from pensions in South Africa’, World Bank Economic Review,
17(1): 27–50.
Blank, L., Devereux, S. and Handa, S. (2011) ‘The case for social protection for
children’, in S. Handa, S. Devereux and D. Webb (eds) Social protection for Africa’s
children, Abingdon: Routledge: 15–28.
Brooks-Gunn, J. and Duncan, G. (1997) ‘The eects of poverty on children’, The
Future of Children, 7(2): 55–71.
Coady, D., Grosh, M., and Hoddinott, J. (2003) ‘Targeting outcomes redux’, The
World Bank Research Observer, 19(1): 61–85.
Coady, D., Grosh, M. and Hoddinott, J. (2004) Targeting of rransfers in developing
countries: Review of lessons and experiences, Washington DC: World Bank.
Corak, M. (2006) ‘Do poor children become poor adults? Lessons from a cross
country comparison of generational earnings mobility’, IZA Discussion Paper No.
1993, Bonn: Institute for the Study of Labor (IZA).
Das, J., Do, Q. and Özler, B. (2005) ‘Reassessing conditional cash transfer programs’,
The World Bank Research Observer, 20(1): 57–80.
de Janvry, A. and Sadoulet, E. (2006) ‘Making conditional cash transfer programs
more ecient: designing for maximum eect of the conditionality’, World Bank
Economic Review, 20(1): 1–29.
Devereux, S. (2000) Social safety nets for poverty alleviation in Southern Africa, ESCOR
Research Report R7017, London: Department for International Development.
Devereux, S. and Sabates-Wheeler, R. (2010) ‘Transformative social protection for
Africa’s children’, in S. Handa, S. Devereux, and D. Webb (eds) Social protection for
Africa’s children, Abingdon: Routledge: 216–38.
Devereux, S., McGregor, J.A. and Sabates-Wheeler, R. (2011) ‘Introduction: social
protection for social justice’, IDS Bulletin, 42(6): 1–9.
Duo, E. (2003) Grandmothers and granddaughters: old‐age pensions and
intrahousehold allocation in South Africa’, World Bank Economic Review, 17(1):
1–25.
Duncan, G. and Brooks-Gunn, J. (1997) Consequences of growing up poor, New York:
Russell Sage Foundation.
Edmonds, E. (2005) ‘Child labor and schooling responses to anticipated income in
South Africa’, Journal of Development Economics, 81(2): 386–414.
Ellis, F., Freeland, N., Ntseane, D., Seleka, T., Turner, S. and P. White (2010) Preparation
of a Social Development Policy – Phase I and Phase II, Gabarone: Government of
Botswana.
Esping-Andersen, G. and Sarasa, S. (2002) ‘The generational conict reconsidered’,
Journal of European Social Policy, 12(1): 5–21.
Fiszbein, A. and Schady, N. (2009) Conditional cash transfers: reducing present and future
poverty, A World Bank Policy Research Report, Washington: The World Bank.
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
304 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
Grantham-McGregor, S., Cheung, Y.B., Cueto, S., Glewwe, P., Richter, L., Strupp, B.
and the International Child Development Steering Group (2007) ‘Developmental
potential in the rst 5 years for children in developing countries’, The Lancet
369(9555): 60–70.
Hanlon, J., Barrientos, A. and Hulme, D. (2010) Just give money to the poor: The
development revolution from the global south, Sterling, VA: Kumarian Press.
Harper, C. and Jones, N. (2008) ‘Why are children’s rights invisible?’, Opinion 119,
London: Overseas Development Institute.
Haveman, R. and Wolfe, B. (1995) ‘The determinants of children’s attainments: a
review of methods and ndings’, Journal of Economic Literature, 33(4): 1829–78.
Holmes, R. and Jones, N. (2010) ‘Rethinking social protection using a gender lens’,
ODI Working Paper 320, London: Overseas Development Institute.
Holmqvist, G. (2010) ‘Fertility impact of social transfers in Sub-Saharan Africa: what
about pensions?’, BWPI Working Paper 119, Manchester: Brooks World Poverty
Institute.
Horton, S., Alderman, H. and Rivera, J. (2008) ‘Hunger and malnutrition’,
Copenhagen Consensus 2008 Malnutrition and Hunger Challenge Paper,
Copenhagen: Copenhagen Consensus Center.
Jones, N. and Holmes, R. (2010) ‘The politics of gender and social protection’, ODI
Brieng Paper 62, London: Overseas Development Institute.
Jones, N.A. and Sumner, A. (2011) Child poverty, evidence and policy: Mainstreaming
children in international development, Bristol: The Policy Press.
Kabeer, N. (2005) ‘Social exclusion: concepts, ndings and implications for the
MDGs’, Paper commissioned as background for the Social Exclusion Policy
Paper, London: Department for International Development.
Kakwani, N., Son, H. and Hinz, R. (2006) ‘Poverty, old-age and social pensions
in Kenya’, Working Paper 24, International Poverty Centre: United Nations
Development Programme, Brasilia.
Kenya OVC-CT Evaluation Team (2012a) ‘The impact of the Kenya Cash Transfer
Program for Orphans and Vulnerable Children on household spending’, Journal of
Development Eectiveness, 4(1): 9–37.
Kenya OVC-CT Evaluation Team (2012b) ‘The impact of Kenya’s Cash Transfer
for Orphans and Vulnerable Children on human capital’, Journal of Development
Eectiveness, 4(1): 38-49.
Rafael Lalive and M. Alejandra Cattaneo (2009) ‘Social interactions and schooling
decisions’, The Review of Economics and Statistics, MIT Press, vol 91(3): 457-77.
Lister, R. (2004) Poverty, Cambridge: Polity Press.
Lund, F. (2011) ‘A step in the wrong direction: linking the South Africa Child Support
Grant to school attendance’, Journal of Poverty and Social Justice, 19(1): 5–14.
Molyneux, M. (2009) ‘Two cheers for CCTs’, IDS Bulletin 38(3): 69–74.
Morris, S., Olinta, P., Floris, R., Nilson, E. and Figueiró, A. (2004) ‘Conditional
cash transfers are associated with a small reduction in the rate of weight gain of
preschool children in northeast Brazil’, Journal of Nutrition, 134(9): 2336–41.
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
305
Keetie Roelen and Rachel Sabates-Wheeler
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
Moser, C. (1989) ‘Gender planning in the Third World: meeting practical and strategic
gender needs’, World Development, 7(2): 1799–825.
Ospina, M.P. (2010) ‘The indirect eects of conditional cash transfer programs: an
empirical analysis of familias en accion’, unpublished PhD dissertation, Georgia
State University, http://digitalarchive.gsu.edu/econ_diss/67
Roelen, K. (2010) ‘Child poverty – what’s in a word?’, in Vandenhole, W., Vranken,
J. and De Boyser, K. (eds) Why care? Children’s rights and child poverty, Antwerp:
Intersentia Publishing: 113–30.
Roelen, K. and Delap, E. (2012) ‘Researching the links between social protection
and children’s care in Sub-Saharan Africa: a concept note’, Brighton: Centre for
Social Protection, Institute for Development Studies and Family for Every Child.
Roelen, K., Edstrom, J., Sabates-Wheeler, R. and Davies, M. (2011) ‘Child and
HIV sensitive social protection in Eastern and Southern Africa: lessons from the
Children and AIDS Regional Initiative (CARI)’, Nairobi: UNICEF Eastern and
Southern Africa Regional Oce.
Sabates-Wheeler, R. and Feldman, R. (2011) Migration and social protection: Claiming
social rights beyond borders, Palgrave Macmillan
Sabates-Wheeler, R. and Roelen, K. (2011) ‘Transformative Social protection
programming for children and their carers: a gender perspective’, Journal for Gender
and Development, 19(2): 179–94.
Sabates-Wheeler, R., Devereux, S. and Hodges, A. (2009) ‘Taking the long view:
what does a child focus add to social protection?’, IDS Bulletin, 40(1):109–19.
Samson, M., Heinrich, C., Kaniki, S. Regalia, F., MacQuene, K., Muzondo, T., van
Niekerk, I. and Williams, M. (2011) ‘Impacts of South Africa’s Child Support
Grant’, in S. Handa, S. Devereux and D. Webb (eds) Social protection for Africa’s
children, Abingdon: Routledge: 117–46.
Standing, G. (2011) ‘Behavioural conditionality: why the nudges must be stopped –
an opinion piece’, Journal of Poverty and Social Justice 19(1): 27–38.
Stewart, S. and Handa, S. (2011) ‘Reaching orphans and vulnerable children through
cash transfers in Sub-Saharan Africa: simulation results from alternative targeting
schemes’, in S. Handa, S. Devereux and D. Webb (eds) Social protection for Africa’s
children, Abingdon: Routledge: 31–56.
Subbarao, K., Mattimore, A. and Plangemann, K. (2001) ‘Social protection of Africa’s
orphans and other vulnerable children: issues and good practice program options’,
Africa Region Human Development Working Paper Series, Human Development
Sector, Africa Region, World Bank, Washington DC
Temin, M. (2008) ‘Expanding social protection for vulnerable children and families:
learning from an institutional perspective’, Prepared by the Interagency Task Team
(IATT) on Children and HIV/AIDS: Working Group on Social Protection, New
York: UNICEF.
UNICEF (2009) Joint Statement on Advancing Child-sensitive Social Protection, DFID,
HelpAge International, Hope & Homes for Children, Institute of Development
Studies, International Labour Organization, Overseas Development Institute, Save
the Children UK, UNDP, UNICEF, the World Bank and World Vision
Delivered by Ingenta to: University of Surrey
IP : 2.121.89.33 On: Mon, 05 Nov 2012 17:04:46
Copyright The Policy Press
306 A child-sensitive approach to social protection
Journal of Poverty and Social Justice • vol 20 • no 3 • 2012 • 291-306 • http://dx.doi.org/10.1332/175982712X657118
White, H., Leavy, J. and Masters, A. (2003) ‘Comparative perspectives on child poverty:
a review of poverty measures’, Journal of Human Development, 4(3): 379–96.
Yates, R., Chandan, U. and Lim Ah Ken, P. (2010) ‘Child-sensitive social protection: a
new approach to programming for children aected by HIV and AIDS’, Vulnerable
Children and Youth Studies, 5(3): 208–16.
Keetie Roelen
k.roelen@ids.ac.uk
Rachel Sabates-Wheeler
r.sabates-wheeler@ids.ac.uk
Institute of Development Studies, University of Sussex, UK
