Protection + prevention + promotion = policy paralysis

 Protection + prevention + promotion = policy paralysis[1]

For the past two decades, the most commonly used conceptual framework for social protection has been the one proposed by Devereux and Sabates-Wheeler[2]. This builds from a proposal made by Guhan in 1994[3], that social protection comprises different measures that serve a variety of functions within the broader reduction of poverty:

·       Promotional measures which aim to improve real incomes and capabilities;

·       Preventive measures which directly seek to avert deprivation in specific ways; and

·       Protective measures that are even more specific in their objective of guaranteeing relief from deprivation.

Devereux and Sabates-Wheeler[4] introduced a fourth function, that of transformation through social justice. So far so good (except for the fact that it would have been much neater if the fourth dimension had also begun with “pro-” or “pre-”!). But they go on to state that “The full range of social protection interventions can be categorised under protective, preventive, promotive and transformative measures.” And they proceed to attempt such categorization:

·       Protective measures include social assistance for the chronically poor and typically include targeted resource transfers – disability benefits, single-parent allowances and social pensions for the elderly poor. Other protective measures can be classified as social services for the poor and groups needing special care, including orphanages and reception centres for abandoned children, feeding camps and the provision of services for refugees and Internally Displaced Persons (IDPs), and the abolition of health and education charges.

·       Preventive measures deal directly with poverty alleviation. They include social insurance for economically vulnerable groups: both formalised systems of pensions, health insurance, maternity benefits and unemployment benefits, and informal mechanisms, such as savings clubs and funeral societies.

·       Promotive measures aim to enhance real incomes and capabilities, which is achieved through a range of livelihood-enhancing programmes targeted at households and individuals, such as microfinance and school feeding.

·       Transformative measures seek to address concerns of social equity and exclusion, such as collective action for workers’ rights, or upholding human rights for minority ethnic groups. Transformative interventions include changes to the regulatory framework to protect socially vulnerable groups as well as sensitisation campaigns.

The 2004 paper was ground-breaking, particularly in highlighting the inadequacy of the preceding “Social Risk Management” framework and the precariously threadbare nature of “safety nets”. It put the “social” back in “social protection”. The Transformative Social Protection (TSP) framework, as it came to be known, rapidly became the must-have structure that underpinned multiple national social protection policies and strategies: e.g. Malawi (2009), Sierra Leone (2009), Mauritania (2011), Kenya (2012), Chad (2013), Ethiopia (2014), Liberia (2014), Zambia (2014), the Gambia (2015) and Zimbabwe (2016) spring to mind.

Yet this paper argues that, from a policy-setting perspective, the TSP framework has inherent flaws; and it has also become less relevant over the twenty years that have elapsed since its publication.

First, the inherent flaws. If it were possible to map different social protection instruments, or broad types of instruments, to the four functions outlined above, that might be helpful. If “social assistance”, say, always served only a protective function, “social insurance” was just preventive, and “social justice” alone delivered transformation, then it would be clear which set of instruments was suited to deliver each required outcome. But the reality is far more complex.

Even in the original paper the authors recognize that such categorization is often unsatisfactory and ambiguous. It is dotted with provisos to the effect that: “some measures could simultaneously ‘promote’ incomes as well as ‘prevent’ deprivation”; “most preventive mechanisms could be argued to have promotive effects”; “some social protection instruments…can be both promotive and transformative”; “some preventive mechanisms can be transformative, and vice versa”; “many interventions that can be considered as social protection measures have more than one objective, and are therefore discussed as both protective and promotive…or as both promotive and transformative”; and so on. This ambiguity is one reason that the proposed TSP framework is less helpful operationally than it is conceptually.

Another is that it is more intuitive to academics than to beneficiaries. Ultimately the function of social protection depends on the situation and inclination of its recipients, rather than on scholarly theory. For some beneficiaries of a cash transfer, the function may be purely protective, to buy basic food for the household; for others it may be preventive, to participate in a funeral club or savings and loans group; for others it may be promotive, even transformative, to invest in a productive livelihood that may lead to independence and dignity. Similarly, school meals might be seen as protective (basic consumption), preventive (avoiding loss of concentration), promotive (improving nutrition and learning), or transformative (keeping girls in school to avoid pregnancy and early marriage). In each case the transfer is the same, but the function varies depending on the status and characteristics of the recipient. As a consequence, it is likely that one would receive an understandably bewildered response if one asked a beneficiary whether the transfer s/he has just received is protective, preventive, promotive or transformative.

So the problem faced by countries that have adopted the TSP framework for their policy is that it doesn’t help them with the selection of instruments with which to implement that policy. Nor does it help them to explain their social protection policy to their citizens, who have no real concept of whether they are in need of protection, prevention, promotion, transformation or some combination of these. Nor does it facilitate the monitoring of success: there are no practical metrics to measure the number of individuals who have been protected, prevented, promoted or transformed.

Second, compounding these intrinsic challenges, there have been substantial changes in the understanding of social protection since 2004. This paper argues that – in addition to the inherent conceptual weaknesses of TSP for policy purposes – there have been three fundamental shifts, which do not invalidate the framework, but which make it even less useful operationally in setting national social protection policy:

1)      Social protection is now seen less as a way of alleviating current poverty, and more as a mechanism to invest in human development by building resilience to a range of vulnerabilities.

2)      Those vulnerabilities tend to reflect the risks faced by an individual rather than a household, and social protection is increasingly seen as being founded on the rights of individuals.

3)      An individual faces changing vulnerabilities from birth to death, so the objectives of social protection also need to change through the lifecycle[5].

Let us consider each of these in turn, in the context of the TSP framework.

Social protection is not just about alleviating current poverty. The 2004 paper is explicit that “The key objective of social protection is to reduce the vulnerability of the poor”. And it circumscribes “three categories of people in need of social protection: 1) the chronically poor 2) the economically vulnerable 3) the socially marginalised”. The danger here is the implication that social protection is for “them”, not for “us”. Yet we all need social protection: we have all been babies, we have all wanted to benefit from an education, we all face the risk of unemployment, we may all acquire a disability, we hope we will all grow old.

Social protection should be based on the individual. The paper is also ambivalent about whether social protection is for the household or the individual – indeed social protection for households is referenced more frequently than social protection for individuals. It recognises that “it is important that social protection is delivered on the basis of claims being made by citizens with entitlements”, but it doesn’t resolve how social protection for households will ever be justiciable. The Universal Declaration of Human Rights (UDHR) should be the foundation for social protection, and it is framed around the rights of individuals.

Social protection should address lifecycle vulnerabilities. The focus on “reducing the vulnerability of the poor” obfuscates the important distinction between the very different vulnerabilities faced by individuals at different stages of the lifecycle, which in turn determine which objectives social protection should contribute to achieving. Social protection should link with other policy interventions to ensure that pregnant women and infants are well-nourished, that older children attend school, that adolescents gain necessary skills, that adults are protected from unemployment, exploitation and violence and that older persons and persons with disabilities participate fully and in dignity. In essence, social protection is about protecting “everyone” (as per the UDHR), not just the “poor and vulnerable”, because we are all vulnerable at different stages of our life.

This paper argues that, at least from a practical policy perspective, a lifecycle framework has become more useful than the TSP framework for framing national policy. The UDHR itself is structured explicitly on a lifecycle approach: its Article 25 stipulates “the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood”, and specifies that “motherhood and childhood are entitled to special care and assistance”. This is the basis for social security in high-income countries, but has often been ignored when implementing social protection in international development contexts. The lifecycle is a more intuitive basis because every citizen knows where s/he fits within it, and policymakers know clearly what they are trying to achieve, and therefore which types of intervention are most appropriate, at each particular lifecycle stage. It allows a much clearer focus on the very different vulnerabilities faced through the lifecycle, some examples of which are shown in the figure below.

Source: Author

The lifecycle framework facilitates the definition of primary and secondary objectives for each lifecycle stage, helps to identify coverage gaps once all existing interventions have been mapped, and permits the identification of key linkages, thereby making the scope of strategy development more manageable by focusing collaboration on just the key partners for that particular lifecycle group. It also makes it much easier to monitor and measure impacts, because implementers know what they are aiming to achieve, and can quantify improvements (reduction in maternal deaths or stunted infants, improved school attendance rates, employment rates, pension coverage, etc.).

The lifecycle approach also allows clearer differentiation within the so-called “cross-cutting themes” of social protection, such as disability, gender and nutrition[6]. Key objectives here, too, evolve through the lifecycle. So, for example, in the case of disability, the primary concerns for young children are around early detection, for older children around inclusive, quality education, for youth around appropriate skills, for adults around access to employment and compensation for the additional costs of equal participation. Or in the case of gender, the emphasis shifts through the lifecycle from equity at birth to equality of opportunity at school, to child protection, to avoidance of early marriage and pregnancy, to prevention of gender-based violence and sexual harassment, to compensation in older age for contributions to the care economy.

There are promising signs that this message is being heard and observed. We are witnessing a welcome shift in the framing of social protection policy.

Lesotho was one of the early explicit adopters of a lifecycle approach in its National Social Protection Strategy 2014/15-2018/19[7]:

Bangladesh’s 2015 National Social Security Strategy has at its core “programme consolidation along life cycle risks, with programmes for children, working age people – including specific focus on youth and vulnerable women – the elderly, and persons with disabilities”. It envisages a “transformation towards a lifecycle system by consolidating progress in a small number of priority schemes”, as depicted below.

Zambia adopted the following for its 2018 Integrated Framework for Basic Social Protection Programmes:

It has incorporated this vision into its new 2023 National Social Protection Policy, which “therefore, identifies five levels in the life cycle approach as: maternity, childhood, youth, adulthood and old age”.

Botswana’s 2021 National Social Protection Framework (and subsequent Implementation Plan) is similarly built around lifecycle pillars.

 Cambodia’s new 2023 National Social Protection Policy Framework has adopted a lifecycle structure:

Sierra Leone has shifted from a classic TSP framework in its 2011 Social Protection Strategy (note the somewhat arbitrary distribution of instruments in the purple ellipse, suggesting, for example, that “unconditional cash transfers” can only be protective), as shown in the figure below…

…to a lifecycle structure in its 2022 Social Protection Strategy:

 Similarly, Malawi’s 2009 National Social Support Policy was recognisably derived from TSP, with four key components: Provision of Welfare Support, Protection of Assets, Promotion through Productivity Enhancement, and Policy Linkages and Mainstreaming. Its 2023 draft National Social Protection Policy, by contrast, states that: “Another inclusion in the Policy is the life cycle approach to social protection, reflecting that individuals face different risks and vulnerabilities at different stages in life”. Its new draft 2024 National Social Protection Strategy is explicitly structured around the lifecycle.

These are all positive developments which demonstrate that – from a policy-setting perspective at least – it is time to move on from the TSP framework that has dominated the last two decades.

So my parting message to all you social protection practitioners out there is to get on your bikes and peddle the lifecycle wherever you can! Good luck!

---

[1] A version of this paper was first portrayed as an “arts-based” presentation at the Centre for Social Protection International Conference on “Reimagining social protection in a time of global uncertainty” (Institute of Development Studies, Brighton, 12-14 September 2023). The full version was then published on the Development Pathways website as a "Pathways Perspective" (Issue No 34, April 2024).

[2] Devereux, Stephen, and Rachel Sabates-Wheeler. 2004. Transformative Social Protection. IDS Working Paper 232. Brighton: Institute of Development Studies.

[3] Guhan, S. 1994. Social Security Options for Developing Countries, International Labour Review, 133 (1): 35-53.

[4] I would like to declare a serious conflict of interest. I have known Stephen and Rachel for even longer than their conceptual framework has been the gold standard for social protectionistas. I am fortunate to count them both as friends (at least until now!), and I have the most enormous respect for all the incisive contributions they have made to the theory and practice of social protection over the years. I just don’t think that their conceptual framework is the optimal one for operationalising social protection policy.

[5] I tend to prefer the term “life-course” because – unless you are Buddhist – life is not a “cycle”: it gets more and more alarmingly linear the older one gets!

[6] Otherwise, these tend to be “mainstreamed”, which is development-speak for “ignored”.

[7] Interestingly, in the revised second version of this Strategy, covering the period 2021 to 2031, Lesotho retained the lifecycle structure, but attempted to overlay it with the TSP framework. Whether this simplifies or complicates policy implementation remains to be seen!