Many intergovernmental organizations (IGOs) now place a high priority on universal social protection as a means for achieving sustainable development. Is this shift toward universal social protection just talk, or does it signify a more substantial emphasis on welfare within development policy? We present a theoretical framework for understanding discursive changes in global policy as rebranding, fads, trends, or paradigm shifts. We then conduct a comparative, semi-structured review of official language related to social protection used by six key IGOs (International Labour Organization, International Monetary Fund, United Nations Children’s Fund, United Nations Development Programme, World Bank, and World Health Organization) across five dimensions of social protection (labor market, health, family, housing, and education) before the introduction of the United Nations 2030 Agenda for Sustainable Development. Then, employing the framework, we analyze the findings of this review to determine the significance of the discursive shift toward universal social protection in the context of the 2030 Agenda. We document that, at present, universal social protection is an influential policy trend that has shaped how IGOs understand and act on social issues. These findings inform theoretical debates on the relationship between discursive and substantive policy change and contribute to a growing literature on transnational social protection. They also have implications for efforts across agencies and sectors to enhance social protection and achieve the Sustainable Development Goals.
There is an emerging consensus among intergovernmental organizations (IGOs) that strengthening social protection is critical for achieving sustainable development. In 2015, the International Labour Organization (ILO) and the World Bank announced a partnership in pursuit of “universal social protection” to achieve the Sustainable Development Goals (SDGs) and accompanying targets related to social protection (Box 1). Together, they define universal social protection as “the integrated set of policies designed to ensure income security and support to all people across the life cycle—paying particular attention to the poor and the vulnerable” (ILO and World Bank Group 2015a). This includes “adequate cash transfers for all who need it, especially: children; benefits/support for people of working age in case of maternity, disability, work injury or for those without jobs; and pensions for all older persons.” It can be achieved in multiple ways, including through “social insurance, tax-funded social benefits, social assistance services, public works programs and other schemes guaranteeing basic income security.”
|Goal 1 -||End poverty in all its forms everywhere. Target 1.3: Implement nationally appropriate social protection systems and measures for all, including floors, and by 2030 achieve substantial coverage of the poor and the vulnerable.|
|Goal 5 -||Achieve gender equality and empower all women and girls. Target 5.4: Recognize and value unpaid care and domestic work through the provision of public services, infrastructure and social protection policies and the promotion of shared responsibility within the household and the family as nationally appropriate.|
|Goal 10 -||Reduce inequality within and among countries. Target 10.4: Adopt policies, especially fiscal, wage and social protection policies, and progressively achieve greater equality.|
Twenty years ago, it would have been difficult to foresee the ILO and the World Bank partnering to enhance social protection (Deacon 2007, 2013a; Orenstein 2008; Voipio 2011). Traditionally, the ILO has championed a universal approach to enhancing social welfare in developing countries anchored in substantiating the rights of workers, largely through a variety of social insurance schemes (Deacon 2005, 2007). In contrast, the World Bank and its sister institution, the International Monetary Fund (IMF), have tended to argue for targeted, residual approaches to social programs deemed to be “pro-poor” (Stubbs and Kentikelenis 2018; Stubbs et al. 2017). In the 1980s, IMF lending policies often required borrowing countries to reduce social spending and to restructure welfare provision (Babb 2013; Babb and Kentikelenis 2018). To ameliorate the social consequences of these policies, the World Bank began promoting minimalist social safety nets for the poor (Babb and Chorev 2016:90; Kentikelenis 2017; World Bank 2001), in contrast to the universal approach favored by the ILO.
How, then, should the shift toward universal social protection be understood? Is it simply new rhetoric used to rebrand existing social programs focused on poverty alleviation? The most recent policy fad in social assistance implementation? A robust trend advancing social policy in a development context? Or does it signify a paradigm shift, or transformation (United Nations Research Institute for Social Development 2016), in global development policy?
To approach these questions, we first develop a theoretical framework for understanding discursive changes in global policy as either rebranding, fads, trends, or paradigm shifts. We argue that changes in policy discourse are significant, or indicative of substantive policy change, when they reflect a change in how key actors understand and act on a given policy problem.
Using this framework, we then determine whether the discursive shift toward universal social protection is best characterized as rebranding, a fad, a trend, or a paradigm shift in global policy. Our data come from a comparative, semi-structured review of official outputs related to social protection at the ILO, IMF, United Nations Children’s Fund (UNICEF), United Nations Development Programme (UNDP), World Bank, and World Health Organization (WHO) across five dimensions of social protection: labor market, health, family, housing, and education. Then, using our theoretical framework of discursive change in global policy, we evaluate the data gathered from these organizations on social protection to determine the nature of this discursive change across agencies and sectors at the global level.
Our findings suggest that the discursive shift toward universal social protection is best characterized as an influential policy trend that has shaped how IGOs understand and act on social issues. However, thus far, the move toward universal social protection does not appear to be paradigmatic, or associated with a shift in the norms guiding policymaking at the global level.
The article is structured as follows. In the next section, we present a theoretical framework for understanding discursive change in global policy. Then we summarize the methods for reviewing and analyzing the social protection activities of IGOs. The fourth section presents the findings of our analysis. We conclude by discussing the implications of these findings, both for our theoretical understanding of the relationship between changes in policy discourse and substantive policy change and also for enhancing social protection as part of broader efforts to achieve the SDGs.
DISCURSIVE CHANGE IN GLOBAL POLICY: REBRANDING, FAD, TREND, OR PARADIGM SHIFT?
Social scientists have long been interested in how policy norms emerge or change at the global level (Babb 2009; Dobbin, Simmons, and Garrett 2007; Drori and Meyer 2006; Kentikelenis and Babb 2019; Kentikelenis and Seabrooke 2017; Meyer 2000; Meyer et al. 1997). This line of research focuses on the related but distinct issues of what key actors say, mean, and do in the field of policymaking, which is critical for understanding how they approach and respond to social problems (Kentikelenis and Shriwise 2016:2). We argue that changes in policy discourse are substantial when they reflect a shift in how key actors understand and act on a given policy problem and present a unifying framework, where changes in policy discourse are best described as rebranding, fads, trends, or paradigm shifts. To do this, we draw from the discursive institutionalism, organizational sociology, and policy change literature (Figure 1).
First, policy ideas—the substantial building blocks of policies—are distinct from policy discourse (Schmidt 2008), as evidenced by the fact that key actors may not always say what they mean. Furthermore, what key actors say about a policy problem may be decoupled from what they do about it (Bromley and Powell 2012; Meyer and Rowan 1977). Decoupling may result from a lack of clarity on what key actors mean to do in the first place (Bromley and Powell 2012) or from conflict within an organization. Alternatively, decoupling may emerge as a longer-term strategy for managing a diverse range of interests—a phenomenon known as organized hypocrisy (Brunson 2002, 2007; Weaver 2008). Finally, as described by Hall (1993), the ideas and principles of key actors can be translated into policy in a number of ways, ranging from the recalibration of existing policy instruments (first-order change), to the introduction of new instruments (second-order change), to the instillation of new norms and ideas guiding the use of policy instruments altogether (third-order change, or a paradigm shift).
As a unifying framework, we suggest that changes in policy discourse are best described in one of four ways: as rebranding, a fad, a trend, or a paradigm shift (Figure 2). Rebranding, fads, and trends are akin to what Hall (1993:279) refers to as “normal policymaking,” which is characterized by first- and second-order change.
Rebranding refers to discursive change without substantive policy change. In other words, it describes instances where key actors change what they say but not what they mean or do about a policy problem. Rebranding may be used to (re)claim policies after a change in political leadership or to disassociate and revive a policy from a previous mishap or failure. For example, very recent publications by the World Bank, including the 2019 World Development Report, have advanced the model of progressive universalism for social protection. Borrowed from the public health literature (Jamison et al. 2013), progressive universalism is the idea that the expansion of social protection should “benefit the poor and vulnerable first” (Rutkowski 2018). However, as close observers quickly pointed out, these ideas amount to“an iteration of the much criticized narrowly-targeted assistance schemes that the international financial institutions have supported for years, in some cases in replacement of existing universal programmes” (International Trade Union Confederation 2018:3).
Policy fads have been described as “an emulation of form rather than of substance” (Rosenau 1968:308); that is, they represent a tactical response to the external environment (Butterfill 2002). Often, they lead to policies being mimicked across countries, with little attention to context or functionality (Andrews, Pritchett, and Woolcock 2017; Pritchett, Woolcock, and Andrews 2013). Policy fads may also take the form of new tools or instruments for addressing long-standing global policy problems. For example, donor efforts to enhance security and justice in the Solomon Islands in the early 2000s made the legal system look modern by improving jail and courthouse infrastructure, but “barely engage[d] in any substantive sense with the prevailing justice problems that most Solomon Islanders encounter most of the time” (Pritchett, Woolcock, and Andrews 2013:6).
A policy trend refers to instances where a change in policy discourse corresponds to a change in what key actors do and mean but in a way that aligns with, rather than challenges or subverts, dominant norms in global policymaking. Policy trends are distinct from policy fads in at least two ways. First, they tend to be longer-lasting forms of second-order change. Second, discursive shifts that correspond to how key actors understand and act on a given policy problem are critical indicators of purposive action. One example is the securitization of the HIV/AIDS crisis, which began to be discursively framed as a security concern as early as 1987, peaked in 2000 with UN Security Council Resolution 1308, was reaffirmed in 2011 with UN Security Council Resolution 1983, and continues to date (Shadyab, Hale, and Shaffer 2017). In this case of discursive change, “the desire to affect policy responses was central … a deliberate attempt to change the way in which the disease was thought about, leading to different possibilities for action” (McInnes and Rushton 2010:239). It is important to mind that a discursive policy trend is not necessarily an indicator of coherent behavior from key actors—there can be organized hypocrisy—nor that they will be aware of and have clearly articulated the rationales underpinning changes in how they act on and speak about policy problems (Butterfill 2002).
Last but not least, a paradigm shift occurs when a change in policy discourse corresponds to a change not only in what key actors do and mean but also in the dominant norms in a policy field. Building on the literature in political science and sociology, we understand paradigm shifts as “radical changes in the overarching terms of policy discourse” (Hall 1993:279), and we suggest that paradigm shifts at the national and global levels are distinct. Unlike at the national level, where there is a recognized government, IGOs compete to set standards and norms for governance at the global level (Halliday and Carruthers 2007, 2009). Therefore, paradigm shifts at the global level are not so much about rearranging “the organization and standard operating procedures of the policy process” (Hall 1993:281) but about instilling a new hierarchy of norms across the policy field, thereby shaping—and constraining—the field’s ideational terrain. Recent historical work suggests that radical efforts were made after World War II to make social security and “freedom from want” global priorities—but they did not come to fruition (Helleiner 2014). At present, these efforts are echoed in the United Nations (UN) 2030 Agenda for Sustainable Development. It has been argued that “the SDGs define a new paradigm for development, with ecological sustainability, human well-being, inclusive economy and ‘leaving no one behind’ at the centre” (United Nations Research Institute for Social Development 2016:213).
Furthermore, while we acknowledge that the cumulative impact of paradigm shifts at the global level depends on how global policies are implemented within national contexts (Babb 2013), we wish to emphasize that paradigm shifts at the global level are best understood not by the number of countries that adopt their policies but by the degree to which a new set of norms shapes the global context in which national policies are created and implemented.
Finally, the field of global development policy is increasingly diverse, fragmented, and multipolar, which may affect the degree to which paradigm shifts are likely to emerge in the first place (Babb and Chorev 2016). In this context, von Gliszczynski and Leisering (2016) suggest that a series of “micro-paradigms” or a variety of policy models may emerge, potentially widening the gap between second- and third-order change. In other words, global policy change, including paradigm shifts, tends to arise gradually from the exchange of policy ideas through discourse between key actors in the field over time.
Returning to the topic at hand, it is unclear whether universal social protection represents a significant change in policy, or just in rhetoric. Is universal social protection simply a rebranding of activities already undertaken by IGOs, the latest fad in poverty alleviation strategies, a substantial trend in social policymaking in the developing world, or possibly a paradigm shift in approaches to global development?
IGOs affect the scope of social protection provision and play an active, if not dominant, role in policymaking at the global level. In line with their mandates, each IGO has a unique set of power resources—monetary, technical, and ideational—that shape, and are also shaped by, the nature of the policy problems they seek to address. From this position, IGOs are particularly well placed to influence the discursive process by legitimating policy ideas and fixing meanings (Barnett and Finnemore 2004). As the glue between policy ideas and actions (Figure 1), policy discourse emerging from IGOs must be understood within its organizational and institutional context (Zittoun 2009).
To determine the significance of the discursive shift toward universal social protection, we conducted a comparative, semi-structured review of the mandate and official outputs related to social protection of six IGOs from 2005 to 2015. The semi-structured nature of the review enabled a comparative approach not only within but also across IGOs and policy areas, which is key to illuminating trends and establishing a comprehensive understanding of the social protection agenda across the global policy field before the introduction of universal social protection in 2015.
This section clarifies the case selection and the methods for data collection and analysis, including details of how distinctions were made between what key actors mean, say, and do.
We focus on policy discourse on social protection within and between six key actors: ILO, IMF, UNICEF, UNDP, World Bank, and WHO. In line with their mandates and unique role in supporting social protection provision, our justifications for selecting these IGOs are as follows.
The ILO has been the lead advocate and supporter of social protection systems dating back well into the twentieth century (Barrientos and Hulme 2009; Usui 1994). Its mandate provides the clearest link to social protection, as it was established to promote social justice and labor rights (ILO 2016a). The ILO also has a dedicated Department of Social Protection. As thoroughly illustrated by Deacon (2013a, 2013b, 2014), the ILO’s work on social protection culminated with the endorsement of its Social Protection Floors Recommendation by both the UN Chief Executive Board in 2008 and the International Labour Conference in 2012. It has also served a critical coordination function between the Social Protection Floors Initiative (SPF-I), a joint initiative under the UN Chief Executive Board (2018) and the Social Protection Interagency Coordination Board (SPIAC-B, established in response to a request by the Group of 20)—both of which emerged in the wake of the 2008 global financial crisis (Deacon 2013a).
The World Bank and the IMF were included given their central role in financing development policies. Both the World Bank and the IMF lend to countries in exchange for policy reforms, a practice known as conditionality (Stubbs et al. 2018), which directly and indirectly affects social protection policies and interventions as well as their resource base. In particular, the World Bank’s mandate to support economic development, which encompasses poverty reduction, shapes its engagement with social protection (World Bank 2016a). It also contributes directly to the design of social protection policies, which it has traditionally referred to as social safety nets (World Bank 2001). The IMF is concerned with providing financial assistance to countries experiencing macroeconomic problems, thereby affecting the design of a range of national policies (Reinsberg et al. 2018). In addition to any conditions it places on lending, IMF financial support often functions as a stamp of approval that yields additional foreign aid from donors (Stubbs, Kentikelenis, and King 2016), which in turn affects the funds available for social protection policies and programs.
UNDP and UNICEF were selected given their central role in social protection within the UN System. UNDP plays a key role in coordinating and setting goals for global development policy and implementation across UN agencies, framing the context in which social protection policies are formulated and implemented. The UN Development Group, chaired by the UNDP administrator, fulfills this coordinating function at the global level, and it includes representatives from specialized agencies with their own governing bodies, including the ILO and WHO. In addition to guiding global development policy, UNDP’s mandate includes a special focus on the least developed countries and on crisis situations, such as conflicts and natural disasters (UNDP 2015a).
UNICEF’s mandate to advance the rights of children encourages broader discussions of social protection away from basic income and the labor market alone (UNICEF 2016). In particular, UNICEF offers technical assistance to ensure that social protection policies are “child-sensitive,” with special emphasis on the role of nutrition in children’s health and development (UNICEF 2017b). In conjunction with its efforts to ensure child protection, UNICEF also recognizes the role that social protection plays in reducing the risk of children being exploited or harmed (UNICEF 2017a). Together, UNDP and UNICEF assist with social protection for some of the most vulnerable, with UNDP leading largely in situations of extreme poverty, often as a result of conflict and disasters, and UNICEF leading the expansion of social protection in relation to child protection and poverty reduction through the provision of rights-based services and interventions (Commission for Social Development 2001).
WHO is the sixth organization included in this analysis given the explicit focus on health in social protection floors and its role as a co-chair of the SPF-I. WHO’s mandate seeks to ensure “the attainment by all peoples of the highest possible level of health,” where health is defined as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity” (WHO 2006a). In particular, WHO is focused on a rights-based approach to health, where “the enjoyment of the highest attainable standard of health is one of the fundamental rights of every human being without distinction of race, religion, political belief, economic or social condition” (WHO 2006a), which includes addressing health determinants (Commission on the Social Determinants of Health 2008). WHO supports social protection primarily through the extension of universal health coverage (WHO 2012), as recognized by both the World Health Assembly and the UN General Assembly (2013).
Finally, while several other IGOs, including the Food and Agricultural Organization, UN Educational, Scientific and Cultural Organization, the Office of the UN High Commissioner for Refugees, and the UN Centre for Human Settlements, to name a few, have implicit links to social protection across sectors, these organizations were not included in this review given their relatively peripheral role in the social protection agenda. However, the results do mention these IGOs where the data gathered highlight their role and contribution in key sectors related to social protection.
Data Collection and Analysis
We chose 2005 to 2015 as the time period for data collection for three reasons. First, in the years leading up to and including 2005, Deacon (2005:19) identified signs that the discourse on welfare policies within IGOs was beginning to shift from one of “a targeted and means tested safety net future” to one where “a universal approach to social welfare provision is again being recognized.” Second, the consolidation of the ILO’s Social Security Department in 2005 created an institutional space in which to shape this (re-)emerging discourse, leading to many critical advances in the social protection agenda (Deacon 2013a:34, 2013b). Third, von Gliszczynski and Leisering (2016:330) also cite 2005 as the year in which social cash transfers were recognized as a fiscally feasible instrument for implementing social protection policies.
Figure 1 served as the analytical frame guiding data collection and analysis for each IGO. What IGOs mean by social protection was approximated by definitions and policy positions articulated in legal instruments, such as conventions, protocols, the adoption of recommendations, and resolutions, as well as in policy frameworks focused specifically on social protection. What IGOs do about social protection was approximated by documents such as financial frameworks, annual reports, interagency communications outlining the tools and instruments used to support the design and implementation of social protection policies, the exercise of convening power (particularly in the case of the ILO), and descriptions of projects and activities meant to enhance social protection. What IGOs say, or the policy discourse on social protection, while present in the previous two dimensions as well, was captured predominantly in the speeches, remarks, and joint statements by high-level officials as well as press releases on the topic of social protection.
To capture the policy discourse as well as the key ideas and actions related to social protection across these IGOs, we collected data using the search function on each organization’s website, searching for the terms “social protection,” “social protection speech,” “social protection press release,” “social protection factsheet,” and “social protection brief.” While search results depend on the technology behind each website, they also reveal what IGOs make publicly available on the topic of social protection, thereby providing information on the instances in which they cloak certain ideas or actions within the discourse on social protection, and when they do not.
Once retrieved, each item was reviewed for definitions, understandings, and actions taken in relation to social protection. Then, each item was classified under one of five policy areas in line with the definition of transnational social protection provided by Levitt et al. (2017): labor market1, health, family, housing, and education. A sixth category of “general” was also included for instances where social protection was discussed either without mention of a specific policy area or in relation to more than one of these policy areas. Finally, a seventh category of “other” was included where social protection was discussed repeatedly in relation to policy areas not included in the definition provided by Levitt et al. The search results were also vetted for geographic specificity; while references to social protection in regional and country contexts were reviewed, references to social protection on a global scale were privileged in the analysis, given the focus on policymaking at the global level.
While the analysis aims to assess qualitative changes in how key actors understand and act on policy problems related to social protection, rather than to determine the frequency with which the term “social protection” is used, a total of 912 webpages and documents were collected and reviewed. Table 1 provides a breakdown by sector.
|.||ILO .||IMF .||UNICEF .||UNDP .||World Bank .||WHO .|
|.||ILO .||IMF .||UNICEF .||UNDP .||World Bank .||WHO .|
To put the total number of items collected in perspective, the IMF search revealed no instances in which the term “social protection” was part of the title of a webpage or document in the “general” category during the time period examined, while the ILO, followed by UNICEF and the World Bank, returned hundreds of webpages or documents explicitly focused on social protection. UNDP, followed more distantly by WHO, fell between these extremes.
Furthermore, definitions of each of the policy areas derived from the literature on transnational social protection were tailored and standardized throughout the review, which led to some fairly predictable differentiation. The ILO, followed by the IMF, returned the most items related to the labor market2; the WHO returned the largest number on social protection in relation to health3; and UNICEF, surpassed just barely by the ILO, had a large number in the area of family4. None of these organizations returned a substantial number of items on the relationship between social protection and education or social protection and housing; however, the relationship between these policy areas and social protection was covered in larger, more comprehensive documents discussing the social protection policies of these key actors across a number of policy areas and sectors.
This initial data collection also discovered key documents used to anchor understandings of what these IGOs mean by social protection, as summarized in Table 2.
|ILO||Social Protection Floors Recommendation, 2012 (No. 202) (ILO 2012); World Social Protection Report, 2014–2015 (ILO 2014b)|
|UNICEF||Social Protection Strategic Framework (UNICEF 2012)|
|World Bank||Social Protection and Labor Strategy, 2012–2022 (World Bank 2012c)|
|WHO||Social Protection Floor for a Fair and Inclusive Globalization (Social Protection Floor Advisory Group 2011)|
|ILO||Social Protection Floors Recommendation, 2012 (No. 202) (ILO 2012); World Social Protection Report, 2014–2015 (ILO 2014b)|
|UNICEF||Social Protection Strategic Framework (UNICEF 2012)|
|World Bank||Social Protection and Labor Strategy, 2012–2022 (World Bank 2012c)|
|WHO||Social Protection Floor for a Fair and Inclusive Globalization (Social Protection Floor Advisory Group 2011)|
Table 3 summarizes the findings related to the inclusion of social protection in the financial frameworks, budgets, and annual reports of the six IGOs examined. These documents were used to anchor assessment of what IGOs do in relation to social protection. Frameworks guiding financial resource allocation were privileged; when such a framework was not found, annual reports were used in their place.5
|ILO||ILO program and budget (biennial)|
|World Bank||Annual reports|
|WHO||General program of work|
|ILO||ILO program and budget (biennial)|
|World Bank||Annual reports|
|WHO||General program of work|
Secondary literature and unofficial sources, such as Deacon (2013a), working papers, and blog posts by high-level officials and staff, were also reviewed for context, to include all key sources that should be considered, and in some cases, to aid in triangulating between sources. However, given the nature of the research design and its theoretical grounding, official language was given preference in the review.
Finally, we excluded sources such as UN development assistance frameworks and funding reports related to joint and pooled funds supporting social protection arrangements, on the grounds that: (1) we are more interested in discerning the unique positions of each IGO in relation both to other IGOs and to key policy areas of social protection, which joint endeavors and reporting may obscure, rather than illuminate; and (2) these reports represent a discursive and substantive synthesis unique to each country, which may or may not be an accurate way to assign meaning to policy discourse at the global level. The findings from the six IGOs were then reviewed and synthesized, first with regard to social protection generally and then with respect to each of the five policy areas related to social protection.
This semi-structured review is intended to serve as a first step in assessing the shift toward universal social protection, and it is important to recognize its limitations. First, in the absence of deeper qualitative work, it is not possible to make causal claims, or to explain why there has been a shift toward universal social protection. Second, this study relies heavily on the degree to which the published output of IGOs accurately reflects what these key actors mean, say, and do when it comes to social protection. It is of course possible that these organizations are engaged in activities that are unreported or under-reported in official language and publications. However, addressing this problem lies beyond the ambitions of this analysis.
How Key IGOs Understand and Act on Social Protection
The six IGOs differ in how they define and understand social protection as well as in the policy areas and the means through which they prefer to achieve it. Taken together, they define social protection mainly in terms of poverty alleviation, as opposed to the substantiation of rights or the need for comprehensive welfare provision more broadly, with only the ILO arguing for a slightly broader understanding of social protection. This effect can be explained, in large part, by the economic focus of the IMF and the explicit poverty reduction mandates of the World Bank and UNDP. Yet even with the ideational flexibility of a rights-based mandate, UNICEF also defines social protection in relation to poverty. Similarly, WHO has commented less on social protection directly and instead has focused on how social protection can contribute to financial protection as part of broader efforts to achieve universal health coverage (Reich et al. 2016; Thomson et al. 2016).
The ILO (2016g) currently understands social protection to include a range of universalistic policy interventions in the areas of income security and healthcare, albeit with a focus on social insurance and managing risks related to the labor market. The other IGOs focus more narrowly on poverty reduction in distinct ways. UNICEF (2012) defines social protection broadly as the public/private mix of policies and programs aimed at eradicating poverty and deprivation. It recognizes both social transfers (also referred to as cash transfers or social assistance) and social services as forms of social protection, and it does not require that social protection be delivered directly to children to be considered “child-sensitive” (UNICEF 2017b). UNDP (2015b) distinguishes between social transfers and social services, defining social protection as the policies and strategies that support basic income/social assistance programs, arguing separately that social services should be available to all. Interestingly, WHO has commented less on social protection directly. When WHO does discursively engage with social protection, it has focused predominantly on how financial risk protection, or “a reduction in the number of people facing financial hardship or impoverishment because of the need to pay directly for the services they receive” (Bayarsaikhan 2012), contributes to universal health coverage as well as to social protection.
The World Bank (2012c:xiii) defines social protection as achieving income security across the life cycle through either social assistance or social insurance mechanisms. While this definition recognizes a wider range of instruments that can be used to ensure basic income provision when compared to the UNDP’s, the poorest strata of the population rarely contribute to social insurance schemes and are far more likely to receive social assistance. Furthermore, the World Bank also targets social protection to the extreme poor (World Bank 2013a, 2015b), which again results in a greater focus on social assistance. The IMF (2015) views its role in social protection as ensuring that those most vulnerable are protected in times of economic change, primarily via social safety net programs targeting the poor.
Finally, joint coordinating mechanisms, including the SPF-I and the SPIAC-B, along with several joint statements, have led to an increase in discursive exchange and alignment related to social protection. In 2009, all of the IGOs examined here, with the exception of the IMF and the WHO, signed on to UNICEF’s joint statement on child-sensitive social protection (Department for International Development et al. 2009). More recently, the World Bank has released joint statements on social protection not only with the ILO but also with UNICEF, outlining common ground in their respective efforts to advance social protection (ILO and World Bank Group 2015b; UNICEF and World Bank 2013). Several costing tools and methods for assessing benefit incidence and potential impact have been created with interagency input (Drouin 2015; Yuster 2015). Together, this suggests that the discourse on social protection must be evaluated not only with respect to the exchange of ideas and development of technical assistance and interventions, but also with respect to the contours of the policy field as they change over time.
In sum, understandings of social protection appear to have narrowed in ideational and instrumental terms since the 2008 global financial crisis and the introduction of the Social Protection Floors Recommendation in 2012 when compared to the definition of social protection provided by the Social Protection Floor Advisory Group in 2011. The Advisory Group, which was convened by the ILO and WHO to provide conceptual and policy guidance during the formation of the Social Protection Floors Recommendation (ILO 2019), conceptualized social protection in broad terms, including both cash transfers and a wide range of services, in its definition of social protection floors as “basic income security, in the form of various social transfers (in cash or in kind), such as pensions for the elderly and persons with disabilities, child benefits, income support benefits and/or employment guarantees and services for the unemployed and working poor; [and] universal access to essential affordable social services in the areas of health, water and sanitation, education, food security, housing and others defined according to national priorities” (Social Protection Floor Advisory Group 2011:xxii). Since that time, views on the role of services in social protection provision appear to be mixed, particularly among IGOs with mandates focused on poverty alleviation. However, all IGOs examined here recognize cash transfers and poverty alleviation as part of social protection. As a result, policy discourse on social protection between these key actors appears to have narrowed.
How Key IGOs Understand and Act on Social Protection across Policy Areas
Conditions of employment and responses to unemployment are core components of social protection policies, and different organizations have deployed interventions to support and address these components in potentially conflicting ways. The ILO has long advocated employment protection, offering recommendations to member states regarding the optimal means of initiation and termination of individual or collective contracts (ILO 2016e). In contrast, the IMF generally favors “flexible” labor markets. Its conditionalities have included the termination of collective contracts, reductions in public service employment or wages, and reforms of dismissal procedures (Forster et al. 2019; Kentikelenis, Stubbs, and King 2016). This stems in part from IMF research showing a negative association between flexible labor markets and unemployment (Bernal-Verdugo, Furceri, and Guillaume 2012, 2013; Crivelli, Furceri, and Toujas-Bernaté 2012), although these findings have been challenged by the ILO (Aleksynska 2014). The ILO also has a range of programs for the elimination of child labor through policy reform, improved monitoring, and awareness raising (ILO 2016b, 2016c). As one of the largest employers, WHO has focused on how the health sector can lead by example by ensuring social protection for health workers, particularly those at risk of acquiring HIV and TB via occupational exposure (Shriwise and Stuckler 2015; WHO 2010c, 2014b, 2015).
Cash transfers are a relatively common component of social protection policy across the IGOs examined. UNDP, UNICEF, and the World Bank all support cash transfers as part of social protection strategies, but they prefer different implementation mechanisms. UNDP specifies that it prefers non-contributory social assistance transfers for poor households (UNDP 2012; UNDP and National Planning Commission 2011). Though not targeting children directly, UNICEF supports all forms of cash transfers to households, from social assistance to disability to pensions, to ensure basic income across the life course, as these policies have been shown to benefit children (Singizi 2012; UNICEF 2015). The World Bank supports both conditional and unconditional cash transfer programs for the most vulnerable, depending on the country, along with employment services for job seekers, technical education, and vocational training to help individuals connect to the labor market (World Bank 2010b, 2014b, 2014d, 2015b). It also supports cash transfers in exchange for participation in public works projects in a number of sub-Saharan African countries (World Bank 2011, 2012b, 2013c, 2014e).
Good health and access to health care are relevant to understandings of social protection across IGOs, but this manifests in different ways. While WHO (2012) recognizes a bidirectional relationship between social protection and health, more is known about how social protection can improve health (Commission on the Social Determinants of Health 2008) than about how better health can improve social protection (Shriwise and Stucker 2015). As definitions of social protection have narrowed to focus more on cash transfers than services, WHO appears to have rebranded social protection as “financial protection” when discussed in the context of universal health coverage. This has two key effects. First, unlike social protection floors, which recognize both cash transfers and basic health services as part of social protection, rebranding social protection as financial protection in the context of universal health coverage draws attention to how universal social protection has narrowed to focus predominantly on cash transfers and poverty relief. Second, in so doing, WHO clarifies where and how it engages with understandings of social protection across IGOs in relation to universal health coverage (predominantly in the area of health financing) while also reserving space to discuss social protection in broader terms. Services are a central component of WHO’s (2018) understanding of universal health coverage, and WHO’s rebranding of narrow definitions of social protection as “financial protection” could be interpreted as a strategic move in support of broader definitions of social protection that include service provision.
The ILO has also long recognized the importance of achieving universal health coverage (in addition to its advocacy of social protection floors). Since 1944, the ILO has had a clear agenda of promoting high-quality, universal health coverage for workers and their families and dependents (ILO 1944). The ILO supports countries in assessing gaps in health care coverage as well as in devising and implementing strategies to expand access to health coverage, which it often refers to as “social health protection” (ILO 2008).
Key IGOs besides WHO and the ILO focus on specific aspects of health and vulnerable groups. UNDP focuses on improving the health of the most vulnerable, such as pregnant women, and also of those in crisis situations, as in Sierra Leone during the Ebola outbreak (UNDP 2015c). UNICEF focuses primarily on the link between health and nutrition for child development (UNICEF 2012). Both UNDP and UNICEF recognize that, in some countries, as many as half of all households benefiting from social protection are AIDS-affected, and they support the development of HIV-sensitive social protection programs as promoted by UNAIDS (UNICEF 2010; UNDP 2014b). The World Bank (2014c) advocates a number of health financing mechanisms to increase health care access, including community-based health insurance, social health insurance, and fee exemptions. The IMF (2008a, 2015) also focuses on health care financing, albeit with an emphasis on fiscal sustainability and the cost-effectiveness of healthcare interventions. Finally, cash transfers are often made conditional on visits to health clinics and the receipt of health interventions such as vaccinations, but this is determined on a case-by-case basis in each country.
All of the IGOs reviewed here focus on families, but they take different approaches. First, the ILO has developed labor standards for workers with families, which are intended to ensure that workers do not face discrimination and that they are able to balance work and family responsibilities. To meet these standards, governments must develop a range of social protection policies, including the provision of community services (e.g., child care) and vocational training, as well as employment protection legislation (ILO 2016f). Second, the importance of child-sensitive social protection strategies has been recognized by UNICEF, UNDP, ILO, the World Bank, and others (Department for International Development et al. 2009). In particular, UNICEF supports child grant programs, such as South Africa’s (Singizi 2012), and recognizes that social protection reduces risk factors that often lead to child harm through violence, exploitation, abuse, and neglect in unstable families and communities (Bloemen 2010). Finally, considering the relationship between social protection and gender also affects the advice of IGOs. For example, UNICEF (2012) promotes gender-sensitive social protection policies, particularly those to do with educational attainment, and the ILO (2016d) supports policies for maternity protection. WHO (n.d.a., n.d.b.) also provides advice on healthy aging and health over the life-course in relation to social protection, which requires action across sectors.
The search results suggest that the IGOs reviewed here focus little, if at all, on improving housing under the heading of social protection. The ILO (n.d.) is primarily concerned with workers’ housing meeting international standards, including health and safety regulations. UNDP recognizes that housing may become a bigger consideration as the relationship between climate change and social protection continues to evolve, as emphasized by UNHABITAT (UN Human Settlements Programme and UNDP 2014). UNICEF (2011) recognizes that rapid urbanization often strains social protection systems, and the WHO has focused increasingly on health policy implementation in cities and urban areas and also at the intersection of social protection, housing, and health policy (Clair and Stuckler 2016; WHO 2010b). The World Bank has funded housing in conjunction with broader social protection efforts (World Bank 2015b), but housing does not seem to be a central pillar of the social protection strategies of any of these organizations.
The role of education in enhancing social protection, and vice versa, was unclear or ambivalent across these organizations. UNICEF conceptualizes improvement in education outcomes as a product of social protection strategies through at least two pathways: by ensuring that household income is high enough to enable children to attend school instead of working to ensure family livelihood; and by conditioning receipt of cash transfers on school enrollment (Bloemen 2010). The World Bank (2015c) recognizes education as a social investment that increases the skills and earning potential of children, and by extension, also sees education as a way to end cycles of poverty through reliable, income-generating work in both the formal and the informal sectors. WHO (2010a) also recognizes links between education and health, but it has not discursively linked this relationship to social protection. Similarly, the IMF (2008a) considers investment in education one of the “most productive and responsible expenditures by a country.” To this end, it now incorporates spending targets in its programs that purportedly protect this policy area from overall austerity measures (IMF 2015), although recent evidence has cast doubt on whether these targets are actually met (Kentikelenis et al. 2014; Stubbs and Kentikelenis 2018; Stubbs et al. 2017; Thomson, Kentikelenis, and Stubbs 2017).
Overall, IGOs have engaged with social protection predominantly through the labor market, followed by the health sector. IGOs consider social protection in relation to family somewhat indirectly, through both child-sensitive and gender-sensitive social protection policies. IGOs discursively engage the least with education and housing in relation to social protection, though the latter, in particular, may change over time as the effects of climate change become more pronounced. Furthermore, the need to work across agencies and sectors was reinforced by the number of policy areas with which social protection has been discursively linked beyond the five areas discussed above, including climate change and the environment, energy, food and nutrition, healthy settings, migration, and violence prevention.
How, then, should the discursive shift toward universal social protection be characterized and understood in the context of the 2030 Agenda? Through the lens of our theoretical framework (Figure 2), our findings suggest that universal social protection is best characterized as a policy trend.
First, looking across policy areas, the ILO and World Bank have rebranded a number of existing activities under the heading of universal social protection. However, neither organization has rebranded all of the activities they link to social protection under this new heading; instead, they focused more on policies related to the labor market and basic health care, and less on other services when compared to the definition of social protection articulated by the Social Protection Floor Advisory Group (2011). This has occurred despite the presence of intersectoral and interagency coordination mechanisms related to social protection, such as the SPF-I and the SPIAC-B. In contrast, WHO responded to the narrowing of definitions of social protection by rebranding efforts towards income support and security as “financial protection” in the context of universal health coverage, pointing to how much understandings of social protection have narrowed.
Nevertheless, universal social protection has moved beyond rebranding, with a number of IGOs embracing a new instrument with which to enhance social protection: cash transfers. While cash transfers have proliferated widely since the early 2000s (von Gliszczynski and Leisering 2016), our review found little evidence that their adoption has led to a change in the dominant norms guiding the broader field of social protection policies, as reflected in their focus on poverty alleviation, in line with the mandate of several of the IGOs examined here. If anything, cash transfers have been successful specifically because they align with and buttress existing norms (Babb and Chorev 2016), even being referred to as “the typical social policy for neoliberalism” (Saad-Filho 2016:69). This finding also aligns with von Gliszczynski (2015:103), who demonstrates that the expansion of cash transfers “did not involve changes in the basic ideas or the introduction of more policy models” in global development, despite consensus on the use of cash transfers as a tool for poverty alleviation.
Finally, the commitment of the ILO and the World Bank to universal social protection is “universal” insofar as it aims to bring a functional, subsistence income and basic livelihood to all individuals in extreme poverty. However, as currently defined, universal social protection is arguing neither for comprehensive, contributory welfare provision nor for a rights-based social minimum. Therefore, in a classic sense, “universal” social protection is, at best, a narrow, subsistence-oriented welfare provision that just escapes equating social protection with financial protection, as evidenced by the lack of a strong partnership between the movements to achieve universal health coverage and universal social protection. This finding also aligns with von Gliszczynski and Leisering (2016:325), who characterize social protection as only “a fragmented and incomplete universalism.” Furthermore, the World Bank has not changed what it says about basic income security in all instances; it continues to promote the concept of social safety nets despite its espoused commitment to “universal” social protection, though these are two fundamentally contradictory policy approaches, resulting in incoherence in what the Bank says and ambiguity in what it means when engaging in policy discourse on social protection (World Bank 2018). In sum, the shift toward “universal” social protection, while a substantial policy trend, is neither complete nor indicative of the kind of wholesale, radical change associated with a paradigm shift.
Given the degree to which understandings of universal social protection have consolidated around the labor market and financial protection through the use of cash transfers, in line with dominant norms, we argue that universal social protection looks more like a robust policy trend than a policy fad or paradigm shift. Distinguishing between policy fads and policy trends serves as a reminder that discursive change happens both in a context and over time, both of which must be taken into account when trying to understand the significance of a policy change. Not only does universal social protection align with the dominant norms across the policy field, it can also morph and adapt in technical design and implementation both within IGOs and in a variety of countries—emblematic of what von Gliszczynski and Leisering (2016:327) refer to as “micro-paradigms,” or global models of second-order change. In an increasingly diverse and multipolar policy field (Babb and Chorev 2016), universal social protection provides a discursive umbrella that is flexible enough to accommodate not only a range of needs in its technical implementation but also pockets of normative change and differentiation throughout the field.
While universal social protection is a substantial policy change that is more than just talk, it is not yet indicative of a paradigm shift in global development policy. While, in theory, it may be possible for this policy trend to “tip” into a paradigm shift, our theoretical framework for understanding the relationship between discursive change and substantive change in global policy should not be mistaken for a linear process. In other words, it is not meant to suggest that policies discursively progress from rebranding, to fads, to trends, and finally paradigm shifts. Paradigm shifts may well be driven by exogenous rather than endogenous factors, and in theory, any shift in the dominant norms across the global policy field, whether focused directly on social protection or not, may affect the future of this policy trend.
Regarding SDG attainment, universal social protection has been discussed overwhelmingly in relation to SDG 1 (ending poverty), with little mention of SDG 5 (achieving gender equality) or SDG 10 (reducing inequality)—a trend that continues to date (ILO 2016h). In other words, universal social protection is more “pro-poor” than it is “pro-welfare” in a broader sense, in part because the mandates of several of the IGOs examined here are focused explicitly on poverty reduction. As a result, universal social protection is focused far more on alleviating material deprivation through the provision of basic income security to the extreme poor than on equalizing power relations between groups, and the lack of a more holistic understanding of social protection could hamper its attainment (Yi 2015). For instance, attempts to strengthen social protection as basic income protection may prove futile in the absence of agricultural and housing adaptation that is able to keep pace with climate and environmental changes (World Bank 2016b).
These findings suggest that the advancement of social protection as a policy agenda and the attainment of all SDGs and targets related to social protection will depend on the ability of policymakers to overcome the ideational constraints posed by organizational mandates, to reach across agencies and sectors to form effective partnerships (as called for in SDG 17), and to recognize and be prepared to take advantage of windows of opportunity to transform the policy field. In particular, the emergence of a comprehensive, less fragmented, and more complete form of universal social protection may depend on the extent to which IGOs with rights-based mandates, such as the ILO, UNICEF, and WHO, are able to continue to display ingenuity and to work collectively to ensure that social protection is about more than financial protection and poverty alleviation.
Future research could offer insight into the trajectory and development of the social protection agenda. First, intersectoral action, particularly with service-based sectors such as health and education, may be indicative of whether or not universal social protection transitions from a “pro-poor” to a “pro-welfare” orientation. Despite WHO’s early involvement in the Social Protection Floor Initiative alongside the ILO, the literature suggests that more is known about how social protection can improve health than about how better health enhances universal social protection. With regard to education, both the IMF and the World Bank—key players in financing social protection—recognize the value of education for labor market development and economic growth, and strong coordination and cooperation will be required to ensure that education policy and vocational programs lead to decent work. Second, the current efforts by the World Bank to “harmonize” and “consolidate” social protection systems are worth further scrutiny. On the one hand, these efforts could institutionalize progress made on social protection in relation to the labor market; on the other, they could introduce barriers and constraints to the development of effective partnerships to expand social protection across agencies and sectors. Finally, quantitative content analysis, exploring the extent to which social protection has become embedded across these IGOs and across sectors, could help determine whether or not the social protection agenda is well placed to challenge or subvert dominant norms across the policy field, and thereby catalyze a transformation in global development policy.