The rise of big philanthropy in global social policy: implications for policy transfer and analysis

Abstract

A new group of Western development donors has emerged as increasingly influential actors in global social policy. Big philanthropies have begun implementing social protection projects on a vast scale across the Global South and have become integrated within global governance structures. It is essential to examine whether their approach to social policy in the South is effective, legitimate and desirable for the substantive agendas and programmes in these countries and for analysis of social policy in a development context. This study investigates contemporary big philanthropies through a qualitative case-study of the Bill and Melinda Gates Foundation and its role in the health sector in Tanzania. It examines the ways in which big philanthropies engage and seek to influence policy on the ground, directly exploring the views and experiences of local stakeholders. The study finds that big philanthropies have distinctive features and mechanisms as global social policy entrepreneurs. In contrast to the vertical and linear processes associated with traditional policy transfer, a more messy and complex set of mechanisms are observed. The study also indicates that despite considerable resources and authority, philanthropic donors may not be effective in securing policy reform within aid-receiving countries due to a lack of transparency and embeddedness.

Introduction

A new group of Western development donors has emerged as increasingly influential actors in global social policy since the early 2000s. Large, tax-exempt private foundations established by super-rich individuals, referred to as big philanthropies in this study, have begun implementing social protection projects and programmes on a vast scale across countries of the Global South and have become embedded and assimilated within structures of global governance. They have further diversified the contemporary philanthropic landscape, which comprises different types of entities ranging from well-known organisations such as the Rockefeller and Ford Foundations, to their more contemporary counterparts such as the BBVA Microfinance, MasterCard, or the IKEA Foundations, as well as individual prominent politicians and celebrities. Moreover, this new generation of ‘Silicon Valley philanthropies’, established and driven by famous individual tech tycoons and hedge-fund billionaires such as Mark Zuckerberg, Pierre Omidyar (Facebook and Ebay founders respectively) and, most famously, the Microsoft billionaire Bill Gates are gaining increasing momentum and traction.

These big philanthropies differ from charities focused on humanitarian assistance or emergency disaster response and, instead, make large investments to address fundamental social protection needs including global health, education, and extreme poverty. Typically led by influential individuals, they also differ from other types of foundations in various ways. They have unique access to a broader and cross sectoral range of networks, and charismatic appeal in the eyes of the public and policy makers alike (Harman, 2016). From an organisational perspective, they are largely driven by the views and interests of their leaders (or small boards), affording them exceptional latitude in processes of internal agenda-setting. Additionally, given their significant independent resources and private sector background, big philanthropies are geared towards risky investments and ‘big bets’ (of millions of dollars) aiming to achieve quick and significant development results. Rather than investing in established social policy instruments and mechanisms, big philanthropies are keen to mobilise their ‘catalytic capital’ for innovative and novel approaches often inspired by and/or involving private sector models and actors (Gelles, 2015; The Bridgespan Group, 2019).

Wider literature on private Foundations has articulated longstanding concerns about the ambition, power and influence they wield in domestic and global policy processes, enabled by their significant material resources and elite social capital (Edwards, 2011; McGoey, 2015; Morvaridi, 2015; Reich, 2018). Conceptualised as possessing ‘hyperagency’ (Schervish, 2003) wealthy philanthropists have been characterised as fashioning both the desired outcomes and ‘rules of the game’ through their material resources and dispositions, largely at the expense of other citizens with more limited ‘agency’ within the set rules. At the same time, the involvement of traditional private foundations in processes of global social policy making has received growing scrutiny particularly in the field of global governance (Harman, 2016; Jung and Harrow, 2019; Youde, 2018, 2019).

However, despite the fact big philanthropies have become progressively influential actors in social policy formation and delivery in the Global South and within global social governance, relatively less is still understood about their goals, operational methods or longer-term influences. Furthermore, and most crucially (for the purposes of this topic), analysis of big philanthropy remains limited and under theorised within traditional Social Policy scholarship. Examining whether the approaches of Western private foundations to social policy formulation and delivery in Low- and Middle-Income Countries (LMIC) are effective, legitimate and desirable, becomes particularly essential given the extent to which the social policy agendas and programmes in these countries have been and continue to be, predicated on international aid and influence. Moreover, the action of philanthropies also has implications for analysis concerning the processes and politics of social policy making in a development context. Established theoretical frameworks for understanding social policy formation have been rooted in the post-war dynamics of the Global North, attributed largely to domestic actors and processes. Conventional explanations for welfare state emergence and social policy reform have typically encompassed theories of industrialisation (Wilensky, 1974), power resources (Korpi, 1983), and the veto-points and path dependencies related to existing institutions (Pierson, 1994). Since the 1990s, the field has largely evolved around comparative social policy and study of welfare regimes in Europe and North America (Esping-Andersen, 1990).

There is now mounting recognition that these analytic frameworks have proven less relevant in explaining processes outside the firstly industrialised countries of the ‘West’ (Midgley et al., 2019; Surender and Walker, 2013). Rather, in LMIC, colonial legacies, informal labour markets, under-developed modes of production, and weak state capacity and legitimacy have opened up spaces for other, mostly external actors, to shape and implement social welfare (Deacon, 2007; Gough and Wood, 2004; Kaasch and Martens, 2015; Schmitt, 2020). It means that policy transfer from external sources constitutes a key component in social policy processes in the Global South and social policy analysis of LMIC necessitates an understanding of both the processes and external sources (Hulme, 2005).

Policy transfer has been commonly defined by the classic interpretation of Dolowitz and Marsh (2000) as a “process in which knowledge about policies, administrative arrangements, institutions and ideas in one political setting (past or present) is used in the development of policies, administrative arrangements, institutions and ideas in another political setting” (2000, p. 5), and the analysis has mainly evolved around inter-state processes, whether evaluating ‘hard’ (regulations and programmes) or ‘soft’ (intellectual frameworks, policy knowledge, values) policies. In a development context, it has generally focused on governmental and multilateral institutions imposing more or less non-voluntary, top-down linear processes often through conditional mechanisms (Appuhami et al., 2011; Jung and Harrow, 2015). More recent works on policy transfer have expanded this framework by drawing attention to the growing knowledge networks and ‘global market places’ of ideas and practices as an axis of social policy transfer (Stone, 2017); transnational forms of governance in which non-state actors operate alongside state actors (Stone and Moloney, 2019); the role of local actors in interpreting, translating and facilitating policy adoption (Clarke et al., 2015; Hadjiisky et al., 2017) and the role of culture and context in policy transfer processes (Porto de Oliveira, 2021).

This research contributes to this literature by examining the particular social policy transfer strategies and processes in the context of aid from big philanthropy. It investigates contemporary big philanthropies primarily through a qualitative case study of the Bill and Melinda Gates Foundation and its role in the health sector in Tanzania. It assesses the means through which the Foundation propels social policy transfer globally and impacts domestic levels of governance. In doing so, it seeks to contribute to a better understanding of the strategies, and models of engagement utilised by big philanthropy to further their agenda and influence policy, and the concomitant implications for transnational social policy analysis. The research builds on and adds to earlier work on the Gates Foundation, most notably McGoey’s (2015), which highlighted the growing social policy influence of the Foundation at the global level, questioned the effectiveness of its adopted policy solutions and presented a rounded critique of the driving motivations behind philanthropic aid.

This study investigates the Foundation’s approaches to policy influence through empirical evidence from an aid-receiving country, Tanzania, and from within the Foundation itself. The study is distinctive in exploring directly the views and experiences of stakeholders in aid receiving countries: most crucially, government bureaucrats, civil society organisations and individual development practitioners in Tanzania as well as in the global development community. It examines the ways in which the Gates Foundation seeks to engage and influence policy on the ground and the views and responses of local stakeholders, and explores the implications for social practice and theory in a development context.

Background

There has been a notable increase of philanthropic funding for welfare and development since the beginning of the century, though the data is likely to be an underestimate since several foundations with significant resources are yet to report to the OECD-DAC database (OECD, 2020a). It is anticipated that the impact of this group will expand as the number of ultra-high net-worth individuals (net-worth over 30 USD million) rose from 78,000 globally in 2009 to 200,900 in 2020 (Capgemini, 2021). In 2018, the assets held by philanthropies in 23 countries (and Hong Kong) surpassed $1.5 trillion, with highest concentration in North America and Europe (Johnson, 2018). While the share of philanthropic aid is still relatively slim compared to the overall ODA aid flows ($ 9 billion versus $ 152.8 billion allocated by OECD DAC members in 2019 ¹ ), its importance is undoubtedly increasing. Significantly, these resources are concentrated in traditional social policy areas, with over 40% of overall philanthropic aid directed to the health sector alone, in comparison to overall ODA flows which are more widely spread across sectors including infrastructure and debt relief (OECD, 2020a, 2018).

The increase in financial flows has been accompanied by the expanding and organised participation of big philanthropies in global social policy institutions. These have shown growing interest in private foundations and expressed political support for their role in social policy making, financing and delivery, establishing bodies and frameworks to facilitate dialogue with these new actors (for example, the OECD’s Global Network of Foundations Working for Development (netFWD) or the SDG Philanthropy Forum run by the UNDP and WINGS). Their active incorporation into global governance accelerated after the Busan High Level Forum on Aid Effectiveness in 2011, which brought together for the first-time traditional donors, South-South co-operators, the BRICS, civil society organisations and private funders in an agreed development cooperation framework. The Post-2015 Development Agenda report on the SDGs by the High Level Panel further endorsed the view that the fight against poverty requires a ‘new global partnership’ involving philanthropic donors (UN, 2013). This rapid amalgamation into structures and processes of global social policy has enhanced their authority as transnational social policy makers and intensified the reach of their policy transfer in the Global South.

The Bill & Melinda Gates Foundation: The rise of a family philanthropy

The speed and magnitude with which big philanthropy has entered global social policy making and finance has been amplified by one particular organisation – the Bill and Melinda Gates Foundation, the world’s wealthiest and most influential private philanthropy. The Foundation’s endowment assets approximate $50 billion, and in 2019, it donated $3.9 billion in Development Assistance for Health (DAH). ² Its yearly endowments bypass entire budgets of leading international organisations such as the World Health Organisation (WHO) or the United Nations Development Programme (UNDP).

The Foundation also stands as the single largest donor to various international organisations and UN bodies, with considerable influence over their policy directions (Youde, 2018). The Foundation has provided grants to the OECD and OECD-DAC on a yearly basis since 2008 and in 2018 alone, it allocated $253.5 million to the WHO (being its second largest donor after the US); $109.3 million to the World Bank, and $93.9 million to UNICEF (OECD, 2020b). It also initiated the launch of Gavi, the Vaccine Alliance (GAVI) in 2000 with an initial commitment of $750 million and has contributed $2.25 billion to the Global Fund’s budget to date (having pledged a further $760 million for 2020-2022), whilst holding seats at the boards of these new global PPPs. Other recipients of its direct, charitable grants include research institutions and universities, global private sector actors and businesses (including pharmaceutical companies), and social enterprises.

Most of the Foundation’s aid investments are directed to sub-Saharan Africa with significant investments in India and Pakistan also. Its primary areas of health investment are infectious diseases (HIV&AIDS, malaria, tuberculosis), immunisation, maternal, new-born and child health and family planning, and the Foundation is known for its ambitious targets to eliminate polio and eradicate malaria in Bill Gates’ lifetime. In keeping with other contemporary big philanthropies, the Gates Foundation is driven by an ethos of ‘making capitalism work for the poor’. Characterised as ‘philanthrocapitalism’ in the literature (Bishop and Green, 2009; McGoey, 2015) it prioritises innovative, techno-scientific solutions and business approaches in order to achieve inclusive economic growth. Rather than pursuing social equality through direct wealth redistribution, this approach is focused on enhancing human capital and individual assets for the poor to ‘lift themselves out of poverty’. In this sense, the Gates Foundation’s normative framework is most closely aligned with a productivist or instrumentalist approach to social policy, driven less by notions of social justice and rights, and more in terms of social investment. Using a logic of entrepreneurship and enterprise, the approach seeks to address both market failures and State ‘institutional voids’ and positioned as an ‘apolitical’ paradigm – it emphasises pragmatism, science and technology and evidence-based policy. Its emphasis on investment in innovative health care instead of immediate poverty alleviating measures such as social pensions and cash transfers reflects its normative framework and goals.

Method and Data Sources

This research consists of a qualitative case study of the Gates Foundation and their work in the health sector in Tanzania. Tanzania provides a particularly information-rich context for several reasons. First, it is a ‘busy’ aid-receiving setting where diverse donors and domestic institutions compete for policy influence. The fifth largest recipient of Gates’ funds (OECD, 2020b), it has a longstanding relationship with the Foundation dating back to the 1990s, and was bequeathed a further $350 million five-year investment in 2017. Tanzania also remains a ‘donor darling’ and holds long-term relationships with other traditional multi- and bilateral development agencies, as well as China. Second, despite inward aid and some progress in improving its health outcomes, Tanzania faces significant ongoing challenges with 42% of all aid received in the country, directed into the health sector. Third, with its strong post-colonial socialist foundation, Tanzania has strived to harmonise donor activities and ensure its own policy preferences are respected in the course of their implementation. The government has made significant efforts to reduce fragmentation of external development investments by establishing structures such as the Development Partners Group (DPG), and to control aid efforts through consensually agreed frameworks. It thus provides a valuable context for investigating the Foundation’s engagement models in a comparative context and their potential for policy influence on the ground.

The core data consists of 40 in-depth and semi-structured in-person interviews conducted in 2017 with senior officials and professionals, and experienced stakeholders. These included Gates’ officials at their Seattle headquarters; representatives from its implementing partner organisations both in their US headquarters and in Tanzania; global development experts from international organisations, bilateral donor agencies in Tanzania and from global headquarters; and, Tanzanian government officials and leading civil society representatives in the country. Interview data was supplemented by a comprehensive documentary analysis of over 50 publicly available sources dating between 2015-2020, including the Foundation’s annual reports; meeting reports and policy documents; statements on the Foundation webpages; speeches and interviews with the Foundation leaders, and, their personal blog posts.

Interviewees from the Gates Foundation were identified via publicly available databases and some ‘snowballing’, while Tanzanian interviewees were contacted through personal and professional connections and recommendations from other interviewees. Interviews were taped and transcribed and transcripts coded and analysed using the software package NVivo 10. A thematic framework analysis was adopted (Gale et al., 2013; Srivastava and Thompson, 2009), which avoided both a positivist hypothesis testing approach or a firmly inductive grounded theory method. Five broad steps of familiarisation, indexing, charting, mapping and interpretation was undertaken across both documentary and interview data. It allowed identification of key themes to emerge from both pre-determined questions as well as novel and emergent accounts. The processes for arriving at reliability and verification of conclusions was strengthened by initially using both structural and descriptive coding and by subsequently using a matrix to compare the content of data between different individuals, interview groups and sources of information. This detailed approach to mapping linkages and triangulating information reinforced interpretation and explanation of the findings. To ensure confidentiality, participants’ names and professional positions have been anonymised. The methods were approved by the Research Ethics Committee of the University of Oxford and the Commission for Science and Technology in Tanzania.

A further review of reports and policy documents, publications and webpages for 10 other leading philanthropic donor organisations operating in the social sector in LMIC was also conducted. ³ Finally a ‘data bank’ was developed compiling information about the Foundation’s health-related project investments in Tanzania and other LMIC between 2015-2020. Information was largely drawn from Foundation’s awarded grants database and webpages, which provided important contextual knowledge about the Foundation’s priorities and approaches in the health sector.

Findings

Instigating policy transfer though domestic gatekeepers, multi-actored and covert processes

The Gates Foundation’s delivery of welfare assistance to individual countries entails a specific format and is typically streamed through various ‘middlemen’ or intermediaries. Its programmatic aid involves 3-5-year pilot projects implemented by third parties such as private sector companies specialising in development work or large, professionalised NGOs and INGOs, rather than funds directed to government agencies. These independent mediators were the primary point of contact for the Tanzanian government and, notably, often ‘masked’ the identity of the Gates Foundation as the original funding/policy source. The rationale for this model from Gates interviewees was that partner-led work allowed ‘domestically based’ actors (even if headquartered in the US) to better bolster ownership of policy initiatives with the government and other domestic development actors. Foundation employees and their implementing partners stressed the benefits of projects which were visibly staffed by local Tanzanians.

“…our partners all have capacity to be the best advocates for the issues that they work on – whether those are small NGOs or being big multinational organisations … they are generally in the best position to tell their own stories and explain why it makes sense to invest in certain issues… scale up programmes…” (Gates Representative 9).

“The Gates Foundation, they don’t push…their partners to use their name, so they don’t exist as ‘Gates Foundation’ on everything, you know… So that countries feel it’s their thing, and it is good for sustainability as well.” (Implementing Organisation Representative 6).

The Foundation also frequently contracts established and locally well-connected senior development practitioners as consultants and advisors. According to interviewees, these respected and well positioned domestic associates allow the Foundation to influence policy making in a ‘roundabout manner’ by gaining access to valuable internal knowledge on policy developments or tensions within the domestic sphere that might otherwise be inaccessible to foreign development practitioners. It also enhances its local credibility through collaborations with influential domestic actors.

“When you have a good consultant who is knowledgeable, well known, who can interact with senior policy makers in the country, it is to your advantage… you can easily get to know what is happening, you also get some input into the thinking of the government” (Global development expert 1).

“We are also increasingly interested in local philanthropy…. because we really see that local buy-in and skin in the game can be incredibly valuable…. [Local philanthropist] is highly influential, highly respected and he brings us lots of credibility” (Gates representative 8).

Another, and important, example of strategically capitalising on the advocacy power of local actors involves the building of domestic civil society movements. The Foundation’s health sector investments show that it is increasingly funding and training civil society organisations for bottom-up promotion of its priority causes and policy objectives, in Tanzania as well as elsewhere in sub-Saharan Africa. One example of this is the Immunization Advocacy Initiative, which trains civil society organisations in evidence-based advocacy techniques in order to enhance bottom-up promotion of increased immunisation in Kenya, Ghana and Ivory Coast. This strategy, known as ‘astroturfing’, relies on masking the original actor behind policy initiatives through engagement of local grassroots actors as policy promoters, and has previously been deployed by the Gates Foundation when investing in US education (Barkan, 2012).

However, Tanzanian officials and other stakeholders on the ground expressed frustration and concern about the Foundation’s lack of direct personal engagement and participation in established donor platforms and forums. Health and finance officials responsible for donor collaboration, aid flows and policy planning repeatedly stated that they knew very little about the Foundation since it remained absent from the structures of collaborative policy making such as budget support roundtables, sector-specific working groups, or the Development Partners Group – considered the most important structure for harmonising investments by aid agencies and alignment with government-approved policy priorities. The lack of engagement, information sharing and use of multiple proxy implementing partners, caused confusion and frustration amongst both policy makers and practitioners.

“Bill Gates is hardly ever working in their own name; they always have sub-contractors. When I hear ‘PATH’, I will know [assume] that Bill Gates is here. So, we go to the PATH event…I suspect [assume]… if it is malaria – then it is Gates’ money.” (Development expert 5, emphasis added).

“I often get requests to meet with the Foundation… or with their consultants; […] So it’s almost like they come and they want to pry some information because they aren’t physically here, and then they go off and they do their own thing but without much feedback really.” (Development expert 7).

It was also widely acknowledged by stakeholders, including those with the most positive views about the Foundation, that this lack of embeddedness meant that simple shifts in opinion or leadership within philanthropic organisations could drastically change their policies and approaches, as they remain largely ‘cut off’ from the populations and countries they serve. According to interviewees, this risk was further intensified by the idiosyncratic nature of individually led private foundations. Lack of bureaucratic constraints and the leverage of individual founders led to strong concerns about the potential risk of abrupt, complete disappearance of philanthropic aid. ⁴ Government officials especially reported that philanthropic donors – with their absence of physical offices and local staff – gave government actors a greater sense of vulnerability than bilateral aid agencies which are bound to their embassies and representative countries and therefore physically accessible at all times and more permanent.

Local officials and partner agencies suggested that while the poor visibility and engagement of the Foundation resulted in part from the outsourcing of projects and its ‘light in-country presence’, the result was not altogether unintentional. According to senior development experts, some of the Foundation’s ‘secrecy’ stemmed from its competition driven, private-sector mindset. The experimental and entrepreneurial nature of some of its research projects and the race to introduce new tools and software: for example, testing the effectiveness of treatments, resulted in cautious information sharing. Whatever the drivers, it was evident that limited information sharing, use of deputised commissioned proxies, and in many cases, deliberate masking of its identity and secrecy was a specific policy transfer strategy of the Foundation.

Instigating policy transfer through financial pressure: deploying ‘soft’ conditionalities?

Social policies have been transferred to the Global South for decades in a non-voluntary manner by colonial empires as well as “agencies of modern internationalism” (Rodgers, 2014). These include traditional aid donors who have imposed social policy reforms through conditionalities attached to the allocation of financial resources. Literature regarding transnational policy transfer demonstrates different degrees of coerciveness; a continuum that has fully voluntary policy learning by policy ‘adopters’ at one end, and coercive policy imposition by policy ‘senders’ at the other. Many differing forms of formal and informal pressures and conditionalities lie in between. (Dolowitz and Marsh, 1996; Evans, 2013; Marsh and Sharman, 2009). Southern donors have also recently attracted attention about application of commercial pressure in the context of development assistance, and brought into question the alleged unconditional nature of South-South cooperation (Asmus et al., 2017). It is relevant therefore to examine the extent to which major philanthropic actors may also apply policy pressure to aid receiving countries.

The findings of this study suggest the Gates Foundation’s aid delivery includes elements of what might be called ‘soft conditionalities’. Through a range of strategies, it utilises its resources to press for adoption of policies or programmes that were not actively or voluntarily being sought.

At the most fundamental level, any donor funding inherently entails policy transfer mechanisms of inducement (Collier et al., 1997); donor control over what is funded and how the resources are deployed. There was a strong perception that through its projects, the Gates Foundation directed resources into the policy areas and approaches of its preference and the Tanzanian government was induced to comply. Both government and civil society representatives felt strongly that the Foundation’s projects and approaches commonly bypassed some of the existing policy agendas and preferences of the Tanzanian government. Recurring themes focused on the issue of vertical health interventions which were seen as narrowly defined and less effective than the system-wide approaches widely preferred on the ground. The Foundation’s conspicuous lack of interest or alignment with Tanzania’s Health Sector Strategic Plan IV (HSSP 2015-2020) which emphasises public and primary health care, together with its planned launch of a new, universalist health insurance scheme, was criticised.

There were several other divergences. Key priorities for the Foundation, such as infectious diseases (HIV, tuberculosis or malaria) were viewed as out of line with the increasing needs (and plans) to address non-communicable diseases such as diabetes, cardio-vascular diseases and cancer.

“In the review of the landscape and the burden of health in Tanzania, we have now aligned to the burden of disease, now shifting from communicable diseases into non-communicable diseases. […] There’s no donor; the government runs and runs and feels this is important, but yet there is no donor.” (Development expert 3).

The Foundation’s population targeting – for instance directing scarce resources towards minority groups such as sex workers, sexual minorities and drug users in the context of HIV – was seen to compromise the needs of mothers and children and undermine Tanzanian strategic planning, as was the Foundations neglect of regional policy plans. Donors were repeatedly criticised for autonomously implementing projects in pre-decided locations without consultation with local authorities.

“There was a very big programme of methadone replacement, and also needle exchange…I see the technical rationale… but here we are struggling to get syringes and needles for treatment of important conditions! And if people see me giving needles for people to inject illicit drugs, as part of a public programme, how can I justify it? […] So that’s donor push…Given the choice, the government might choose the mother and child” (Development expert 6).

“Someone will come and tell me ‘I want to go work in a certain region’. I say ‘how did you know? I know my country better than yourself, but your project is already said you are going to work in a certain region only, why? (…) there is a health management team in that region, you should sit together, and discuss with them and see their priorities and their system, to achieve what they want’.” (Government official 3).

Several domestic development actors also observed that the Foundation pursued close connections to private sector actors whose products they sometimes promoted within Tanzania. A repeated example was the introduction of the self-injectable contraceptive Sayana Press in the context of the Foundation’s family planning investments and the conflicts of interest given Gates’ known investment in Pfizer Inc. It was argued this represents a tangible example of policy transfer tying aid to specific products or services determined by the donor. This form of conditional aid delivery is commonly associated with traditional and Southern donors pursuing national commercial interests (e.g. Asmus et al., 2017; Lauria and Fumagalli, 2019), but evidence of its use in the context of philanthropic aid underscores the need for further analysis of the motivations and mechanisms of big philanthropies.

“They can’t provide assistance unless they know that you will buy, medicines or whatever… So this is two way; they are helping you, but you have no option. You need to take in the expertise, you need to buy the medicine from them; you need to work with them.” (Government official 4)

Other administrative and contractual practises were also described as forms of pressure. Differing from typical project aid directives by INGOs, Gates Foundation’s Memorandums of Understanding (MOUs) with recipient governments generally include clauses detailing government measures to sustain projects after the pilot phase. Arguably ‘administrative conditionality’ rather than direct coercion, it was nevertheless felt to constitute another layer of donor influence upon domestic policy making and interpreted as another form of ‘soft’ conditionality.

“… for the Gates Foundation, it is important for them to see the government interest and plans that they want to take over. So they encourage the government to see how and what to do, they want to see what is contributed to sustainability or what is being done: change of policy, putting in money….” (Implementing organisation representative 6).

“We might say ‘we are going to support this programme but [only] if you accept that you are going to hire people, to continue implementing this programme…’ those kinds of things. It is really towards sustainability and ownership of the programme down the road.” (Gates representative 6).

Finally, it is noteworthy that Gates Foundation interviewees characterised the organisation’s evidence-based approach as non-political and credited this with enhancing its credibility and ability to specify the terms of aid delivery. In addition, interviewees emphasised the Foundation’s exceptional comparative advantage that came with the fame of the co-chairs and compelling brand name of ‘Gates’. It was often asserted that the charismatic authority of the Foundation leaders allowed the organisation to raise media visibility for their agenda and capture the attention of policy practitioners. It promoted the Foundation’s work globally as well as in Tanzania – and was seen to galvanise and attract other donor agencies to support its policy directions. It further illustrates the use of important epistemic and normative authority alongside soft conditionalities.

“They have an enormous convening and messaging power, right, partly because of the resources and partly because of who they are. Because of this, they can get attention and headlines, and attention on issues, which is powerful.” (Development expert 9).

“(…) because they have a brand name – they can easily influence other donors (…).” (Implementing organisation representative 6).

Discussion

It is clear that Western big philanthropies represent a new powerhouse as increasingly influential transnational social policy actors; not private service providers competing according to regulatory frameworks established by the state but as rivals to states and international social governance organisations. This has important implications for the dynamics of global social policy, and challenges our understanding of transnational social policy transfer. In the context where state-centric analyses of policy making are progressively contested by the proliferation of non-state agents and arenas (e.g. Kaasch and Martens, 2015; Stone, 2017; Stone and Moloney, 2019), especially in LMIC, it is essential that social policy analysis expands its frameworks to incorporate and critically analyse the transfer practices and mechanisms instigated by these new actors.

This paper explores the means of policy influence and mechanisms of social policy transfer instigated by big philanthropies through a case study of the Gates Foundation’s engagement in Tanzania, as well as some of the comparative risks of philanthropic engagement from the perspective of local stakeholders with direct experience of working with the Foundation. Though not the first of its kind (Fejerskov, 2018; Mahajan, 2018), it is one of relatively few works that incorporates the experiences and perspectives of domestic stakeholders and recipients of aid delivered by big philanthropies.

This study suggests that big philanthropies have distinctive features as global social policy entrepreneurs that both challenge our understanding of the repertoire of approaches to social policy transfer typically associated with aid delivery, while also confirming existing critiques of external ‘imposition’ and ‘intervention’ in defining policy priorities in the Global South. Harnessing the power of charismatic influence and championing independent and evidence-based solutions, the Gates Foundation emphasised partnership with local agents. Nevertheless, the findings from this study suggest that their interventions were experienced as (if not explicitly coercive) still ‘imposed’ – externally derived and driven policy priorities which are still not fully collaborative. Besides the examples of MOU sustainability clauses and resources ‘tied’ to specific products, described in this study as ‘soft conditionalities’, the Gates Foundation routinely deploys Performance Based Funding and ‘Cash-On-Delivery’ agreements which have been criticised as analogous to the logics of coercive SAPs in the 1980s and good governance conditionalities in the 1990s (Fraser and Whitfield, 2008).

Viewed as even more problematic was the Foundation’s deliberate strategy of ‘covert’ and ‘long arm’ engagement in Tanzania. In contrast to other external development actors in Tanzania, who directly engaged with government officials and collaborative donor platforms, the Gates Foundation adopted indirect and circuitous mechanisms to deliver programmes and influence government direction; mobilising intermediary agents including INGOs, influential consultants and local civil society actors. With the exception of Ethiopia, where the government has demanded greater transparency and harmonisation with domestic policy processes, this model appears representative of the Foundation’s approach elsewhere. It is noteworthy, that other leading big philanthropies, such as the Children’s Investment Fund, Bloomberg Philanthropies and the Hewlett and Packard Foundations have similarly increasingly emphasised their interest in enhancing local policy advocacy to achieve their set policy goals, suggesting a growing trend of ‘astroturfing’ strategies in the context of philanthropic policy transfer.

Perversely, the Foundation justified its strategy as strengthening local buy-in and ownership of projects, ensuring they were informed by indigenous knowledge and preferences. They appeared acutely aware of, and keen to distinguish themselves from, past coercive policy practices by Western development agencies, particularly the Bretton Woods institutions in Africa, which have been widely criticised and discredited for coercive conditional lending and policy imposition (Surender & Walker, 2013). However, stakeholders in Tanzania stressed that, because of lack of transparency and embeddedness, confidence in the Foundation’s motivations was undermined rather than reassured. Rather than seeming impartial, the Foundation seemed unaccountable and undermining of democratic governance processes. Characterised as ‘strategic camouflage’, the approach expediently avoided wider government and public scrutiny. Local stakeholders also argued that the absence of collaboration and democratic accountability resulted in a lack of alignment with donor priorities and a fragmentation of aid landscape. The Foundation’s externally designed policies and programmes were perceived to be driven by ‘individual and personal preference’, ignorant of Tanzanian cultural norms, political traditions and institutions, or domestic needs and priorities. It suggests that despite their considerable resources and charismatic authority, philanthropic donors may not be as effective in promoting policies and programmes within aid-receiving countries as might be expected. Lack of alignment with domestic priorities and direction of travel are likely to affect their feasibility and long-term sustainability on the ground. As such, the philanthropic paradigm may entail considerable risk of incomplete transfer and inappropriate transfer i.e. socio-economic and politico-ideological mismatch of social policies Dolowitz and Marsh, 2000).

These findings also indicate that in contrast to the policy transfer processes traditionally associated with aid delivery, the processes uncovered here exhibit messy and complex policy transfer mechanisms. Conventional analysis postulates that social policies typically travel from donor countries and organisations to aid-receiving governments utilising financial incentives and hegemonic ideas under largely vertical and linear processes (Appuhami et al., 2011). Moreover, traditional social policy analysis has largely organised its analytic frameworks around somewhat discrete and compartmentalised levels of governance such as the global, regional, national and sub-national. Challenging this view, our findings show that social policy transfer from Western big philanthropies to countries of the Global South incorporates diverse types of financial, epistemic and charismatic influences and involves intermediary actors across the spectrum from international NGOs and development organisations to sub-national civil society spheres. Together with the Gates Foundation’s engagement and financial flows to multilateral and global governance institutions, it points to a much more multi-levelled, multi-actored and multi-dimensional form of social policy transfer.

To this extent, these findings support and add to the alternative notion of ‘complex multilateralism’ (O’Brien, 2002) which demonstrates how the actor landscape in global social policy has rapidly diversified from traditional state and multilateral institutions to include non-state actors. It is evident that individual philanthropists and big philanthropies, working together with other private sector entities and civil society organisations – occasionally amalgamated under new powerful structures – are adding to this actor proliferation and are increasingly integrated as ‘key players’ in global social governance. In addition, this new dynamic further reinforces the idea of ‘politics of scale’ (Papanastasiou, 2019; Stubbs, 2005). Instead of driving policy making through established, compartmentalised and hierarchical levels of governance, we observe that the multi-levelled and circuitous ways in which the Gates Foundation operates transcends traditional notions of space. As such, the operational model of the Gates Foundation demonstrates the fluidity of ‘politics of scale’, where strategic engagement and influence is achieved in highly entwined arenas and networks across global, national and sub-national levels.

It is clear from the empirical findings in this study that, whereas traditional welfare state analysis is predicated on the idea that the nation state is the primary agent for social policy formulation and implementation, in the context of development, social policies and programmes are frequently determined by forces outside national borders – in this case via large philanthropic donors. The Gates Foundation, together with other new big philanthropists, have significant impact on the quantity, distribution and types of aid flowing to individual LMIC and increasingly represent and drive new forms of multilateralism. As such, it is essential that social policy theory and critical analysis concerning external actors in social policy making within the Global South includes a focus on big philanthropies. Since these new donors are less embedded and bound by democratic or bureaucratic structures both in their countries of origin, or as evidenced in this study, within the countries they seek to assist, it is necessary to problematise their decision-making processes and the mechanisms they use to transfer and implement their social policy ideas and programmes.

Competing interests

The author(s) declare none