Any alternative to the Wall Street Consensus? Comparing the infrastructure financing models of the US, the EU, and China

ABSTRACT

Global South countries (excluding China) need USD 40 trillion in infrastructure finance by 2035. Most of this must be sourced externally. The US’s Partnership on Global Infrastructure and Investment, the EU’s Global Gateway, and China’s Belt and Road Initiative all aim to provide parts of this capital. Conceptualising the state-capital relation of the US’s approach, the ‘Wall Street Consensus’ appoints host countries to catalyse private capital by de-risking infrastructure projects. Here, I conduct a comparative study of the state-capital relationships within the financing models of the three initiatives, drawing from 123 interviews conducted across the global North and South, official documents, and existing literature on infrastructure finance. I find that while the three models assign different roles to host countries, they all entail a heavy financial risk burden. Highlighting differences between initiatives, I conceptualise the US‘s model as ‘host country de-risking’, the EU’s as ‘ODA bargained de-risking’, and China’s as ‘state-organised risk-sharing’. Positioned between these models, I find that global South countries face a dilemma between the need for a ‘green state’ approach to drive infrastructure development and pressure for a limited risk-carrying state role imposed by the three models.

KEYWORDS:

• US
• EU
• China
• infrastructure finance
• Wall Street Consensus

Introduction

To meet economic and sustainability goals, developing countries need $40 trillion for infrastructure by 2035 (White House 2021a). Most are unable to meet this financing need without external financing (Songwe et al. 2022). Different ideas exist on how this external capital should be channelled to global South countries, with each idea assigning different roles to state and private actors in host and origin countries. These ideas are operationalised as financial models with a set of financial tools to be used. The most influential of these ideas are those embodied in the infrastructure financing initiatives of the world’s largest economic powers: the US’s Partnership for Global Infrastructure and Investment (PGII), the EU’s Global Gateway (GG), and China’s Belt and Road Initiative (BRI). Here, a dilemma arises: While a consensus is emerging that an expansive and developmentalist state role is needed to steer sustainable development trajectories (Johnstone and Newell 2018, Deleidi et al. 2020, Hausknost and Hammond 2020, Meckling and Allan 2020, Green 2022), certain financing models restrict the policy scope and limits the roles of both origin and host states in their state-capital relationship (Gabor 2021, Alami et al. 2022, Schindler et al. 2023). Consequently, global South countries may be squeezed between the need for an active state role to steer development and the need to adhere to a certain limited state role to access global capital.

In this paper, I investigate this dilemma, contributing to the proliferating debate arising from Gabor’s (2021) conceptualisation of the Wall Street Consensus. I specifically speak to the question raised in the last sentence of Gabor’s (2021, p. 454) article, ‘If the Wall Street Consensus is the strategic Western response to China’s Belt and Road Initiative, how will these competing development paradigms co-evolve?’. Since the publication of Gabor’s article, scholarship has moved the debate forward, such as by questioning the severity of the suggested limitations (Alami et al. 2021, O’Sullivan and Rethel 2023, Bernards 2024) and by providing greater agency to host countries in addressing the limitations (Chiyemura et al. 2023, Schindler et al. 2023, Van Wieringen and Zajontz 2023). From that, a comparative analysis of differences between financing models promoted by the US, the EU, and China, can provide valuable insights into the dynamics of both state-capital relations and host-origin country relations.

The article makes one central finding: The three models all assign a heavy financial risk-burden on global South host countries – in different ways. I conceptualise the US’ model as ‘host country de-risking’ through guarantees covering political, currency, project, and other risks. This model adheres to the principles of the Wall Street Consensus. However, while Gabor (2021) ties the US and EU together as a ‘Western’ financing model, I suggest that there are noteworthy differences. I conceptualise the EU’s model as ‘overseas development assistance (ODA) bargained de-risking’, constituting de-risking through guarantees by both host and origin country in combination with substantial financial aid and technical assistance. While private investors from the US and EU are indifferent as to who does the de-rising, the EU and its member states are willing to carry part of the financial burden. As a third way, I conceptualise China’s model as ‘state-organised risk-sharing’, in which the host country de-risks as the sole recipient of loans and China de-risks by providing patient capital with few conditions through state-sponsored package deals. As opposed to the US and EU models, the purpose of China’s model is not to attract private finance but to reduce risks for state-actors in both the home and host country. I find that the models all entail heavy financial burdens on host countries’ and might limit their use of developmentalist policies. Finally, I argue that host countries have agency in selectively engaging with each model to limit the effects of the potential constraints.

The question of manoeuvrability within a dilemma

How much manoeuvrability do host countries have between the pressure to adhere to a certain role within the three financing models and the need to actively use the state to steer their development trajectory? This question lies at the core of the problem raised by Gabor (2021) regarding the state-capital relationship of the Wall Street Consensus. While the importance of this question remains central, recent literature has added nuance to several important dimensions. I build on this recent literature to address this question, which is at the core of an emerging research agenda.

First, recent literature questions how much the financing model of the Wall Street Consensus actually limits policy options of host countries. Scholarship suggests that host countries have a certain degree of agency and manoeuvrability of policy choice while still adhering to the de-risking role imposed by the financing model. Schindler et al. (2023) show how de-risking states under the Wall Street Consensus financing model have policy scope wide enough to conduct industrial policy. Similarly, O’Sullivan and Rethel (2023) show that while host countries are conditioned by the Wall Street Consensus, they use state financial activism to also pursue the interest of the domestic middle class. Adding to that, Alami et al. (2021) show that multilaterals have a discursive role in promoting a greater state role, such as, but not limited to, the de-risking state of the Wall Street Consensus. Additionally, recent literature questions whether the Wall Street Consensus does, in fact, represent the interest of private global North capital. Bernards (2024) and Copley (2022) both argue that it is rather a result of states trying to grapple with contradictions in contemporary capitalism, such as overaccumulation in the global North and underdevelopment in the global South. If this is the case, the dilemma can not be solved by switching to another financing model, but only by reforming capitalism.

Second, existing literature highlights key differences and similarities between overseas infrastructure finance in the US, the EU, and China. According to Gabor (2021), the US approach is aligned with the interests of Wall Street as a ‘development as de-risking’ model, limiting the role of states in the global South to de-risking assets in order to catalyse private capital. The EU’s approach has been characterised by an aid focus, integrating grant and concessional elements into development finance. In recent years, the rationale has changed from combating global poverty to aspects more directly aligned with the EU’s foreign policies, such as migration, security, and climate change (Carbone 2017, Mah 2023). Literature on Chinese development finance emphasises its differences from the US and the EU. For example, it stresses Chinese lending’s patient nature, characterised by longer time horizons than US lending (Kaplan 2018). Furthermore, the literature highlights that the limited conditionality of Chinese lending amplifies existing political institutions in host countries rather than changing them (Caskey 2022). Finally, some scholars argue that differences are limited and that the three parties are converging in their practice (Skalamera Groce and Köstem 2021, Van Wieringen and Zajontz 2023). Overall, literature is abundant on the development finance of all three parties, thus providing key input for a comparative study.

Third, to understand the three models, it is simultaneously critical to address their interconnectedness. With the rise of China, development finance has become an arena of competition between states and ideas (Henning 2019, Kring and Gallagher 2019, Pape and Petry 2023). For decades, Western countries have provided financing through multilateral agencies, bilateral mechanisms, and private financiers. Facilitated by China’s rising economic might, the country entered the stage with the launch of the BRI in 2013. With a direct state-to-state approach, fast rollouts, and designated construction companies, the BRI was welcomed in many countries. In some areas, China’s approach complements Western activities, while in others, it challenges the Western approach and influence (Chin and Gallagher 2019). As China challenges Western global dominance, these dynamics spill into development financing. It was, therefore, partially in response to the BRI that the US launched the PGII, and the EU launched the GG. Given these models’ influence on states’ policies, global South countries risk getting caught in geopolitical conflicts beyond their control. In this way, the article addresses questions raised by Kring and Gallagher (2019) regarding the coherence, stability, and conflicts in the changing institutional order of development finance.

Finally, regarding the second part of the dilemma, a consensus has emerged that an expansive and developmentalist state role is needed to steer sustainable development trajectories (Johnstone and Newell 2018, Deleidi et al. 2020, Hausknost and Hammond 2020, Meckling and Allan 2020, Green 2022). Scholars use a variety of terms to capture this active, interventionist, and expansive state role, which involves using the state’s balance sheet and state apparatus through fiscal policy, monetary policy, financial regulations, as well as state-owned financial institutions and enterprises. They emphasise green transition as the core issue to be prioritised alongside economic development. Examples of such terms include ‘green developmental state’ (Gabor 2021) ‘big green state’ (Gabor and Braun 2023), ‘environmental state’ (Hausknost and Hammond 2020), ‘ecostate’ (Meadowcroft 2005), and, in more popular discourse, a ‘green new deal’ (Chomsky and Pollin 2020). While differences exist between the terms, they all promote a state role larger than the role imposed on host countries by the Wall Street Consensus. As a shorthand for the collective idea behind these terms, I use the term ‘green state’ in the remainder of the paper.

Three initiatives, three different financing models?

To address the manoeuvrability of host countries within the dilemma, I conduct a comparative case study of the financing models of three infrastructure finance initiatives (PGII, GG, and BRI). First, I question the homogeneity of each initiative by looking for diverse opinions and approaches within state-organs, the private sector, and international organisations. This approaches how, underneath the surface of the neat rhetoric of the state-led initiatives, there exists a web of complexity of stakeholders, a history of development policies, potentially conflicting interests, and overlaps with other initiatives. For example, large infrastructure projects in the global South often have a wide array of participating stakeholders that may prefer a certain financing model. In practice, while the financing models advanced by each participant may be apparent, projects sometimes end up with a financing model that mixes participants’ models. This intersection of models may create manoeuvrability for host countries, which is explored in the discussion section. Second, I examine the change of financing models over time between current and past infrastructure initiatives of the three parties, especially given that the US’s and EU’s initiatives were recently launched. Third, I investigate potential gaps between rhetoric and practice and address these by triangulation of sources and data as far as possible. Fourth, I assess how the initiatives relate to each other, as this tells us how origin countries, host countries, and private financiers can potentially engage with the initiatives simultaneously. Finally, tying these four aspects together allows me to assess how the financing models shape the dilemma that host countries may find themselves in.

I use primary data collected in 123 interviews as well as secondary material in the form of official state-issued documents and existing literature on infrastructure finance. Interviews were conducted across the global North and South, covering public and private financial institutions, government representatives, technology providers, construction companies, law firms, and researchers. The interviews were conducted in 2022 and 2023 during stays of one to two months in China, Vietnam, India, Ethiopia, Brazil, the US, and Denmark. All interviewees were asked specifically about whether the three infrastructure finance initiatives constitute distinct ‘models’ and, if so, about their characteristics and impacts on host countries. A difficulty with a state-centred study is tying a financing model to an origin state and subsequently showing how this model is imposed on a host state. Addressing this issue, the interviews allow me to go beyond what can be seen in public rhetoric, quantitative data on financing, and single-country case studies, and grasp how people in origin countries perceive and operationalise their financing model as well as how people in host countries understand their financing options and their implications. From those insights, I can comparatively investigate heterogeneity within initiatives, change over time, rhetoric-practice gaps, and interrelations between the initiatives.

The US’s Partnership for Global Infrastructure and Investment

The PGII was launched at the G7 in 2022, building upon Build Back Better World (B3W) launched a year earlier. From what can be known given the short history, the PGII appears to be a continuation of the US model of infrastructure finance already in existence. It builds upon and adds to two previous initiatives, namely, the Blue Dot Network (BDN) and the Better Utilisation of Investment Leading to Development (BUILD). Therefore, past and ongoing practice is considered representative of the PGII since no policy change is indicated in the interviews or official documents and statements. As a summary of the below assessment, I find that the PGII can be conceptualised as a ‘host country de-risking’ model, largely in line with what Gabor (2021) refers to as the Wall Street Consensus: The PGII uses host countries’ balance sheets to de-risk projects in order to attract international private capital.

Two core aspects are emphasised in the PGII: It is launched as a response to the BRI and relies on catalysing private capital. On the first account, the B3W was launched based on a G7 discussion on both strategic competition with China and infrastructure in developing countries (White House 2021a). Furthermore, according to Meeks (2021), the Chairman of the House Foreign Affairs Committee, the US under its last administration, ‘retreated into an America-alone policy that ceded leadership to rival nations that were more than eager to fill the void’, suggesting that it is a clear response to the BRI. Many of the characteristics officially highlighted in the PGII are also in contrast to US official criticism of the BRI. This includes emphasis on transparency in processes and investments, broad stakeholder consultation and participation, high-quality components, and prioritisation of climate change. As a second core principle, the initiative aims to mobilise private capital through host country state guarantees. Rather than committing a certain amount from the US state budget, the US commits to ‘mobilising’ USD 200 billion over the next five years through a combination of grants, federal financing, and leveraging private sector investments (White House 2022).

The ‘host country de-risking’ model of the PGII severely inhibits countries from using the full policy toolbox of a green state. Both existing literature and extensive interviews with a broad range of participants show that the only role within the model is for states to de-risk projects through guarantees. At the global level, interviews with past and current US government and World Bank staff members show that the US and the bank have promoted privatisation and marketisation of energy and electricity markets, emphasising that the state should only provide guarantees for prices and uptake.¹ On the host country side, government officials and consultants to government bodies expressed in interviews that they were under great pressure from the US and the World Bank to reduce the state's role in the market and to expand the issuance of guarantees to attract capital from the West.² When asked about the US financial model, investors from across Western countries expressed in interviews that they would not invest without the host country states providing guarantees covering risks of currency, political, electricity uptake, and more. For example, as summarised: ‘Without guarantees, life is just too uncertain, it’s just unbankable’. ³

This model identified by interviews supports findings of existing scholarship regarding the US government’s relation to the Wall Street Consensus and the role imposed on host countries. For example, Dafermos et al. (2021) argue that beyond guarantees, the policy tools of host countries suggested by the US and the World Bank are limited to developing PPP infrastructure as an asset class, using the central bank to purchase climate-affected assets, regulating disclosure of climate-related financial risks, and implement carbon pricing. From that, Mustaq (2021) demonstrates how emerging countries shape bond markets to accommodate overseas private institutional investors. As an illustrative case, Dafermos et al. (2021) show that the Cameroonian government had to provide uptake, currency, and political risk guarantees to the World Bank to get financing for the Nachtigal Hydropower Project.⁴ Overarchingly, US ODA is focused on humanitarian and social development, while the World Bank is the US preferred institution for promoting infrastructure development in the global South. Accordingly, the US only mobilised between USD 4 and 7 billion of private finance annually since 2012 through the US Development Finance Corporation and US Agency for International Development, which are inconsequential amounts compared to the financing need (OECD 2024). As such, while some financing from the US Development Finance Corporation works as origin country de-risking, its limited scale makes it an insignificant part of the overarching host country de-risking approach predominantly promoted by the US.

While the PGII rhetorically promotes sustainability as part of infrastructure development, it argues that this is best achieved by the state using a limited number of tools to decrease projects’ risks. The model reinforces the existing dynamics of relations between states and the private sector in the global economy, largely accepting the status quo while promoting a slightly increased role of states in supporting infrastructure. From that, Bernards (2024) argues, that the model can be interpreted either as the interest of private finance or as states dealing with an inherent North–South imbalance in modern capitalism. In essence, I find that the US’s PGII model is well conceptualised by the Wall Street Consensus. This means that the model is largely aligned with the de-risking proposal from the global private financial sector and multilateral development banks, promoting a minimal role of the state by arguing that states should focus on using their financial resources to de-risk assets and leave the rest up to private actors.

The EU’s Global Gateway

The GG was launched on 1 December 2021, committing to mobilising EUR 300 billion by 2027 (EC 2021a). Similarly to the PGII, the EU’s GG is a continuation and expansion of the existing development model of the EU. As a clear expression of this, European Commission (EC) President von der Leyen states that ‘With Global Gateway, the European Union redoubles its efforts … ’ (EC 2022), while the EC Vice-President Borell states that the GG is ‘reaffirming our vision’ (EC 2021b). Analysts from the Center for Global Development confirm this, arguing that the GG is not ground-breaking but rather a packaging exercise of activities already planned (Gavas and Pleeck 2021). Consequently, though the initiative is new, it is possible to take previous efforts into account when assessing the GG model. In summary, the GG can be conceptualised as ‘ODA bargained de-risking’: It formalises and strategises development efforts, balancing and bargaining between the existing ODA focus with other dynamic foreign policy goals, placing de-risking responsibilities on both origin and host countries.

The GG has clear relations with both the PGII and the BRI. Regarding the first, EC President von der Leyen states that the GG and the B3W will mutually enforce each other (EC 2021b). Towards the BRI, EC President von der Leyen expresses that ‘Indeed, countries … need better and different offers [to China’s initiative],’ calling the GG a ‘true alternative’ (Strupczewski 2021). As another similarity with the PGII, the characteristics highlighted in the GG serve as a direct contrast to central criticisms of the BRI. These include high-quality, environmental, and social standards, democratic values, concessional financing, and a level playing field (EC 2021b). However, researchers argue that the GG is more than merely a response to the BRI but a reflection of a changing overarching foreign and development policy of the EU (Okano-Heijmans 2022). This trend confirms the literature’s argument of increasing alignment with EU interests and ODA policies (Carbone 2017).

Conceptualised in this article as ‘ODA bargained de-risking,’ the GG limits the range of state policies but allows substantial state backing from both host and origin countries. The approach of sharing the de-risking burden was expressed across several interviews with public officials and private financiers working on EU-led projects.⁵ The GG fundamentally agrees with the PGII in that the state should work to de-risk infrastructure assets but differs in three ways emphasised in both interviews and current literature: The substantial ODA element means that the grant element in de-risking is larger, leading to lower burdens for host countries by increasing de-risking by the origin state. For example, as the EU is the world’s by far largest ODA provider (OECD 2021), this is often channelled to the host country's fiscal resources, so they increase their fiscal capacity, which is required for many state tools like state-backed development banks, subsidies, and state-owned enterprises (SOEs). Second, The EU’s GG model is characterised by bargaining with host countries, which leaves more room for countries to implement developmentalist state policies as there is room and flexibility to tailor GG efforts to local needs and circumstances. Third, state-backed development finance institutions of origin and host countries play a larger role than in the PGII, such as by taking equity stakes rather than just de-risking. In this way, the model encourages the use of more tools than the PGII and allows for a larger role of the state across the given tools. The GG model allows a more active state, given its emphasis on fiscal policy as a driver as opposed to the market-emphasising neoliberal approach of the PGII.

At the core of how the EU’s model differs from the US, the GG has stronger concessional elements in overseas development finance (Tagliapietra 2021). Grants, favourable loans, and budgetary guarantees to de-risk assets and improve debt sustainability are explicitly promised (EC 2021b). The financing mechanism is phrased as a Team Europe approach, where EU bodies, member states, and development finance institutions mobilise private capital and coordinate projects in partner countries (EC 2021b). However, the GG provides limited new funding commitments, with most already committed before the official launch of the initiative. Still, the 3.4 rate at which the EU expects public capital to leverage private capital is a more achievable rate than the 15 rate for the Juncker Plan (Tagliapietra 2021). In comparison, MDBs claim to catalyse 2–5 dollars of private capital for every 1 dollar of their own investment (World Bank Group 2015).

Considering the budget numbers and phrasing in the GG's official documents compared with the EU’s current practice of ODA, it suggests that the GG will rely on concessional financing with a substantial grant component to de-risk projects to attract private capital (Mah 2023). For example, in 2020, the grant component of EU ODA loans totalled EUR 55 billion, making up 58 per cent of the world’s total. This is more than twice that of the US and equal to three times as high a proportion of GDP (0.5 per cent in the EU and 0.17 per cent in the US) (OECD 2021). However, despite large ODA amounts a gap between rhetoric and practice appears to exist. While interviewees from European development finance institutions emphasised how EU origin countries and EU organs play a role in de-risking infrastructure assets, interviewees in host countries expressed that they themselves still accounted for the majority of the burden and that EU countries should do more.⁶ Part of this discrepancy is likely to stem from the still limited scale of EU ODA compared to the global South infrastructure financing need.

The GG encourages some developmentalist policies while limiting others, focusing efforts around ODA and de-risking tools of both origin and host countries. As an example of ODA and direct financial support from origin countries, the Lake Turkana wind park in Kenya received equity and concessional debt financing from a long list of European development finance partners such as the Finnish and Danish governments and the EU Africa Infrastructure Trust Fund. At the host country side, the state-owned Kenya Power de-risks the project through a 20-year fixed price guarantee as part of the power purchase agreement (AfDB 2015). Further support from the EU and EU countries comes from export credit agencies and national development banks, with the logic of ODA to make projects feasible and profitable to private investors.⁷ As part of that, European countries provide ODA in the form of technical assistance to improve host countries’ state capacity to organise different aspects of development, such as in the case of the Danish Energy Authority. The explicit purpose of such ODA is to de-risk infrastructure projects, to encourage for-profit involvement of Western financial institutions and companies.⁸ Indeed, European investors expressed in interviews that such technical and financial support is critical for reducing their assessment of project risks and for providing the conditions under which they can scale up their equity and debt involvement.⁹ While emphasising the need for de-risking assets, these investors had no preference for whether this was done by the host country, origin country, or multilaterals.¹⁰ Consequently, global North investors are not neatly aligned with either the US or the EU, focusing merely on the assets being de-risked.

China’s Belt and Road Initiative

The BRI was launched by China’s President Xi Jinping in Kazakhstan in 2013. As with the PGII and GG, it is more a formalisation of existing efforts rather than a new policy initiative. Since the BRI’s official launch, the number of signatory countries has grown to 147, as measured by how many countries have signed a Memorandum of Understanding affirming their membership (Nedopil 2022b). Since then, the total value of BRI lending and investments has reached approximately USD 2 trillion (BU 2023, AEI 2024). However, since its peak in 2017, both lending and investments have fallen by at least 50 per cent (BU 2023, AEI 2024). From that, the BRI has evolved to focus on smaller and more targeted projects, which is an approach called ‘small is beautiful’ (Ray 2023). In spite of this emerging new approach, the financing amounts are substantially smaller. Hence, irrespective of the financing model, this means that China is now less of a scalable alternative to the US and EU. Indeed, across 100 interviews in recipient countries in the global South, no one expressed the opinion that China was systematically changing their financing options.¹¹ Therefore, China’s role today is shifting to be less a provider of capital and more a provider of cost-competitive construction and equipment for projects. To summarise, the BRI’s model can be conceptualised as ‘state-organised risk-sharing: Projects are developed through state-to-state coordination and implemented through Chinese state finance and insurance, with Chinese SOEs and private companies involved in technology, project development, and construction.

The BRI is often interpreted as an initiative launched to increase China’s role and influence in the world and to do so in a way that challenges Western dominance. As a concrete and often-studied example, the Asian Infrastructure Investment Bank directly overlaps with and challenges Western MDBs such as the World Bank and Asian Development Bank and was therefore strongly opposed by the US (Zhu 2019). However, in official rhetoric, the BRI is open for collaboration, as expressed by China’s Foreign Minister Wang Yi: ‘China is open to the United States participating in the Belt and Road Initiative. We are also willing to consider coordinating with the US Build Back Better World initiative to provide the world with more high-quality public goods.’ (Tian 2022). Still, in practice and as of today, Chinese BRI infrastructure projects are mostly state-to-state, coordinated, and implemented by Chinese organisations, with the private sector primarily involved as technology providers and sub-contractors.

There are two sides to de-risking within ‘state-organised risk-sharing’. The host country de-risks the assets since loans are given only to the state or state-backed organisations, and China de-risks by providing patient capital, few conditions, and standardised state-sponsored package deals. As China provides such package deals there is limited leeway for local tailoring (Liu and Lim 2022). Host countries can choose between project types, but if China is the financier, the project comes as a package of Chinese state-owned institutions with little room to choose international collaborators (Larsen and Oehler 2022). As part of that, the BRI uses bilateral state-to-state relations to develop projects (FT Editorial Board 2022).

Problematically, the reliance on state-owned institutions actively limits the private sector’s financial involvement, while still allowing the private sector to follow in the footsteps of financiers and play a role in equipment provision, construction, and sub-contracting. Even China’s recent use of PPPs in Africa entails a relationship between government and Chinese SOEs rather than private companies (Van Wieringen and Zajontz 2023). In this way, not leveraging private capital limits the total amount of projects that can be financed. This becomes problematic when an increasing number of global South countries struggle with over-indebtedness. Consequently, the model’s heavy reliance on state resources means that it cannot close the financing gap. The use of state-to-state coordination as well as state finance for infrastructure projects was voiced in interviews with government officials and project developers in Vietnam and Ethiopia as well as by representatives of Chinese construction companies operating abroad.¹²

In terms of financing models, the standard practice for BRI projects is to combine financing from several state-owned Chinese financial institutions. Loans from Chinese policy banks such as the Chinese Development Banks and commercial banks such as the Industrial and Commercial Bank of China, as well as project insurance from Sinosure.¹³ While the GG uses concessional financing to de-risk projects, Chinese financing is, in most cases, on commercial terms. For example, an average Western country loan to a global South country has an interest rate of 1 per cent compared to a Chinese loan of 4 per cent (Malik et al. 2021), leading analysts to conclude that Chinese lending is more likely to lead to over-indebtedness (Tagliapietra 2021). Furthermore, Chinese financing is characterised by longer time horizons (Kaplan 2018) and has less conditionality for the host country government (Caskey 2022). However, one conditionality often insisted on by Chinese lenders is that the loan terms or even the loan itself are kept confidential to the public (Gelpern et al. 2021). In this way, the Chinese government and financial institutions do not publicly disclose their project involvement or their financing terms. Instead, only annual totals are disclosed by the Chinese Ministry of Commerce (2022), with numbers often differing drastically from independent sources (Malik et al. 2021).¹⁴ While Chinese financing is the least transparent of the three initiatives, financing from US and EU state-organs are also critiqued for being opaque (Ingram and Paxton 2023). Lastly, BRI energy infrastructure projects are mainly in fossil fuels (Larsen and Oehler 2022) and apply host countries’ environmental regulations, which are substantially lower than the practice seen in MDBs (Voituriez et al. 2019).

The BRI limits private financial participation in projects and imposes limits on host countries in several central ways. For example, in a similar fashion to the PGII, China insists that host country governments rather than private companies act as recipients and guarantors of projects (Caskey 2022), adding risks and tying fiscal resources that could be used in other ways. Furthermore, China insists on confidentiality in contract terms (Gelpern et al. 2021), limiting the possibility of host countries implementing policies on mandatory environmental disclosure by financial institutions. In addition, as expressed by host country project developers and government officials, Chinese institutions have inadequate practices on environmental protection and civil society involvement,¹⁵ encouraging host countries not to increase such standards as part of industry regulations.¹⁶ For example, the Lamu coal-fired power plant project did not include sufficient environmental impact assessment or civil society consultation and was subsequently cancelled for these reasons (Shi 2021).

Discussion: stuck in the middle or a window of opportunity?

The three infrastructure financing models all impose a heavy financial de-risking burden on host countries. However, they do so in different ways, differing on key dimensions, as summarised in Table 1. Overall, greater similarities exist between the PGII and GG than between those and the BRI. This is to be expected given the similarities between Western countries and their formal support for each other’s initiatives (EC 2021b, White House 2021b). These similarities are also apparent in the role of the state in de-risking financial assets, as both the PGII and GG rely heavily on the state using certain financial tools to leverage the private sector in order to meet their total financing target. Within the de-risking approach, the clearest difference between the PGII and GG is the latter’s greater emphasis on grant elements that come from the integration with ODA efforts. In addition, the three initiatives differ in terms of scale. In 2020, the EU’s total ODA grant equivalent of EUR 73 billion was higher than all of China’s investment and construction in the BRI combined (OECD 2021, Nedopil 2024). If the GG allows the EU to direct more of this capital towards infrastructure and is able to catalyse the stated EUR 3.4 from the private sector for each EUR 1 of ODA, then the GG can further outsize the BRI. Though the tools and rate of leveraging private capital are unclear for the PGII, it may be similar to the EU, though the US’s commitment is two-thirds the size of the EU’s. Furthermore, given the BRI’s state-to-state model, it differs from the PGII and GG by being less welcoming to the private sector within financing, less consultative with stakeholders, and less transparent.

Table 1. Conceptualisation and comparison of the PGII, GG, and BRI.

	Partnership for Global Infrastructure and Investment	Global Gateway	Belt and Road Initiative
Role of origin state	Coordination of private actors	De-risking	Coordination of complete package deal
Role of host state	De-risking	Coordination and de-risking	Coordination and political support
Role of private sector	Almost all of capital	Majority of capital	Limited capital, substantial contracting and technology
Financial tools	De-risking tools	Aid and de-risking tools	Commercial terms on debt and equity
ODA financing	Limited	Large	Limited
International collaboration	Emphasised	Emphasised	Limited, so far
Stakeholders	Government, private, IO, NGO	Government, private, IO, NGO	State-to-state, including SOEs, private sub-contracting
Transparency	Medium	Medium	Low
Green transition	Increasingly emphasised	Emphasised	In rhetoric, not practice
Conceptualisation	Host country de-risking	ODA bargained de-risking	State-organised risk sharing

Here, we can return to the dilemma presented at the outset of this paper: Is there any manoeuvrability left for a green state approach in global South countries within the limitations of the infrastructure financing model? I find that host countries have agency to selectively engage with each initiative in ways that increase their manoeuvrability and allow them to more actively use the state apparatus. This finding supports previous arguments questioning the severity of limitations (Alami et al. 2021, O’Sullivan and Rethel 2023, Bernards 2024) and recognising agency of host countries (Chiyemura et al. 2023, Gabor and Sylla 2023, Schindler et al. 2023, Van Wieringen and Zajontz 2023). I find that selective engagement is possible given that neither model is hegemonic and since the three models can be combined in single countries and even projects.

As neither model is hegemonic, host countries may be able to pick and choose their engagement with each model. Essentially, the models each assign a specific role to host states that shape their ability to use developmentalist state tools. Regarding the US, Gabor (2021) argues that the PGII limits host countries to only providing guarantees and other types of de-risking. The findings of this article support this argument but find that the GG’s ODA focus may provide a number of such de-risking mechanisms through the origin rather than the host country side. For example, government officials organising Danish ODA explained in interviews that a core purpose of climate-related ODA is to help host countries de-risk infrastructure projects to attract global capital.¹⁷ Furthermore, while the BRI cannot provide sufficient financing, Chinese companies have cost-competitive construction and equipment and have great ambition to expand overseas.

By strategically engaging with each model, it may be possible for host countries to maintain some of the policy manoeuvrability required for an expansive state approach. It may be possible for host countries to provide a guarantee to de-risk a project to work with the US to mobilise private capital, receive concessional financing for a project from the EU, and take a sovereign-backed loan from China to build a Chinese SOE-led project. This can even be combined within a single project. However, interviews with global South countries’ government officials suggested that using Chinese construction and technology companies carries greater security risks as such companies have closer ties with the Chinese state than their Western counterparts.¹⁸ This situation supports Kring and Gallagher’s (2019) argument that China’s development finance simultaneously brings additional resources and tensions.

Still, recent history suggests that it is possible to design projects in ways that allow the three models to support the same project. For example, the world’s largest solar farm is being built in the United Arab Emirates in a consortium led by the French SOE, EDF, and the Chinese private solar panel company, Jinko Solar (EDF 2020). Even without funding from the EU itself, this project is within the GG and counts towards its financial commitment through its French component. Another example can be seen in the Peljesac Bridge in Croatia, which received EUR 357 million of its total of EUR 420 million from the EU, while the construction tender went to the China Road and Bridge Corporation. This was the first time a Chinese company won a bid for an EU-funded project, suggesting that more projects at the GG and BRI intersection are possible. Furthermore, research shows that about 30 per cent of the World Bank’s procurement contracts for the construction of infrastructure in Africa are awarded to Chinese companies (Zhang 2021).

These cases demonstrate that the models are not mutually exclusive and that host countries can take advantage of promoting projects where they overlap. This could include using US involvement to catalyse private capital, using EU involvement to access development finance, and Chinese involvement to acquire low-cost infrastructure construction. It suggests that global South countries may be able to retain manoeuvrability by only selectively and strategically engaging with each model. A research agenda is emerging on this issue that requires further scholarship (Colgan and Miller 2022, Gabor and Sylla 2023). Interviews also suggest that host countries are aware of this and intentional about how to engage strategically, as summarised by an official from the Ethiopian Ministry of Finance ‘ African countries are trying to play the US and EU and China against each other. The US thinks we are dumb. We are poor, but we are not dumb.’¹⁹

Conclusion

In this article, I have shown that the three infrastructure finance models of the US, EU, and China differ, but all entail a heavy financial debt burden for host countries. The US does so through ‘host country de-risking, the EU by ‘ODA bargained de-risking’ and China as ‘state-organised risk-sharing’. Relating this to the question in the title of the paper, there is indeed no alternative to the financial risk burden of the Wall Street Consensus amongst the three initiatives, as the EU and China also impose large burdens. These models all limit the manoeuvrability of host countries to use developmentalist tools. Still, in practice, host countries may have agency to pick and choose between the models on a project basis which reduces the severity of each model’s individual constraints. To move this research agenda forward, future scholarship can draw from these findings to better grasp the heterogeneity of financing options and adequately capture the agency of host countries in manoeuvring between the financing models.

Acknowledgements

I wish to thank the reviewers for their constructive and engaging input for improving the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Mathias Larsen

Mathias Larsen is a postdoctoral research associate at Brown University’s Watson Institute for International & Public Affairs. His research concerns the political economy of the role of the state in financing green transition, focusing on the case of China and other global South countries.

Notes

1 Interviews focused on the World Bank’s role in Vietnam, in particular with regard to reforming energy markets and the state-owned monopoly company Vietnam Electricity (EVN). 2022.

2 Interviews were conducted in Vietnam in 2022 and India in 2023

3 Interviews numerous equity funds and institutional investors from the EU and the US. 2021–3.

4 See Gabor (2021) Dafermos et al. (2021) and Gabor & Sylla (2020) for an extensive list of cases.

5 Interviews covered Ministries of Foreign Affairs, export credit agencies, private equity funds, institutional investors, and project developers. 2021–3.

6 Interviews with European export credit agencies and ODA agencies, as well as with host country government and project developers. 2022–3.

7 Interviews with European export credit agencies and ODA agencies. 2022.

8 Interviews with the Danish Energy Authority in Vietnam, India, and Denmark. 2022–3

9 Interviews with European investors active in global South countries. 2022.

10 Interviews with European investors active in global South countries. 2022.

11 Interviews with host country actors across all countries of fieldwork. 2022-3.

12 Interviews were conducted in Ho Chi Minh City and Hanoi, Vietnam. 2022.

13 See Chin and Gallagher (2019) for details on the roles of different Chinese organisations involved in overseas project finance

14 See Jepson (2021) for an assessment of the accuracy of official and unofficial accounts of Chinese overseas lending.

15 See Larsen et al. (2023) for an account of sustainability practice by Chinese sovereign backed development funds.

16 Interviews were conducted in Vietnam and India. 2022–3.

17 Interviews with the Danish Energy Authority in Vietnam, India, and Denmark. 2022–3

18 This was expressed by both government officials, financiers, and project developers in Vietnam and India. 2022-3.

19 Interview with official from the Ethiopian Ministry of Finance. 2023.