Development and Sustainability in Economics and Finance 8 (2025) 100096

Contents lists available at ScienceDirect

Development and Sustainability in Economics and Finance

journal homepage: www.sciencedirect.com/journal/dsef

Climate finance and infrastructure development in Africa: Direct effects and

transmission mechanisms
Bruno Emmanuel Ongo Nkoa a,* , Pascal Ndyanabo Rutazihana b,c,** ,
Arsene Mouongue Kelly b,*
a
Faculty of Economics and Management, University of Yaoundé II, Yaoundé, Cameroon
b
Dschang School of Economics and Management (DSEM), University of Dschang, Dschang, Cameroon
c
Université Catholique la Sapientia de Goma, Goma, République Démocratique du Congo

A R T I C L E I N F O A B S T R A C T

JEL Classifications: Climate finance has emerged as a critical instrument for advancing the Sustainable Development Goals,
C33 particularly in Africa where infrastructure deficits remain acute, climate vulnerabilities are severe, and the de­
H54 mand for resilient development is urgent. While existing studies have examined the broader developmental
O55
impacts of climate finance, little is known about its specific influence on infrastructure outcomes across African
Q54
economies, leaving a significant gap in both empirical evidence and policy guidance. In an attempt to fill this gap
Keywords:
and provide actionable solutions, the present study investigates the effect of climate finance on infrastructure
Climate Finance
Infrastructure Development
development in Africa. Specifically, it seeks to examine the direct effect of climate finance on infrastructure,
Sustainable Development assess whether adaptation and mitigation finance exert differential impacts, and explore the role of human
IV-GMM capital and technological innovation as key transmission channels. Using a panel dataset of 50 African countries
Africa over the period 2012–2021, the analysis applies Ordinary Least Squares and Instrumental Variable Generalized
Method of Moments estimation techniques to address econometric concerns and ensure robust results. The
findings reveal that climate finance significantly enhances infrastructure development, with the effect further
amplified through human capital and technological innovation, and the results remain consistent under alter­
native model specifications. Based on these insights, the study offers targeted policy recommendations for
optimizing the allocation and strategic use of climate funds to address Africa’s infrastructure deficits and advance
sustainable development goals.

1. Introduction indispensable for human development (Tongson and Caelian, 2024;

Wassou et al., 2024). Yet, Africa continues to face chronic infrastructure
Infrastructure is a fundamental pillar of economic systems and plays deficits, reflected in limited access to electricity, underdeveloped
a determining role in growth trajectories and the well-being of society. It transport corridors, and poor water and sanitation systems, which
underpins the efficient operation of businesses, connects markets, and significantly constrain economic competitiveness and social inclusion.
improves productivity (Ketu and Tchapchet-Tchouto, 2025). For At the same time, the continent is highly vulnerable to climate change,
example, transport networks, such as roads, railways, and ports, enable with rising temperatures, erratic rainfall, and extreme weather events
the movement of goods and services, reducing costs and increasing threatening existing infrastructure and increasing the urgency for
economic competitiveness (Foster et al., 2023). Similarly, energy climate-resilient investments (Timilsina et al., 2024). It is in this context
infrastructure, including electricity generation and distribution systems, that climate finance becomes a critical tool, as it provides external re­
is essential to industrial operations and economic growth (Thacker et al., sources targeted at supporting adaptation and mitigation measures,
2019). Infrastructure also plays an essential role in improving the thereby enabling African countries to build climate-resilient and sus­
quality of life and social well-being of communities by providing access tainable infrastructure. Despite the growing discourse on climate
to healthcare, education, drinking water, and sanitation, which are finance and its role in advancing sustainable development, empirical

* Corresponding authors.
** Corresponding author at: Dschang School of Economics and Management (DSEM), University of Dschang, Dschang, Cameroon.
E-mail addresses: [email protected] (B.E. Ongo Nkoa), [email protected] (P.N. Rutazihana), [email protected] (A.M. Kelly).

https://doi.org/10.1016/j.dsef.2025.100096
Received 7 May 2025; Received in revised form 25 August 2025; Accepted 23 September 2025
Available online 24 September 2025
2950-5240/© 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

evidence on its effectiveness in addressing Africa’s infrastructure gaps and development, particularly within the African context. First, it ex­
remains limited, particularly in terms of understanding the channels tends the empirical discourse by explicitly linking climate finance to
through which climate finance translates into tangible infrastructure infrastructure development in Africa, a connection that remains
outcomes. This makes it imperative to examine the nexus between underexplored in existing research. While previous studies, such as Kelly
climate finance and infrastructure development, not only to assess the (2024) have focused on the role of climate finance in enhancing food
effectiveness of financial flows but also to identify pathways for security, emphasizing the agricultural and environmental dimensions of
leveraging them in ways that simultaneously enhance economic growth, adaptation and mitigation financing, this study shifts the analytical lens
social well-being, and environmental sustainability across the continent. to physical infrastructure, including transport, energy, and ICT systems,
However, despite a broad consensus on the centrality of infrastruc­ which are essential for both economic growth and climate resilience.
ture, developing countries, particularly Africa, face serious qualitative This distinction is critical, as infrastructure not only serves as a foun­
and quantitative infrastructure deficiencies (Ketu et al., 2022). For dation for productive activities but also determines a country’s capacity
example, transport costs account for 30–40 % of the price of goods in the to absorb and respond to climate shocks. Second, by employing rigorous
region, up to four times more than in developed countries, due to the econometric strategies, specifically, the Generalized Method of Mo­
poor quality and inefficiency of roads, with only 25 % paved (Jedwab ments with Instrumental Variables (IV-GMM) and quantile regression,
and Storeygard, 2019). Access to energy remains very limited, with only the study addresses endogeneity and cross-country heterogeneity,
43 % of the population having electricity in 2021, far behind regions thereby ensuring credible causal inference and revealing how climate
such as North America (99 %) and Southeast Asia (96 %), with rural finance impacts infrastructure outcomes at different levels of develop­
areas disproportionately affected, with an access rate of 25 % (IEA, ment. This approach moves beyond average effects and allows for a
2022). Similarly, only 58 % of the population has access to safely more refined understanding of how the benefits of climate finance are
managed drinking water, a very low percentage compared with other distributed across countries with varying infrastructure capacities.
regions such as Latin America (95 %), which further exacerbates health Third, the study innovatively disaggregates climate finance into adap­
and economic problems (MacDonald et al., 2021). tation climate finance (ACF) and mitigation climate finance (MCF), a
As Africa continues to face these major deficits, the continent needs distinction often overlooked in the literature. By doing so, it offers tar­
approximately 130–170 billion euros per year to fill these gaps, which geted policy insights on the relative effectiveness of different financing
are exacerbated by rapid urbanization and demographic change (Effiom types, enabling policymakers to tailor interventions based on specific
and Agala, 2020). Despite various initiatives, financial constraints and infrastructure needs. Lastly, the study introduces a contextual dimension
ineffective implementation, and environmental concerns continue to by examining the moderating roles of human capital and technological
limit infrastructure expansion in Africa (Haryani and Anjani, 2023). For innovation. These structural factors are shown to significantly influence
example, public budgets, often constrained by debt and competing pri­ the efficacy of climate finance, highlighting the importance of comple­
orities, are insufficient to meet the necessary investments (Estache, mentary investments in education and innovation ecosystems to maxi­
2010). Also, in many countries, public spending is already under pres­ mize development outcomes.
sure due to urgent social needs, making it difficult to allocate sufficient The remainder of the paper is structured as follows: Section 2 reviews
resources to infrastructure (Calderón and Servén, 2010). For their part, the literature on climate finance and infrastructure development. Sec­
private investors often favor high-profitability sectors, leaving strategic tion 3 presents the methodology, data sources, and empirical models.
areas such as water and sanitation on the sidelines, as they are consid­ Section 4 presents the results, including analyses of heterogeneity across
ered less attractive due to their low return on investment (Inderst, infrastructure groups and sectors. Section 5 concludes with policy
2013), as is also the case for climate-resilient infrastructures with low recommendations.
profitability and often long-term prospects. As a result, the infrastruc­
ture deficit persists. 2. Literature review
Faced with these challenges, climate finance represents an oppor­
tunity that can be exploited to bridge the financing gap for resilient and 2.1. Theoretical framework
sustainable infrastructure in Africa. It is understood as any local, na­
tional, or transnational financing, from public, private, and alternative The interaction between climate finance and infrastructure devel­
sources, that seeks to support mitigation and adaptation actions that will opment can be analyzed through environmental economics theory and
address climate change (Kelly, 2025; Newell, 2025). Financial in­ systems theory. Environmental economics provides a fundamental
struments such as green bonds and climate investment funds can offer theoretical framework for understanding how climate finance can cor­
innovative solutions for attracting capital to environmentally friendly rect the market failures associated with environmental externalities
infrastructure projects (Inderst, 2013; Feuzeu and Kelly, 2025). Existing (Pigou, 1920). In the absence of regulatory or market-based in­
literature has examined the role of climate finance in global sustain­ terventions, the social costs of greenhouse gas emissions - such as
ability (Betzold and Weiler, 2017; Kelly and Ndeffo, 2025) and its increased frequency of extreme weather events, biodiversity loss, and
allocation efficiency (Pickering et al., 2017), but few studies disaggre­ climate-induced economic shocks - are not reflected in market prices,
gate its impact on infrastructure sectors. While previous research has leading to underinvestment in low-carbon infrastructure (Hallegatte
acknowledged the potential of instruments such as green bonds and et al., 2016). Climate finance instruments aim to internalize these ex­
multilateral climate funds (Kumar, 2022), there remains limited ternalities by channeling capital flows towards sustainable and
empirical understanding of how various forms of climate finance, spe­ climate-resilient infrastructure systems. Mechanisms such as green
cifically global climate finance (GCF), adaptation climate finance (ACF), bonds, public-private partnerships (PPPs), and carbon pricing create
and mitigation climate finance (MCF), influence infrastructure out­ financial incentives for the deployment of renewable energy networks,
comes across Africa’s diverse economic contexts. The present study thus climate-resilient transport systems, and energy-efficient urban infra­
sets as objective to investigate the effect of climate finance on infra­ structure (Tol, 2002; Batten et al., 2020). By setting an appropriate price
structure development in Africa so as to address these gaps in the for carbon and climate risks, these instruments redirect capital markets
existing literature. By examining both the direct effects and potential towards sustainable investments that not only produce economic returns
transmission channels, this study seeks to provide a comprehensive but also generate positive environmental externalities. Edenhofer
analysis of the role climate finance plays in shaping infrastructure (2015) argue that these mechanisms contribute to intertemporal effi­
development on the continent. ciency by reducing long-term environmental liabilities and aligning
In filling the aforementioned gaps, the present study makes several infrastructure development with intergenerational well-being. The
important and novel contributions to the literature on climate finance deployment of solar energy in developing countries is a concrete

2
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

illustration: targeted subsidies and concessional financing reduce high only directly supports renewable energy infrastructure but also am­
upfront investment costs, accelerating the transition from fossil plifies the impact of AI through technological and innovation channels.
fuel-dependent systems to renewable alternatives (Buchner et al., 2014). This highlights climate finance’s instrumental role in unlocking infra­
These fiscal and financial tools are particularly relevant for African structure co-benefits across sectors. Complementing this, Zhao et al.
economies, where high risk premiums and capital scarcity often (2025) examine the influence of climate finance on electricity accessi­
discourage long-term investment in infrastructure. Climate finance, bility and urban–rural electricity inequality using data from 125 coun­
therefore, helps to ensure that infrastructure not only meets immediate tries between 2002 and 2020. Their results show that climate finance
development needs but also mitigates exposure to future climate risks, significantly improves electricity infrastructure, especially in
thereby building resilience, reducing life-cycle costs, and improving the low-income countries, and contributes to narrowing spatial disparities
overall efficiency of resource allocation. in energy access. These findings reinforce the view that climate finance
From a systems theory analytical perspective, climate finance plays a can serve as a direct and inclusive driver of infrastructure development
central role in enhancing the resilience of interdependent infrastructure by targeting systemic vulnerabilities in energy and utility systems.
networks. Systems theory conceives of infrastructure not as isolated Synthesizing these insights, we propose the following hypothesis:
units, but as dynamic, interconnected systems where disturbances at one
Hypothesis 1. Climate finance has a direct positive effect on infrastructure
node (e.g., power grid) can propagate to other nodes (e.g., transport,
development.
water) - a phenomenon known as “cascading failure” (Rinaldi et al.,
2001; Markolf et al., 2018). Modern critical infrastructure systems -
2.2.2. Indirect effect
energy, water, transport - operate as coupled socio-technical systems,
making them sensitive to climate-induced disruptions, such as heat
− Human capital channel
waves, floods, or forest fires. These stressors can trigger feedback loops
across networks (for example, flood-damaged roads preventing power
Human capital is essential in the relationship between climate
lines from being repaired), exacerbating systemic risk. Climate finance,
finance and infrastructure development, acting as a critical transmission
when allocated to adaptation-oriented investments, injects resilience by
channel that enhances the effectiveness of financial interventions to
enhancing the modularity, redundancy, and flexibility of these in­
build climate-resilient infrastructure. Human capital, which encom­
frastructures (Chelleri et al., 2015). For example, funding decentralized
passes the knowledge, skills, and health of engineers, planners, con­
renewable energy microgrids mitigates dependence on centralized
struction workers, and policy makers, is essential for designing,
power plants, decoupling local risk from systemic grid failure. Similarly,
implementing, and maintaining sustainable infrastructure systems.
natural flood management interventions (e.g., wetland restoration)
When climate finance is directed towards building human capital
mitigate the intensity of shocks before they reach built systems, pro­
through education, technical training, and healthcare, the workforce is
tecting infrastructure from abrupt climate extremes.
better equipped to adopt innovative technologies, apply green stan­
dards, and manage complex projects (Acemoglu and Autor, 2012). The
2.2. Related link and hypothesis construction
result is more sustainable and adaptable infrastructure, such as
flood-resistant transport networks or renewable energy networks, which
2.2.1. Direct effect
are essential for climate resilience. Investment in human capital ensures
In this subsection, we assess the arguments regarding the direct effect
that climate finance is used effectively. Skilled professionals can opti­
of climate finance on infrastructure development. First, drawing on
mize resource allocation, reduce construction delays, and mitigate the
endogenous growth theory (Romer, 1990; Howitt and Aghion, 1998),
risk of cost overruns in infrastructure projects (Hallegatte et al., 2019).
climate finance is theorized to alleviate capital constraints that prevent
Similarly, health initiatives that reduce worker absenteeism due to
the deployment of climate-resilient infrastructure, particularly in
climate-related illnesses (e.g., heat stress) improve productivity in
developing economies where public budgets and private investment are
labor-intensive infrastructure sectors (ILO., 2023; Wassou et al., 2025).
insufficient to meet adaptation and mitigation needs. Romer’s frame­
Human capital also promotes equitable outcomes in infrastructure. Also,
work posits that targeted financial flows can stimulate technology
climate finance that prioritizes gender- and youth-sensitive training
spillovers and economies of scale in infrastructure sectors by reducing
programs ensures that marginalized groups, such as women in rural
the marginal cost of adopting low-carbon technologies. This is consistent
areas, acquire the skills needed to participate in infrastructure projects.
with the empirical analyses of Hallegatte et al. (2016), which show that
This allows us to formulate the following hypothesis:
climate finance - in particular concessional loans and grants from
multilateral institutions - significantly increases investment in renew­ Hypothesis 2. Climate Finance will lead to an increase in the level of
able energy networks and flood-resistant transport systems in infrastructure by encouraging the development of human capital.
low-income countries, where market failures in long-term infrastructure
finance are severe. Similarly, Agrawala et al. (2020) find that climate − Technological innovation channel
finance disbursed through bilateral channels correlates with improved
Technological innovation serves as an essential transmission channel
infrastructure resilience in small island developing states (SIDS), as
between climate finance and infrastructure development, catalyzing the
external finance offsets national budget deficits and encourages
creation, dissemination, and adoption of advanced solutions to address
public-private partnerships. By addressing liquidity constraints and
systemic climate vulnerabilities. Climate finance directly stimulates
reducing project risk, climate finance directly increases the stock of
innovation cycles by funding research and development (R&D) of
infrastructure in sectors where high upfront costs deter private capital
technologies designed to mitigate infrastructure risks - such as extreme
(Meltzer, 2016). However, the effectiveness of these flows depends on
weather resilience and carbon footprint reduction - while bridging the
their alignment with national regulatory frameworks; for example,
‘valley of death’ between experimental phases and commercialization
Pauw et al. (2022) show that climate finance fails to catalyze infra­
(OCDE, 2022; Hallegatte et al., 2019). Public and multilateral climate
structure development in countries with fragmented energy policies, as
funds de-risk innovations at an early stage, enabling private sector
misaligned incentives discourage coordinated investment.
engagement and accelerating the translation of theoretical solutions into
Further support for the direct role of climate finance in infrastructure
scalable systems. This process is facilitated by institutional frameworks
development comes from recent global empirical evidence. Zhao et al.
that prioritize intellectual property agreements, technology transfer
(2024) investigate the role of artificial intelligence (AI) in advancing
mechanisms, and public-private R&D partnerships, ensuring that in­
renewable energy development using panel data from 63 countries be­
novations are aligned with infrastructure needs (IPCC, 2022). The
tween 2000 and 2019. Their findings reveal that climate finance not

3
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

scalability of technological innovations, due to economies of scale, four disaggregated components including: Electricity infrastructure,
further amplifies the impact of climate financing. As technologies quantified by total national electricity generation (including public,
mature - such as solar photovoltaic (PV) or smart grid systems - their unit private and imported sources); Transport infrastructure, measured by
costs decrease, improving affordability for developing economies the total road network (in kilometres) and the proportion of paved
(Zhang et al., 2022). This cost reduction allows governments to upgrade roads; ICT infrastructure, measured by internet penetration (number of
ageing infrastructure or deploy new systems aligned with global climate users per 100 inhabitants) and telephone subscription density; and
goals. At the same time, climate-resilient innovations, such as water Water and sanitation infrastructure, approximated by access to
networks managed by artificial intelligence, generate positive feedback improved sanitation facilities and population coverage of drinking water
loops by reducing long-term maintenance costs and avoiding sources (Cerra et al., 2017; Ningaye and Ketu, 2023). By integrating both
climate-related losses. The World Bank (2023) estimates that every aggregate and disaggregate measures, the analysis captures systemic
dollar invested in resilient infrastructure leads to a fourfold increase in interdependencies while addressing criticisms of oversimplification
avoided damage, providing an incentive to increase climate finance and raised in previous literature.
build systemic resilience. On this basis, we formulate the following To capture our variable of interest, namely climate finance, this
hypothesis: study adopts a tripartite framework aligned with recent prior work (Lee
et al., 2022; Zhao et al., 2022; Kelly, 2024): (i) Global Climate Finance
Hypothesis 3. Climate finance improves infrastructure levels by boosting
(GCF) refers to multilateral financing mechanisms, such as those of the
technological innovation in Africa.
UNFCCC Green Climate Fund, that pool and allocate resources to miti­
However, while recognizing that existing research on finance and gation and adaptation projects in developing economies; (ii) Climate
infrastructure provides valuable insights into the role of financial access Change Mitigation Finance (MCF) encompasses financial flows dedi­
in improving infrastructure levels, there is limited exploration of how cated to reducing greenhouse gas emissions, including investments in
climate finance specifically affects infrastructure development outcomes renewable energy systems, carbon capture technologies and energy ef­
in the African context. Existing research has mainly focused on broader ficiency improvements; and (iii) Adaptation Climate Finance (ACF)
financial mechanisms such as microfinance, foreign direct investment, targets initiatives that build resilience to climate impacts, such as coastal
financial markets, and inclusive digital finance in various global con­ protection infrastructure, climate-smart agriculture and disaster risk
texts (Luu et al., 2024). Less attention has been paid to the unique management systems, as shown by empirical studies on the effectiveness
challenges and potentials of climate finance, which is playing an of finance (Falconer and Stadelmann, 2014; Pauw et al., 2022). The
increasingly central role in addressing the effects of climate change on relevant data is taken from the OECD’s Development Assistance Com­
infrastructural systems. Also, previous studies increasingly focus on the mittee (DAC) database, a standardized repository for bilateral and
contribution of infrastructure to economic growth without examining its multilateral climate-related official development assistance (ODA).
crucial role in adapting to climate shocks (Etensa et al., 2022; Taufik and To enhance the reliability of the estimation results and mitigate the
Markhamah, 2024). Given the continent’s climate vulnerabilities, a risk of omitted variable bias, this study carefully selects control variables
more contextualized approach is needed to align infrastructure devel­ based on a review of the relevant empirical literature on the de­
opment with the imperatives of adaptation and climate resilience. While terminants of infrastructure development (Kumar et al., 2022; Calderon
building on previous work, this study therefore fills these gaps in the et al., 2018; Brichetti et al., 2021; Cerra et al., 2017). The baseline model
literature by examining the direct and indirect effects of climate finance incorporates five core control variables that capture macroeconomic,
on infrastructure development in the African context in order to develop environmental, and institutional dimensions: (i) financial development
targeted policies for interventions that effectively improve the quantity (logFD), which influences infrastructure investment through improved
and quality of infrastructure in the current context of climate change access to credit and capital markets; (ii) CO₂ emissions (lnCO₂), serving
resilience in Africa. as a proxy for economic activity and energy-intensive infrastructure; (iii)
natural resource depletion (logNRD), which may reflect both the
3. Methodology financing capacity from resource rents and the need for compensatory
infrastructure; (iv) population growth (logPop), which can exert pres­
3.1. Data sure on existing infrastructure and reduce per capita availability; and (v)
control of corruption, an institutional factor affecting the efficiency and
The objective of this paper is to analyze the effect of climate finance effectiveness of infrastructure delivery. In line with theoretical expec­
on infrastructure development in a panel of 50 African countries3 over tations and empirical findings, financial development, natural resource
the period 2012–2021. The choice of countries and the study period are exploitation, emissions, and stronger institutional quality are expected
based on the availability of data, in particular those used to capture to positively correlate with infrastructure development, while rapid
climate finance. population growth may undermine it. To further ensure robustness and
The dependent variable of this study is infrastructure development, capture alternative explanatory pathways, we augment the model with
operationalized by the African Infrastructure Development Index (AIDI), four supplementary control variables: (i) urbanization (logUrban),
a multidimensional measure compiled by the African Development reflecting demand for urban infrastructure; (ii) GDP per capita
Bank. The existing literature shows the heterogeneity of conceptual (logGDP), capturing overall economic capacity and development level;
frameworks for defining infrastructure: while some studies adopt a ho­ (iii) industrial development (logIND), indicative of structural trans­
listic approach, others isolate specific sectors (e.g., transport) without formation and infrastructure needs; and (iv) government public expen­
taking into account the interdependencies between types of infrastruc­ diture (logGOV), representing fiscal commitment to infrastructure
ture (Agénor and Moreno-Dodson, 2006). Relying mainly on physical investment.
indicators of public infrastructure further complicates comparability Preliminary results, illustrated in Fig. 1, suggest a positive correla­
between studies. Agénor and Moreno-Dodson (2006) and Seethepalli tion between climate finance and infrastructure development in aggre­
et al. (2008) propose a global definition encompassing transport, water gate, but also with its four composite indices, including transport,
and sanitation, information and communication technologies (ICTs), electricity, water and sanitation, and digital infrastructure (ICT). This
and energy systems. In order to address methodological pluralism and link is empirically examined in more detail in the following sections of
improve robustness, this study uses not only the composite AIDI, but also the study. The summary statistics in Table 1 show that the first two
moments of the variables in our model fall within an appropriate range
for comparability. In addition, Table 2 provides the correlation matrix
3
See Appendix for complete list of Countries results, which indicate a positive correlation between climate finance

4
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

Fig. 1. Climate Finance and Infrastructure development.

Source: Author’s construction.

Table 1
Variables used and summary statistics.
Variable Description Obs Mean Std. Dev. Min Max Source

logAIDI Composite index of infrastructure development and its four subcomponents. 500 2.943 .683 1.028 4.486 AfDB (2024)
logGCF Global Climate finance (constant 2019 US$) 433 8.032 1.193 4.478 13.442 OCDE-DAC (2024)
logMCF Mitigation Climate finance (constant 2019 US$ $) 411 7.185 1.603 1.348 12.684 OCDE-DAC (2024)
logACF Adaptation Climate finance (constant 2019 US$) 403 7.601 1.106 3.786 12.977 OCDE-DAC (2024)
logFD Domestic credit to private sector (% of GDP) 455 2.776 1.435 − 5.793 4.859 WDI (2024)
logCO2 Carbon dioxide (CO2) emissions (total) 439 − 2.113 0.599 − 3.671 − .466 WDI (2024)
logNRD Natural resources depletion (% of GNI) 488 1.045 1.764 − 6.701 3.499 WDI (2024)
logPop Population growth (annual %) 499 0.732 0.664 − 6.079 1.366 WDI (2024)
CC Control of corruption: estimate 500 − 0.652 0.601 − 1.795 1.017 WGI (2024)
logUrban Urban population (% of total population) 500 3.7 0.446 2.415 4.504 WDI (2024)
logGDP GDP per capita growth (annual %) 489 7.255 0.881 5.379 9.363 WDI (2024)
logIND Manufacturing. value added (% of GDP) 427 2.236 0.623 − 0.182 3.631 WDI (2024)
logGov Government final consumption expenditure (% of GDP) 436 2.599 0.459 1.277 3.703 WDI (2024)

Source: Author’s Construction

and infrastructure development and establish the absence of multi­ economics and infrastructure finance, which posit that access to
collinearity in the model. adequate and sustained financing is a key determinant of infrastructure
Legend: Log: Logarithm. WDI: World Development Indicators. growth, particularly in developing countries. Climate finance, both in
OECD-DAC: Development Assistance Committee of the Organization for the form of mitigation and adaptation flows, has emerged as a critical
Economic Co-operation and Development (OECD). channel for financing infrastructure that is not only resilient to climate
risks but also conducive to sustainable development (Bhattacharya et al.,
2012; Bowen, 2016). From a theoretical perspective, the role of climate
3.2. Empirical model finance in infrastructure development is supported by the blended
finance framework and the climate-compatible development paradigm,
This research aims to provide an analysis of the effect of climate which emphasize the mobilization of public and private climate-related
finance on infrastructure development. The empirical model in this funds to close the infrastructure financing gap while promoting
study is motivated by the theoretical underpinnings of development

5
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

Table 2
Matrix of correlations.
Variables (1) (2) (3) (4) (5) (6) (7) (8) (9)
(1) logAIDI 1.000
(2) LogGCF 0.150 1.000
(3) logMCF 0.212 0.775 1.000
(4) logAFC 0.094 0.738 0.558 1.000
(5) logFD 0.461 0.113 0.119 0.041 1.000
(6) logCO2 0.529 − 0.006 0.062 − 0.134 0.331 1.000
(7) logNRD 0.253 − 0.020 − 0.038 0.022 − 0.279 − 0.209 1.000
(8) logPop − 0.580 0.064 − 0.033 0.209 − 0.296 − 0.359 0.439 1.000
(9) CC 0.483 0.010 0.112 − 0.077 0.289 0.327 − 0.329 − 0.424 1.000

Source: Author’s Construction.

long-term environmental and economic co-benefits (Buchner et al., the extent of the bias induced by the omission of variables: the
2011; Rutazihana and Kelly, 2025). Therefore, previous empirical and method introduced by Altonji et al. (2005), known as AET, and its
theoretical research is used as a base for the specification of the extension developed by Oster (2019). Altonji et al. (2005) proposed
empirical functional benchmark model (Cerra et al., 2017; Calderon an approach for assessing the robustness of results by postulating
et al., 2018). As shown in Eq. (1) (X representing the set of control that the relationship between the treatment (i.e. variable X) and the
variables), we investigate the following relationship: unobservable variables can be inferred from the observed relation­
ship between the treatment and the measurable variables. Their
fInfrastructure development it = (Climate financeit , Xit ) (1)
method is based on the hypothesis that if all the unobservable vari­
The literature indicates that empirical data obtained by a log-linear ables were included in the model, the variance of the outcome
specification yields more reliable and efficient results than a simple studied would be fully explained, corresponding to a coefficient of
linear definition (Kelly et al., 2024). Thus, in order to reduce the vari­ determination (R2) equal to 1. This approach is based on the idea that
ability of the data, which can result in a distribution of the error term rigorous consideration of the observable variables makes it possible
that violates the normality assumption, a prerequisite for the OLS to obtain an estimate of the extent of the bias due to the unobserved
regression method, we converted the variables to natural logarithms. variables. It is particularly relevant when no prior information is
The coefficients are then treated as elasticities, as shown in Eq. (2): available to establish the exogeneity of the variable of interest, since
it provides a quantitative assessment of the bias associated with
logAIDIit = β0 + β1 logCFit + β2 logXit + εit (2)
omissions.
Oster (2019) has refined this methodology by relaxing certain
Where index i denotes the country and t the year, β0 is the constant, β1
restrictive assumptions. The AET method is based on a null hy­
and β2 are the parameters to be estimated, and εit is a random error term.
pothesis according to which the treatment effect is zero and
logAIDI represents the dependent variable (infrastructure develop­
implicitly assumes that the inclusion of unobservable variables
ment), logCF the explanatory variable of interest (Climate finance), X
would make it possible to fully explain the outcome studied (i.e., R2
represents the vector of control variables, selected after an extensive
= 1). However, the presence of measurement errors in the dependent
review of the existing literature: (i) financial development (logFD), (ii)
variable may lead to an underestimation of the robustness of the
CO2 emissions (logCO2), (iii) natural resources depletion (logNRD), (iv)
results if this assumption is maintained. The approach proposed by
population growth (logPop) and (v) control of corruption (CC).
Oster overcomes this limitation by adjusting the bias estimate. It is
Eq. (2) can be rewritten as follows:
based on establishing a link between the selection of observed vari­
AIDIit = β0 + β1 logCFit + β2 logFDit + β3 logCO2it + β4 logNRDit ables and that of unobserved variables, while setting a plausible
(3)
+ β5 logPopit + β6 CCit + εit upper limit for R2), corresponding to a situation in which all unob­
served variables would be included in the estimate. For this reason,
Where β1 to β6 , the parameters to be estimated. we use Oster’s stability test, which is gradually becoming a reference
tool in the empirical literature for assessing the sensitivity of results
3.3. Identification and estimation strategy to unobserved variables (Avom and Ngounou, 2025). In addition, to
mitigate the impact of measurement errors, we use alternative in­
3.3.1. Identification strategy dicators of climate finance. More specifically, we use a measure
This study examines how climate finance affects infrastructure expressed as a percentage of GDP in order to improve the robustness
development in Africa. OLS was used to estimate our baseline model in of our analysis.
Eq. (3). However, OLS estimation of such a model does not solve the − Reverse causality
problem of endogeneity, which can arise from three sources: omitted
variable bias, measurement error, and reverse causality. Failure to take Even after dealing with the omitted variable bias, concerns remain
endogeneity into account can lead to biased data and erroneous about reverse causality between infrastructure development and climate
conclusions. finance, which could lead to simultaneity and biased OLS coefficients
(Wooldridge, 2010). The initial assumption is that climate finance
− Variable omission bias and measurement error stimulates infrastructure growth, but reverse causality may distort this
One of the main challenges in estimating Eq. (3) is the risk of bias assumption: countries with pre-existing absorptive capacities or strong
caused by the omission of certain variables. In other words, there is a institutional frameworks may attract more climate finance. Donor
probability that determining factors have not been included in the allocation criteria, which favor countries with developed in­
basic model, which could distort the results obtained. A common frastructures, further reinforce this bidirectional dynamic. Stagnant
approach is to include as many determinants of the dependent var­ infrastructure could also reduce climate finance if funds are misused,
iable as possible. However, for the purposes of this study, it is reinforcing interdependence rather than a unidirectional causal link
extremely complex, if not impractical, to incorporate all the (Nakhli et al., 2020). To mitigate this problem, an instrumental variable
explanatory variables into the econometric analysis. approach could be considered, but its effectiveness depends on the
The literature nevertheless proposes two methods for estimating

6
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

identification of a strong exogenous instrument, exclusively related to issue of inadequate identification of instruments. The IV-GMM estimator
climate finance and unrelated to the error term. Such instruments are has been used in recent empirical studies (Acheampong et al., 2022;
rare in practice, as they should not directly influence the development of Njangang et al., 2024). For robustness purposes, we used an estimation
infrastructures apart from their effect via climate financing. Poor in­ technique similar to quantile regressions introduced by Koenker and
strument selection risks exacerbating endogeneity rather than solving it Bassett (1978) to examine the heterogeneous effects of climate funds on
(Anderson, 2022). Consequently, the study considers the use of lagged different levels of infrastructure development in African countries.
climate finance variables to establish temporal precedence and reduce
the risk of reverse causality (Beck and Katz, 1996). Lagged values test 4. Results and discussion
whether the impact of climate finance on infrastructure persists over
time, given the delayed implementation and economic effects of these 4.1. Baseline OLS estimates
projects (Flyvbjerg, 2014). Immediate correlations between current
infrastructure and future finance could instead signal reverse causality, Table 3 presents the results of the baseline model estimated using the
where existing infrastructure attracts finance. This approach favors time OLS technique. Columns (1) to (6) present the separate effects for each
sequencing to disentangle causality. climate finance indicator: Global Climate Finance (GCF), Mitigation
Climate Finance (MCF), and Adaptation Climate Finance (ACF). Col­
3.3.2. Estimation strategy umns (1), (3), and (5) present the results of the bivariate model without
As no identification approach is perfect, we use a thorough and control variables, while columns (2), (4), and (6) present the results of
appropriate empirical analysis to reduce endogeneity in this work. Eq. the basic model with control variables.
(3) is estimated using the instrumental variable generalised method of In column (1), the coefficient associated with GCF is positive and
moments (IV-GMM) proposed by Baum et al. (2003). This technique is statistically significant at the 5 % level, suggesting that the increase in
particularly well-suited to this study because of its ability to deal climate finance as a whole contributes to boosting the level of infra­
effectively with estimation problems caused by sources of endogeneity. structure for the set of African countries considered. In column (2), when
Using a conventional estimator such as OLS to estimate an empirical the control variables are included, the coefficient remains positive and
model can cause endogeneity problems and result in attenuation bias, significant, and its magnitude and significance increase. All other things
leading to biased OLS estimates (Wooldridge, 2015). In addition, the being equal, a 10 % increase in overall climate finance will result in a
IV-GMM estimator can provide consistent estimates when faced with 0.779 % increase in the level of infrastructure. The results also show that
endogeneity issues. This estimator also takes into account the presence the coefficients associated with climate financing for mitigation and
of first-order autocorrelation (AR(1)) within panels and hetero­ adaptation are positive and significant at the 1 % level. This means that
skedasticity (Baum et al., 2003). The IV-GMM estimator is a static esti­ their 10 % increase translates into an improvement in the level of
mator that produces long-term coefficients, unlike the dynamic infrastructure development of 0.598 % and 0.549 % respectively. These
system-GMM estimator, which is designed to estimate short-term results therefore confirm the first hypothesis of this work, as theoreti­
coefficients. cally discussed in subsection 2.2, on the beneficial role of climate
The robust option was used in the application of the IV-GMM method financing in increasing the level of infrastructure in Africa. They can
in order to take heteroskedasticity into account in this study. In addition, then be put into perspective with the conclusions of various previous
the study used Hansen’s J statistic and the Cragg-Donald/Kleibergen- works (Zen and Saputra, 2023; Foster and Briceño-Garmendia, 2010;
Paap F statistic to assess the validity of the instruments. The Hansen J- Meltzer, 2016).
statistic is used to analyze the issue of over-identification of instruments, Several arguments can explain this positive correlation between
while the Cragg Donald/Kleibergen-Paap F-statistic is used to assess the climate finance and infrastructure development, based on various

Table 3
Baseline OLS.
Dependent Variable: Infrastructure development

​ (1) (2) (3) (4) (5) (6)

Panel A. OLS estimates
LogGCF 0.0611** 0.0779*** ​ ​ ​ ​
​ (0.0325) (0.0204) ​ ​ ​ ​
logMCF ​ ​ 0.0855*** 0.0551*** ​ ​
​ ​ ​ (0.0224) (0.0161) ​ ​
logACF ​ ​ ​ ​ 0.0612** 0.0699***
​ ​ ​ ​ ​ (0.0332) (0.0238)
logFD ​ 0.0948*** ​ 0.0982*** ​ 0.0998***
​ ​ (0.0120) ​ (0.0139) ​ (0.0122)
logCO2 ​ 0.326*** ​ 0.317*** ​ 0.302***
​ ​ (0.0448) ​ (0.0463) ​ (0.0493)
logNRD ​ 0.0363** ​ 0.0346** ​ 0.0345**
​ ​ (0.0128) ​ (0.0134) ​ (0.0136)
logPop_Gr ​ − 0.425*** ​ − 0.402*** ​ − 0.424***
​ ​ (0.0444) ​ (0.0434) ​ (0.0525)
CC ​ 0.230*** ​ 0.224*** ​ 0.220***
​ ​ (0.0289) ​ (0.0293) ​ (0.0282)
Constant 2.454*** 3.232*** 2.348*** 3.420*** 3.394*** 3.734***
​ (0.260) (0.188) (0.160) (0.164) (0.252) (0.199)
Observations 433 347 411 331 403 322
R-squared 0.012 0.579 0.041 0.587 0.010 0.531
Panel B. Selection on observables and unobservable
Delta (δ) statistic for β = 0 ​ 1. 493 > 1 ​ 2.152> 1 ​ 1.114 > 1
Oster bounds (̂ β, β∗ (Rmax = 0.7, δ = 1)) ​ [0.0659; 0.0785] ​ [0.0542; 0.0868] ​ [0.0685; 0.0959]

Source: Author’s construction Notes:

Robust Standard errors in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

7
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

interconnected economic mechanisms. First and foremost, climate R2 max = 1. First, the delta (δ) statistic, which captures the proportion of
financing encourages investment in essential projects, the introduction X, i.e., the treatment variable, explained by the observables relative to
of new technologies, and the attraction of private capital. Thanks to this the unobservables. Next, we defineR2 max, which is calculated taking
dynamic, the continent benefits from a sustainable modernization of its into account X, the observed and unobserved control variables. Note
infrastructures, responding to climate challenges and development that R2 max < 1 as long as ε ∕= 0. The difference between R2 max max and
needs. Climate funds make it possible to inject financial resources 1 captures the degree of idiosyncratic variation in Y.
directly into strategic initiatives such as renewable energies, transport In addition, Oster (2019) provides evidence to support the idea that
infrastructure adapted to extreme climatic conditions, and optimized when the proportionality coefficient (δ)is greater than unity, it means
management of water resources. By supporting these projects, they help that unobserved confounders do not influence the results. The values of
to fill major gaps in access to energy, connectivity, and basic services, the delta statistic (δ) presented in Panel B of Table 3 are greater than
thereby promoting the structural transformation of African economies unity, indicating that unobserved confounders are not driving the
and job creation (Carfora et al., 2019). At the same time, the introduc­ results.
tion of innovative solutions, such as smart solar energy or eco-friendly Furthermore, when the constraint between the estimated climate
building materials, stimulates the transfer of skills and encourages finance coefficient β and the β∗ statistic excludes zero, as Oster (2019)
local innovation. What’s more, the projects supported by climate finance argues, then the baseline estimate is not driven solely by unobserved
often require additional investment. For example, the installation of variables. The results in Table 3 in Panel B show that none of these in­
solar or wind farms requires the construction of access roads or the tervals include zero, suggesting that the estimated effects of climate
deployment of digital systems to efficiently manage energy distribution. finance on infrastructure development are unlikely to be explained by
This complementarity generates a multiplier effect on the economy by unobserved factors.
encouraging the public and private sectors to develop ancillary infra­
structure, thereby boosting the overall competitiveness of the region
(Meltzer, 2016). However, obtaining climate financing often requires 4.2. Endogeneity issue: IV- GMM
compliance with high standards of governance, with requirements in
terms of transparency and environmental impact assessment. These Previous OLS results suggest that climate finance improves infra­
constraints encourage better management of public and private re­ structure development in African countries. In addition, we provide
sources, ensuring a more efficient distribution of investment for infra­ empirical evidence using the Oster (2019) coefficient stability test that
structure development (Cavalieri et al., 2020). the previous results are unlikely to be due to omitted variables. While
The introduction of control variables in models (2) to (4) and (6) this provides interesting information in support of our main hypotheses,
reveals mixed effects on infrastructure development. First, the results this result may still suffer from the well-known reverse causality bias. As
reveal crucial economic and environmental dynamics that shape infra­ we saw earlier, although the aim of this study is to analyze the effect of
structure development. Financial development shows a positive associ­ climate finance on infrastructure development, it is possible that a
ation in all three models, highlighting the role of advanced financial country’s infrastructure development may, in turn, trigger or accelerate
systems in mobilizing capital for infrastructure. Levine (2005) points out climate finance, in which case there is reverse causality. In order to limit
that efficient intermediation reduces transaction costs and risks, this bias, we build on previous studies (Dong et al., 2024; Acheampong
enabling long-term financing and public-private partnerships that are
crucial to bridging Africa’s infrastructure gap. At the same time, carbon Table 4
emissions show significant positive coefficients, reflecting the depen­ Baseline IV-GMM.
dence of early industrialization on carbon-intensive technologies in the Dependent variable: Infrastructure
energy and transport sectors (Stern, 2004). This highlights a develop­ development

ment trade-off, requiring green financing to align infrastructure growth ​ (1) (2) (3)
with climate objectives. Depletion of natural resources also shows a LogGCF 0.179*** ​ ​
(0.0537)
positive relationship, consistent with the ‘growth paradox’ in ​ ​ ​
logMCF ​ 0.132*** ​
resource-dependent economies: infrastructure expansion, often financed ​ ​ (0.0407) ​
by extractive activities, accelerates ecological degradation (Barbier logACF ​ ​ 0.111***
et al., 2019). For example, building roads in forested areas exacerbates ​ ​ ​ (0.0114)
deforestation, underlining the urgency of policies such as resource logFD 0.0754*** 0.0800*** 0.0985***
(0.0133) (0.0161) (0.0193)
taxation or renewable reinvestment to reconcile short-term gains with
​
logCO2 0.333*** 0.321*** 0.297***
long-term sustainability. ​ (0.0510) (0.0528) (0.0635)
Secondly, demographic and institutional factors also play a key role. logNRD 0.0510*** 0.0496*** 0.0412**
Population growth has a significant negative impact on infrastructure, as ​ (0.0155) (0.0167) (0.0173)
logPop − 0.463*** − 0.434*** − 0.451***
rapid population expansion strains existing systems, diluting per capita
​ (0.0523) (0.0538) (0.0873)
investment and exacerbating urban congestion (Bloom et al., 2010). CC 0.224*** 0.197*** 0.221***
Adaptation strategies such as smart cities and decentralized governance ​ (0.0381) (0.0410) (0.0309)
are essential to align infrastructure provision with demographic change. Constant 2.533*** 2.963*** 3.841***
Conversely, controlling corruption shows a positive relationship, sup­ ​ (0.413) (0.308) (0.722)
Observations 386 367 362
porting Mauro’s (1995) argument that reducing corruption increases the
R-squared 0.570 0.577 0.524
efficiency of public investment. Transparent procurement and Hansen j 0.000 0.000 0.000
accountability mechanisms minimize resource leakage, ensuring that Instruments 6 6 6
infrastructure projects achieve development objectives. Kleibergen-Paap rk Wald F statistic 42.585 50.348 23.028
Although the results of the OLS estimations confirm our main hy­ Cragg-Donald Wald F statistic 59.975 53.390 26.707
Stock-Yogo weak ID test critical values
pothesis that climate finance favors infrastructure development in Af­ 10 % maximal IV size 16.38 16.38 16.38
rica, it is possible that this is due to the exclusion of relevant 15 % maximal IV size 8.96 8.96 8.96
confounding factors. We use the Oster stability test to address this issue. 20 % maximal IV size 6.66 6.66 6.66
The results of the Oster (2019) stability test are presented in Panel B of 25 % maximal IV size 5.53 5.53 5.53
Table 3. Oster (2019) suggests recording the proportionality coefficient, Source: Author’s construction.
i.e., delta (δ), and the bias-adjusted β∗ coefficient assuming δ = 1 and Notes: Standard errors in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1.

8
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

et al., 2022) and use the instrumental variable generalized method of have the capacity to influence infrastructure development following
moments (IV-GMM), and the results of these estimates are presented in previous studies (El-Bouayady et al., 2024). Table 5 presents the results
Table 4. of the modified base model, which includes control variables such as
When we examine the different diagnostic tests of our model, we can urbanization captured by urban population as a percentage of total
see that the model is well specified. The Hansen over-identification test population, annual growth rate as a percentage of GDP per capita in
(Hansen J) shows that the instruments used are not over-identified and constant local currency, industrial development captured by
are therefore appropriate. Furthermore, the Cragg-Donald/ manufacturing value added and public expenditure measured by gov­
KleibergenPaap F-statistic also suggests that the instrument is not ernment final consumption expenditure (Kelly and Rutazihana, 2024).
weak, as all its values are above the critical values of the Stock-Yogo All the coefficients for global climate financing, mitigation, and adap­
weak identification test. Finally, too many instruments can distort the tation remain positive and statistically significant at conventional levels,
results, so the golden rule is that the number of instruments should al­ as shown in columns (1) to (12). The results also support the idea that
ways be less than the number of countries (Roodman, 2009). The results climate change mitigation finance (MCF) is more crucial than adapta­
in Table 4 produced a maximum number of 6 instruments, which is well tion finance (ACF). Therefore, the results of our study show the
below the number of countries (50). According to the calculated co­ robustness of our findings when more control variables are included.
efficients, climate finance has a positive and statistically significant ef­ Furthermore, our study provides empirical support for our first hy­
fect on infrastructure development in all specifications, which is pothesis, which proposes that climate finance plays an essential and
consistent with the OLS results presented in Table 3. According to the direct role in infrastructure development.
coefficient calculated in column (1), a 1 % increase in global climate
finance is associated with a 0.179 % increase in the level of infrastruc­ 4.3.2. Robustness to alternative measures of climate finance
ture on average. These empirical estimates lend a reassuring credibility In this section, we assess the robustness of our findings using alter­
to the expected beneficial effects of climate finance in African countries. native measures of climate finance. Drawing on recent work by Lee et al.
In columns (2) and (3), we estimate the effect of the two sub-components (2022) and Zhao et al. (2022), we consider the ratio of the three main
of climate finance, namely the Mitigation Climate Finance (MCF) and categories of climate funds - global climate funds, mitigation funds, and
the Adaptation Climate Finance (ACF). adaptation funds - to the GDP of recipient countries. The results, pre­
Once again, in line with the results obtained in Table 3, we observe sented in Table 6, corroborate our previous analyses: the coefficients
that both types of climate finance positively influence infrastructure associated with these three indicators remain positive and statistically
development in Africa. However, our findings reveal that climate significant. These observations indicate that an increase in climate
finance directed toward mitigation efforts has a significantly greater finance substantially favors the expansion of infrastructure in African
quantitative and qualitative effect on infrastructure development than countries. Therefore, our results remain valid regardless of the metric
adaptation finance. This outcome is consistent with previous empirical used to measure climate finance.
studies, such as Buchner et al. (2011), which highlight that mitigation
projects, particularly in sectors like renewable energy, clean transport, 4.3.3. Robustness to the sub-components measure of infrastructure
and energy efficiency, are more likely to attract substantial financial development
flows due to their alignment with industrialization goals and We extend our robustness analysis by examining the relationship
revenue-generating potential. By contrast, adaptation finance, while between climate finance and infrastructure development using a more
critical for enhancing resilience to climate change, tends to depend more detailed approach. Rather than using an aggregate index of infrastruc­
heavily on public or concessional funding and struggles to attract private ture development, we consider its main components: transport, access to
capital due to its lower and less predictable financial returns (Pauw electricity, information and communication technologies (ICTs), and
et al., 2019). This pattern aligns with the observations of Norman and water supply and sanitation. The results of this analysis, presented in
Nakhooda (2015), who argue that the market logic inherently favors Table 7, show that these different dimensions of infrastructure remain
projects with measurable economic outcomes, thereby creating an positively and significantly influenced by climate finance, with the
asymmetry in the distribution of climate finance. Our results, therefore, exception of ICTs, which are not significantly influenced by climate
reinforce the notion that while both mitigation and adaptation finance funds in Africa. The lack of impact of climate finance on ICTs can be
play complementary roles in sustainable development, mitigation explained by the fact that funds are directed towards sectors more
finance has a more immediate and catalytic impact on infrastructure directly linked to the energy transition, such as renewable energies or
expansion, particularly in contexts where economic transformation and transport infrastructure. Alternatively, the lack of direct effect may
industrial upgrading are priorities. The findings also suggest a need to result from the fact that climate finance indirectly favors ICTs, via im­
strengthen institutional mechanisms that can improve the attractiveness provements to other infrastructures such as electricity or transport
of adaptation finance, especially in vulnerable African economies, a (Zhang et al., 2023). Thus, the validity of our findings remains largely
point similarly emphasized by Atteridge (2011) and Nakhooda et al. intact when the overall infrastructure index is disaggregated, reinforcing
(2015). the evidence of a positive link between climate finance and the devel­
opment of physical infrastructure in Africa.
4.3. Robustness checks
4.3.4. Robustness to alternative estimation technique: quantile regression
This subsection presents the robustness of our results by performing In this study, we came to the conclusion of the positive influence of
several sensitivity tests. First, by taking into account a number of climate funds on infrastructure development using a parametric
additional determinants of infrastructure development, we re-estimate approach based on OLS regressions. However, the relevance of our an­
our preliminary model. Second, we re-estimate our baseline model alyses could be questioned due to the fact that this estimator focuses
using alternative measures of climate finance. Third, we test the effect of only on the average effect and does not take into account the effect that
aggregate climate finance on the four main components of the infra­ the measures of climate funds could have on different intervals of the
structure development index. Finally, we use an alternative estimation level of infrastructure in African countries. This section, therefore, an­
technique (quantile regressions) to assess the reliability of our results. alyses the asymmetric relationship between climate finance and infra­
structure development using quantile regression.
4.3.1. Robustness to additional control variables The purpose of using quantile regression is to examine the effect of
In order to make a preliminary assessment of the resilience of the climate finance on infrastructure development across different degrees
model, we add to the base model a series of variables that are thought to of infrastructure development. This allows us to understand how climate

9
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

Table 5
Additional control variables.
Dependent variables: Infrastructure development

​ (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
LogGCF 0.226*** 0.174*** 0.237*** 0.223*** ​ ​ ​ ​ ​ ​ ​ ​
​ (0.0568) (0.0550) (0.0669) (0.0691) ​ ​ ​ ​ ​ ​ ​ ​
logMCF ​ ​ ​ ​ 0.0801*** 0.0982*** 0.9634*** 0.0988*** ​ ​ ​ ​
​ ​ ​ ​ ​ (0.112) (0.110) (0.116) (0.118) ​ ​ ​ ​
logACF ​ ​ ​ ​ ​ ​ ​ ​ 0.153*** 0.132*** 0.206*** 0.214***
​ ​ ​ ​ ​ ​ ​ ​ ​ (0.0396) (0.0351) (0.0446) (0.0502)
Baseline Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Controls
logUrban 0.390*** 0.163* 0.439*** 0.420*** 0.371*** 0.112 0.417*** 0.437*** 0.339*** 0.0397 0.275*** 0.243**
​ (0.0718) (0.0865) (0.107) (0.118) (0.0714) (0.0841) (0.105) (0.117) (0.0665) (0.0926) (0.101) (0.108)
logGDP ​ 0.273*** 0.197*** 0.189** ​ 0.311*** 0.211*** 0.201*** ​ 0.377*** 0.345*** 0.315***
​ ​ (0.0524) (0.0707) (0.0743) ​ (0.0502) (0.0702) (0.0741) ​ (0.0634) (0.0730) (0.0744)
logIND ​ ​ 0.116*** 0.112** ​ ​ 0.0813* 0.0534 ​ ​ 0.162*** 0.184***
​ ​ ​ (0.0393) (0.0530) ​ ​ (0.0467) (0.0625) ​ ​ (0.0367) (0.0522)
LogGov ​ ​ ​ − 0.145* ​ ​ ​ − 0.0848 ​ ​ ​ − 0.254***
​ ​ ​ ​ (0.0758) ​ ​ ​ (0.0794) ​ ​ ​ (0.0813)
Constant 0.399 − 0.475 − 1.616*** − 0.911 1.117** − 0.174 − 1.225** − 0.979 1.957** 0.805 − 0.484 0.547
​ (0.639) (0.550) (0.626) (0.726) (0.502) (0.453) (0.495) (0.628) (0.875) (0.726) (0.813) (0.974)
Observations 386 386 355 338 367 367 339 325 362 362 333 317
R-squared 0.598 0.670 0.708 0.733 0.608 0.682 0.688 0.688 0.572 0.644 0.721 0.747
Hansen j 0 0 0 0 0 0 0 0 0 0 0 0
Instruments 6 6 6 6 6 6 6 6 6 6 6 6
Kleibergen- 41.063 34.734 27.323 25.966 49.622 47.047 29.176 24.896 22.691 20.881 19.389 19.007
Paap rk
Wald F
statistic
Cragg-Donald 56.893 49.787 39.826 39.045 51.506 49.475 32.667 28.221 25.919 23.858 21.240 21.491
Wald F
statistic
Stock-Yogo weak ID test critical values
10 % max IV 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38
size
15 % max IV 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96
size
20 % max IV 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66
size
25 % max IV 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53
size

Source: Author’s construction

Notes: Standard errors in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1

funds influence infrastructure dynamics beyond the conditional mean depending on the quantile. In the lower and intermediate quantiles (Q25
(Koenker and Bassett, 1978). The use of quantile regression in this study and Q50), the coefficients are small and insignificant, indicating that
makes it possible to assess the consistency of the results between this type of financing does not have an immediate impact on infra­
different levels of infrastructure in different African countries. The structure in countries with a low level of infrastructure. However, at
quantile regression results, presented in Table 8, provide a better un­ higher quantiles (Q75 and Q95), the coefficients become positive and
derstanding of how the relationship between climate finance and significant suggesting that at more advanced levels of infrastructure
infrastructure development varies across different levels of infrastruc­ development, climate finance for adaptation plays a more important
ture in African countries. role. This can be explained by the fact that climate adaptation often
The results first show that, with regard to the effect of global climate involves long-term projects (water and natural resource management,
finance (logGCF), the coefficients remain positive and statistically sig­ resilience to natural disasters, resilient transport, resilient urbanization,
nificant across several quantiles, particularly in Eqs. (1) to (4), indi­ etc.) that require already existing infrastructure to be fully operational.
cating that global climate finance significantly favours infrastructure Overall, the results confirm the beneficial role of climate finance in
development. The effect is particularly marked in column (4). However, infrastructure development in Africa, although the effect is greater for
from the lowest quantiles (Q25) to the highest quantiles (Q95), this ef­ countries with a more advanced and developed level of infrastructure.
fect increases, which could indicate that the effect of global climate The graphic illustration in Figure A1 reinforces these results. Global
finance is greater in countries where infrastructure is initially devel­ funding to combat climate change, and funding for mitigation and
oped. Secondly, the results indicate that the effect of mitigation climate adaptation, all show constant upward trends. Their impact is gradually
finance is small and often insignificant, suggesting that this type of increasing in regions with the best basic infrastructure. This may reflect
finance does not contribute directly to infrastructure development in the ability of more advanced economies to mobilize these funds effec­
countries where infrastructure is still underdeveloped. However, from tively, perhaps through established institutions, technical expertise or
the higher quantiles onwards, the coefficients become positive and sig­ scalable projects. The results imply that climate finance complements
nificant, indicating that at higher levels of infrastructure development, existing capacity rather than compensates for deficiencies. The use of
mitigation climate finance has a positive impact. This could be explained quantile regression therefore, highlights the importance of tailoring
by the fact that mitigation investments (renewable energy, energy effi­ climate finance strategies to local conditions in African countries.
ciency, etc.) require a certain level of pre-existing infrastructure to be
fully effective. Lastly, climate financing for adaptation has similar ef­
fects to financing for mitigation and seems to have a variable effect

10
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

Table 6 We draw on recent work by Tadadjeu et al. (2025), who use a two-stage
Alternative measures of Climate finance. approach to mediation analysis. Referring to the theory of Baron and
​ Dependent variable: Infrastructure Kenny (1986), the methodology used is that based on Eqs. 4 and 5, as
development illustrated in Fig. 2.
​ (1) (2) (3) {
k
logGCFgdp 0.0255*** ​ ​ ∑
​ (0.0487) ​ ​ Model1 : Medit = α1 + b1 Climate Financeit + δh Xit + μit (4)
logMCFgdp ​ 0.0180*** ​ h=1

​ ​ (0.0336) ​ {
logACFgdp ​ ​ 0.0122*** k
∑
​ ​ ​ (0.0717) Model2 : AIDIit = α2 + b2 Climate Financeit + δh Xit + vit (5)
logFD 0.0965*** 0.105*** 0.101*** h=1
​ (0.0119) (0.0174) (0.0165)
logCO2 0.324*** 0.329*** 0.166**
​ (0.0653) (0.0619) (0.0773)
Table 8
logNRD 0.0463*** 0.0479*** 0.0235
​ (0.0175) (0.0157) (0.0162) Alternative estimation technique: Quantile Regression.
logPop − 0.435*** − 0.433*** − 0.182* Dependent Variable: Infrastructure development
​ (0.0620) (0.0544) (0.0976)
CC 0.250*** 0.231*** 0.198*** ​ (1) (2) (3) (4)
​ (0.0309) (0.0306) (0.0393) ​ Q25 Q50 Q75 Q95
Constant 3.812*** 3.817*** 4.399*** Panel A
​ (0.193) (0.155) (0.215) LogGCF 0.0528* 0.0619** 0.0783*** 0.159***
Observations 349 367 364 ​ (0.0283) (0.0270) (0.0176) (0.0433)
R-squared 0.576 0.582 0.577 Control variables Yes Yes Yes Yes
Hansen j 0.000 0.000 0.000 Constant 3.223*** 3.466*** 3.634*** 3.408***
Instruments 6 6 6 ​ (0.272) (0.259) (0.169) (0.416)
Kleibergen-Paap rk Wald F statistic 64.099 61.155 33.513 Observations 347 347 347 347
Cragg-Donald Wald F statistic 94.304 72.462 49.917 Pseudo R2 0.583 0.640 0.648 0.781
Stock-Yogo weak ID test critical values Panel B
10 % maximal IV size 16.38 16.38 16.38 logMCF 0.0252 0.0557*** 0.0334** 0.0994***
15 % maximal IV size 8.96 8.96 8.96 ​ (0.0199) (0.0195) (0.0138) (0.0310)
20 % maximal IV size 6.66 6.66 6.66 Control variables Yes Yes Yes Yes
25 % maximal IV size 5.53 5.53 5.53 Constant 3.512*** 3.561*** 3.947*** 3.882***
​ (0.218) (0.214) (0.151) (0.339)
Source: Author’s construction. Observations 331 331 331 331
Notes: Standard errors in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1 Pseudo R2 0.588 0.645 0.751 0.779
Panel C
logACF 0.0449 0.0166 0.0438** 0.0491**
4.4. Empirical analysis of transmission channels ​ (0.0344) (0.0303) (0.0211) (0.0715)
Control variables Yes Yes Yes Yes
In this sub-section, we empirically test the transmission channels Constant 3.831*** 3.734*** 3.883*** 5.190***
discussed in sub-Section 2.2 to identify the mechanisms through which ​ (0.295) (0.259) (0.181) (0.613)
Observations 322 322 322 322
climate finance contributes to the achievement of SDG 9 with respect to Pseudo R2 0.568 0.603 0.703 0.724
the development of sustainable and resilient infrastructure. The context
provided in subsection 2.2 allows us to identify two variables as po­ Source: Author’s construction.
Notes: Standard errors in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1
tential channels, namely human capital and technological innovation.

Table 7
Alternative dependent variables.
Dependent variable logTransport logElectricity logICT logWSS

​ (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
LogGCF 0.252*** ​ ​ 0.259* ​ ​ 0.0499 ​ ​ 0.00684* ​ ​
​ (0.0971) ​ ​ (0.139) ​ ​ (0.0731) ​ ​ (0.0266) ​ ​
logMCF ​ 0.283*** ​ ​ 0.0770* ​ ​ 0.0448 ​ ​ 0.00731* ​
​ ​ (0.0641) ​ ​ (0.106) ​ ​ (0.0488) ​ ​ (0.0206) ​
logACF ​ ​ 0.310* ​ ​ 0.259* ​ ​ 0.267 ​ ​ 0.117*
​ ​ ​ (0.177) ​ ​ (0.266) ​ ​ (0.162) ​ ​ (0.0630)
Control variables Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Constant 0.737 0.629 2.861** 1.383 2.926*** 1.444 4.120*** 4.165*** 5.383*** 4.537*** 4.533*** 5.341***
​ (0.764) (0.474) (1.158) (1.158) (0.800) (1.765) (0.582) (0.381) (1.047) (0.200) (0.147) (0.398)
Observations 386 367 362 386 367 362 386 367 362 386 367 362
R-squared 0.676 0.658 0.670 0.606 0.618 0.647 0.700 0.633 0.572 0.665 0.693 0.551
Hansen j 0 0 0 0 0 0 0 0 0 0 0 0
Instruments 6 6 6 6 6 6 6 6 6 6 6 6
Kleibergen-Paap rk Wald F 42.585 50.348 23.028 52.585 50.348 23.028 42.585 50.348 23.028 42.585 50.348 23.028
statistic
Cragg-Donald Wald F 59.975 53.390 26.707 59.975 53.390 26.707 59.975 53.390 26.707 59.975 53.390 26.707
statistic
Stock-Yogo weak ID test critical values
10 % max IV size 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38 16.38
15 % max IV size 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96 8.96
20 % max IV size 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6.66
25 % max IV size 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53 5.53

Source: Author’s construction.

Notes: Standard errors in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1

11
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

In column (3), the estimated coefficient of human capital is positive

and statistically significant. In the same column, the Sobel and Delta test
statistics are statistically significant at the 1 per cent level, validating the
mediating role of human capital. These results also show that 10.3 % of
the effect of climate finance is moderated by human capital. With regard
to the mediating role of technological innovation in columns (4) and (5),
the values of the various tests (Sobel and Delta) are also positive and
statistically significant, confirming the mediating role of technological
innovation. In column (5), we also find that technological innovation
Fig. 2. From climate finance to infrastructure development: transmission explains 46.4 % of the effect of climate finance on infrastructure
channels. development.
Source: Author’s construction. These results confirm our hypotheses presented in Section 2.2, con­
cerning the transmission channels through which climate funds improve
Where Medit represents the mediators (human capital and technological the level of infrastructure. From an economic perspective, human capital
innovation). The composition of the effect is derived from the above and technological innovation play a mediating role in the relationship
system estimates as follows: Indirect effect : b1 ∗ b3 ; Direct effect : b2 between climate finance and infrastructure development by improving
and Total effect : b1 ∗ b3 + b2 the effectiveness and efficiency of resource allocation and use. Strong
The estimation process is described as follows. First, we estimate human capital endowments - conceptualised as the combination of
model (1) or Eq. (4) by regressing climate finance on mediators to obtain educational attainment, technical skills and institutional knowledge -
b1 . Second, we estimate Eq. (5) by regressing climate finance on infra­ amplify the productivity of climate funds by equipping stakeholders
structure development while controlling for mediators, which gives us with the cognitive and managerial capabilities required for infrastruc­
b1 andb3 . The indirect effect is calculated as the product of b1 andb3 . This ture planning, execution and management (Acemoglu and Autor, 2012).
value also indicates the size of the mediation effect, which depends on The right kind of human capital can shift the political economy in favor
how climate finance affects the mediators (b1 ) and how these mediators, of climate policies, improving the effectiveness of climate finance in
in turn, affect infrastructure development (b3 ) (see Fig. 2 for illustra­ infrastructure projects (Hsu, 2016). In addition, technological innova­
tion). The results of this exercise are presented in Table 9. tion acts as another key intermediary between climate finance and the
Table 9 presents the results of the mediation analysis. Column (1) structural transformation of the infrastructure sector. Climate funds
presents the basic results of the bivariate model, suggesting that climate directed towards more efficient and sustainable technologies (Gao et al.,
finance is positively associated with infrastructure development. Col­ 2021), lead to the development of low-carbon infrastructure, which is
umns (2) and (4) present the estimates of model (1), considering human also essential for sustainable urbanisation and climate resilience (Trinh,
capital and technological innovation as mediators, respectively. Col­ 2022).
umns (3) and (5) provide estimates for model (2), incorporating medi­
ators as control variables. The overall results indicate that climate 5. Conclusion and policy recommendations
finance has a significant impact on both mediators, with statistically
significant effects at the 1 % level, as shown in columns (2) and (4) of Climate change has become a major obstacle to development,
Table 9. Each mediator also has a distinct effect on infrastructure particularly in regions such as Africa where there are already significant
development, as shown in columns (3) and (5) of the same table. In gaps in infrastructure. Extreme weather events, rising temperatures and
addition, climate finance has a substantial influence on infrastructure changing rainfall patterns are damaging physical infrastructure and
development when the mediators are excluded, as shown in column (1). disrupting essential services such as transport, water supply, energy and
communications. These disruptions further hamper economic growth
and human development. Responding to vulnerabilities in infrastructure
Table 9 systems requires urgent and coordinated action. This includes building
Transmission channels of the effect of climate finance on infrastructure climate-resilient infrastructure, investing in low-carbon technologies
development. and ensuring equitable access to basic services. However, such trans­
Total Human capital Channel Technological formations require significant financial resources. In this respect,
effect Innovation Channel climate finance, particularly from international sources, can play a
Variables
logAIDI
logHC logAIDI logTI logAIDI decisive role in supporting the development of infrastructure that is both
(1)
sustainable and resilient to future shocks.
(2) (3) (4) (5)
Using panel data from 2012 to 2021 and methodologies such as OLS
logGCF 0.0611** 0.0955*** 0.0548** 0.170*** 0.0176*** and IV-GMM to deal with endogeneity and omitted variable bias, the
(0.0273) (0.0245) (0.0266) (0.00996) (0.00262)
​
study found that climate finance significantly improves infrastructure
logHC ​ ​ 0.660*** ​ ​
​ ​ ​ (0.0505) ​ ​ development across African nations. Mitigation finance had a greater
logTI ​ ​ ​ ​ 0.167*** direct impact than adaptation finance. However, this study uniquely
​ ​ ​ ​ ​ (0.0146) demonstrated that climate finance also acts indirectly through the
Constant 2.454*** 0.807*** 1.922*** − 1.868** 2.767*** development of human capital and technological innovation. The het­
(0.222) (0.200) (0.214) (0.777) (0.204)
erogeneity analyses also highlighted the fact that the effectiveness of
​
Observations 433 433 433 433 433
Indirect effect ​ ​ 0.006*** ​ 0.028*** climate finance depends on the pre-existing capacity of infrastructures.
Delta test Robustness checks confirmed the reliability of these results. Our findings
​ ​ ​ (0.0016) ​ (0.0019) remained valid when we took into account additional control variables,
Indirect effect 0.006*** 0.028***
​ ​ ​
alternative measures of climate finance, and infrastructure. In addition,
Sobel test
​ ​ ​ (0.0016) ​ (0.0018) the Oster (2019) test for omitted variable bias confirmed the internal
% of total ​ ​ 10 % ​ 46 % validity of the estimates. The heterogeneity analysis showed that the
effect that is effectiveness of climate finance in promoting infrastructure is stronger
mediated in countries with good absorptive capacity. In addition, quantile re­
Source: Author’s construction. gressions showed that climate finance has a greater marginal effect in
Notes: Standard errors in parentheses; *** p < 0.01, ** p < 0.05, * p < 0.1 countries with a moderate stock of infrastructure.

12
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

Based on the empirical findings of the study, several targeted policy infrastructure transformation across Africa.
recommendations are proposed to enhance the developmental impact of Despite its contributions, this study has some limitations. The anal­
climate finance on infrastructure in Africa. First, given the observed ysis is restricted to data from 2012 to 2021, which may overlook longer-
positive and significant effect of climate finance on infrastructure term dynamics. Future research could extend the timeframe to assess the
development, African governments, working with international donors, sustainability of the observed effects. Additionally, exploring other
could intensify efforts to secure and channel a substantial portion of the transmission channels and conducting comparative regional analyses
$100 billion annual climate finance commitment pledged under the within Africa would offer deeper insights into the climate finance­
Paris Agreement toward infrastructure-related sectors. This funding –infrastructure nexus.
should focus on projects that are both climate-resilient and
development-enabling. Second, the study finds that mitigation finance CRediT authorship contribution statement
has a stronger influence on infrastructure outcomes than adaptation
finance. As such, priority should be given to mitigation-focused infra­ Arsene Mouongue Kelly: Writing – review & editing, Supervision,
structure investments, especially in renewable energy, sustainable urban Investigation, Formal analysis. Pascal Ndyanabo Rutazihana: Writing
transport, and digital connectivity, which simultaneously address – original draft, Methodology, Investigation, Data curation, Conceptu­
climate and economic goals. Third, the role of human capital and alization. Bruno Emmanuel Ongo Nkoa: Writing – review & editing,
technological innovation as significant transmission mechanisms sug­ Visualization, Supervision, Software, Resources, Investigation.
gests that climate finance must be complemented by investments in
education, vocational training, and national innovation systems. Funding
Strengthening research institutions, technical training programs, and
digital infrastructure can improve the absorptive capacity of countries This research received no external funding.
and ensure that climate finance delivers long-term, scalable infrastruc­
ture benefits. Finally, to promote inclusive growth, climate finance
should target underserved regions and marginalized communities, Declaration of Competing Interest
ensuring equitable access to climate-resilient infrastructure. These rec­
ommendations call for a coordinated, inclusive, and innovation-driven The authors declares no real, potential, or apparent conflict of
strategy to leverage climate finance as a catalyst for sustainable interest

Appendix

Figure A1. The magnitude of the effects of climate finance on infrastructure development by quantile
Source: Author’s construction

13
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

List of countries

Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cabo Verde, Cameroon, Central African Republic, Chad, Comoros, Congo Dem Rep,
Congo Rep, Cote d′Ivoire, Djibouti, Egypt, Eritrea, Eswatini, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Liberia,
Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Sierra
Leone, Somalia, South Africa, Sudan, Tanzania, Togo, Tunisia, Uganda, Zambia, Zimbabwe.

Data Availability Effiom, L., Agala, F., 2020. Infrastructure and Africa’s development: the imperative of
alternative funding options. Financing Africa’s Development Paths Sustainable
Economic Growth 101–133.
Data will be made available on request. El-Bouayady, R., Radoine, H., El Faouzi, N.E., Tayi, S., Ozkan, H.C., 2024. Assessing and
modeling the impact of urbanization on infrastructure development in Africa: a data-
driven approach. Cities 155, 105486.
References
Estache, A., 2010. Infrastructure finance in developing countries: an overview. EIB Pap.
15 (2), 60–88.
Acemoglu, D., Autor, D., 2012. What does human capital do? A review of goldin and Etensa, T., Taye, L., Bersisa, M., 2022. Infrastructure development and economic growth
katz’s the race between education and technology. J. Econ. Lit. 50 (2), 426–463. in east Africa: quantity versus quality dimensions using panel-ARDL approach. Int. J.
Acheampong, A.O., Opoku, E.E.O., Dzator, J., 2022. Does democracy really improve Empir. Econ. 1 (04), 2240001.
environmental quality? Empirical contribution to the environmental politics debate. Falconer, A.D., & Stadelmann, M. (2014). What is climate finance? Definitions to
Energy Econ. 109, 105942. improve tracking and scale up of climate finance.
Agénor, P.R., & Moreno-Dodson, B.. (2006). Public infrastructure and growth: New Feuzeu, D.H.D., Kelly, A.M., 2025. Does financial technology fuel economic complexity?
channels and policy implications (Vol. 4064). World Bank Publications. Empirical insights from Africa. Research, 100360.
Agrawala, S., Dussaux, D., & Monti, N. (2020). What policies for greening the crisis Flyvbjerg, B., 2014. What you should know about megaprojects and why: an overview.
response and economic recovery?: Lessons learned from past green stimulus Proj. Manag. J. 45 (2), 6–19.
measures and implications for the COVID-19 crisis. Foster, V., & Briceño-Garmendia, C. (2010). Africa’s infrastructure. A Time for
Altonji, J.G., Elder, T.E., Taber, C.R., 2005. Selection on observed and unobserved Transformation. A co-publication of the Agence Française de Développement and the
variables: assessing the effectiveness of catholic schools. J. Political Econ. 113 (1), World Bank 〈http://webcache〉. googleusercontent. com/search.
151–184. Foster, V., Gorgulu, N., Straub, S., Vagliasindi, M., 2023. The Impact of Infrastructure on
Anderson, B.S. (2022). Endogeneity in strategic entrepreneurship research. In Research Development Outcomes. World Bank, Washington, DC.
handbook on strategic entrepreneurship (pp. 233-263). Edward Elgar Publishing. Gao, Y., Qian, S., Li, Z., Wang, P., Wang, F., & He, Q. (2021, July). Digital twin and its
Atteridge, A. (2011). Will private finance support climate change adaptation in application in transportation infrastructure. In 2021 IEEE 1st International
developing countries?: historical investment patterns as a window on future private Conference on Digital Twins and Parallel Intelligence (DTPI) (pp. 298-301). IEEE.
climate finance. Hallegatte, S., Vogt-Schilb, A., Bangalore, M., & Rozenberg, J. (2016). Unbreakable:
Avom, D., Ngounou, B.A., 2025. Bright cities and intelligences: can access to ICTs building the resilience of the poor in the face of natural disasters. World Bank
sustainably reduce energy poverty in Africa? Energy Policy 202, 114568. Publications.
Barbier, E., Bishop, J., Aylward, B., Burgess, J., 2019. Economic policy and sustainable Hallegatte, S., Maruyama Rentschler, J.E., & Rozenberg, J. (2019). Lifelines: The
natural resource management. Policies for a small planet. Routledge, pp. 65–90. Resilient Infrastructure Opportunity (Vol. 2).
Batten, S., Sowerbutts, R., & Tanaka, M. (2020). Climate change: Macroeconomic impact Haryani, D., Anjani, Z., 2023. The importance of environmental, social, and governance
and implications for monetary policy. Ecological, societal, and technological risks (ESG) principles in public works and housing infrastructure. J. Infrastruct. Policy
and the financial sector, 13-38. Manag. (JIPM 6 (1), 15–31.
Baum, C.F., Schaffer, M.E., Stillman, S., 2003. Instrumental variables and GMM: Howitt, P., Aghion, P., 1998. Capital accumulation and innovation as complementary
Estimation and testing. The Stata Journal 3 (1), 1–31. factors in long-run growth. J. Econ. Growth 3, 111–130.
Beck, N., Katz, J.N., 1996. Nuisance vs. Substance: specifying and estimating time-series- Hsu, S.L. (2016). Capital transitioning: A human capital strategy for climate innovation.
cross-section models. Political Anal. 6, 1–36. Forthcoming, Transnational Environmental Law, FSU College of Law, Public Law
Betzold, C., Weiler, F., 2017. Allocation of aid for adaptation to climate change: do Research Paper, (809), 16-12.
vulnerable countries receive more support? Int. Environ. Agreem. Polit. Law Econ. IEA. (2022). World Energy Outlook. 〈https://www.iea.org/reports/world-energy-outloo
17, 17–36. k-2022〉.
Bhattacharya, A., Romani, M., Stern, N., 2012. Infrastructure for development: meeting ILO (2023). Guidelines for the development of inclusive transport infrastructures. 〈https
the challenge. CCCEP Grant. Res. Inst. Clim. Change Environ. G 24, 1–26. ://www.ilo.org/sites/default/files/2024-09/Guidelines-for-developing-inclusive-tr
Bloom, D.E., Canning, D., Fink, G., 2010. Population aging and economic growth. Glob. ansport-infrastructure_FR.pdf〉.
Growth 297–328. Inderst, G. (2013). Private infrastructure finance and investment in Europe.
Bowen, A., 2016. Raising climate fi nance to support developing country action: some IPCC. (2022). Climate Change 2022: Mitigation of Climate Change. Cambridge
economic considerations. International Climate Finance. Routledge, pp. 114–134. University Press 〈https://www.cambridge.org/core/books/climate-change-2022〉
Brichetti, J.P., Mastronardi, L., Rivas, M.E., Serebrisky, T., & Solís, B. (2021). The mitigation-of-climate-change/2929481A59B59C57C743A79420A2F9FF.
infrastructure gap in Latin America and the Caribbean: investment needed through Jedwab, R., Storeygard, A., 2019. Economic and political factors in infrastructure
2030 to meet the sustainable development goals. investment: evidence from railroads and roads in Africa 1960–2015. Econ. Hist. Dev.
Buchner, B., Falconer, A., Hervé-Mignucci, M., Trabacchi, C., Brinkman, M., 2011. The Reg. 34 (2), 156–208.
landscape of climate finance. Climate Policy Initiative, Venice 27. Kelly, A.M., 2024. Battling for food security in Africa: is climate finance the missing
Buchner, B., Clark, A., Falconer, A., Macquarie, R., Meattle, C., Tolentino, R., bullet? World Food Policy 10 (2), 227–253.
Wetherbee, C., 2014. Global landscape of climate finance 2019. Clim. Policy Initiat. Kelly, A.M., 2025. Does climate finance foster happiness in African economies? Assessing
32 (1), 1–38. the direct and indirect pathways. Glob. Transit.
Calderon, C., Cantu, C., Chuhan-Pole, P., 2018. Infrastructure development in Sub- Kelly, A.M., Ketu, I., Tchapchet Tchouto, Nembot Ndeffo, L., 2024. Investigating the link
Saharan. Afr. a Scorec. World Bank Policy Res. Work. Pap. (8425). between exhaustion of natural resources and economic complexity in sub-Saharan
Calderón, C., & Servén, L.. (2010). Infrastructure and economic development in Sub- Africa. African Development Review 36 (3), 486–502.
Saharan Africa. Journal of African Economies, 19(suppl_1), i13-i87. Kelly, A.M., Ndeffo, L.N., 2025. Aid or sanction? Comparing the effectiveness of climate
Carfora, A., Pansini, R.V., Scandurra, G., 2019. The causal relationship between energy finance and environmental taxation in mitigating environmental degradation in
consumption, energy prices and economic growth in asian developing countries: a Africa. Innov. Green. Dev. 4 (3), 100241.
replication. Energy Strategy Rev. 23, 81–85. Kelly, A.M., Rutazihana, P.N., 2024. Digital divide and its implications for education in
Cavalieri, M., Guccio, C., Lisi, D., Rizzo, I., 2020. Does institutional quality matter for the Post-covid-19 era in Sub-Saharan Africa. Innov. Educ. 6 (2), 108–132.
infrastructure provision? A non-parametric analysis for Italian municipalities. Ital. Ketu, I., Tchapchet-Tchouto, J.E., 2025. Revisiting the infrastructure–industrialisation
Econ. J. 6, 521–562. nexus in Africa: economic uncertainty contingent and thresholds. J. Infrastruct. Dev.
Cerra, V., Cuevas, A., Goes, C., Karpowicz, I., Matheson, T., Samake, I., Vtyurina, S., Ketu, I., Tchouto, J.E.T., Kelly, A.M., 2022. Does infrastructure development drive
2017. Determinants of infrastructure and its financing. Emerg. Econ. Stud. 3 (2), economic complexity in African countries. Econ. Bull. 42 (4), 2043–2053.
113–126. Koenker, R., Bassett Jr, G., 1978. Regression quantiles. É conom. J. Econom. Soc. 33–50.
Chelleri, L., Schuetze, T., Salvati, L., 2015. Integrating resilience with urban Kumar, B.R., 2022. Infrastructure financing instruments. In Project Finance: Structuring,
sustainability in neglected neighborhoods: challenges and opportunities of Valuation and Risk Management for Major Projects. Springer International
transitioning to decentralized water management in Mexico City. Habitat Int. 48, Publishing, Cham, pp. 31–54.
122–130. Lee, C.C., Li, X., Yu, C.H., Zhao, J., 2022. The contribution of climate finance toward
Dong, D.K., Zhao, C., Dong, X., 2024. From hell to heaven: how climate risks hurt the environmental sustainability: new global evidence. Energy Econ. 111, 106072.
poor and climate finance heals them. Bull. Monet. Econ. Bank. 27 (4), 603–630. Levine, R., 2005. Finance and growth: theory and evidence. Handb. Econ. Growth 1,
Edenhofer, O., 2015. Climate change 2014: mitigation of climate change, 3. Cambridge 865–934.
University Press.

14
B.E. Ongo Nkoa et al. Development and Sustainability in Economics and Finance 8 (2025) 100096

Luu, H.N., Nguyen, L.Q.T., & Nguyen, L.T.M. (2024). The impact of foreign direct Rutazihana, P.N., Kelly, A.M., 2025. Climate Finance and Natural Resource Depletion in
investment on infrastructure development. Journal of Economic Studies, (ahead-of- Africa: Does Institutional Quality Matter? Society & Natural Resources 1–25.
print). Seethepalli, K., Bramati, M.C., Veredas, D., 2008. How relevant is infrastructure to
MacDonald, A.M., Lark, R.M., Taylor, R.G., Abiye, T., Fallas, H.C., Favreau, G., West, C., growth in east Asia? World Bank Policy Res. Work. Pap. 4597.
2021. Mapping groundwater recharge in Africa from ground observations and Stern, D.I., 2004. The rise and fall of the environmental kuznets curve. World Dev. 32 (8),
implications for water security. Environ. Res. Lett. 16 (3), 034012. 1419–1439.
Markolf, S.A., Chester, M.V., Eisenberg, D.A., Iwaniec, D.M., Davidson, C.I., Tadadjeu, S., Ndjakwa, M., Kamguia, B., 2025. Improving access to drinking water and
Zimmerman, R., Chang, H., 2018. Interdependent infrastructure as linked social, sanitation in developing countries: do women policymakers matter? J. Dev. Stud.
ecological, and technological systems (SETSs) to address lock-in and enhance 1–21.
resilience. Earth’s. Future 6 (12), 1638–1659. Taufik, I., Markhamah, S., 2024. The effect of infrastructure development on economic
Mauro, P., 1995. Corruption and growth. Q. J. Econ. 110 (3), 681–712. growth in ASEAN countries: dynamic panel data analysis. Nusant. Econ. Entrep. J.
Meltzer, J.P., 2016. Financing low carbon, climate resilient infrastructure: the role of 85–92.
climate finance and Green financial systems. Clim. Resilient Infrastruct. Role Clim. Thacker, S., Adshead, D., Fay, M., Hallegatte, S., Harvey, M., Meller, H., Hall, J.W., 2019.
Financ. Green. Financ. Syst. (Sept. 21 2016). Infrastructure for sustainable development. Nat. Sustain. 2 (4), 324–331.
Nakhli, S., Gaies, B., Guesmi, K., Moisseron, J.Y., 2020. How do institutional frameworks Timilsina, G., Stern, D.I., Das, D.K., 2024. Physical infrastructure and economic growth.
shape financial development in developing countries? Bank. Mark. Invest. 161, Appl. Econ. 56 (18), 2142–2157.
19–41. Tol, R.S., 2002. Estimates of the damage costs of climate change. Part 1: benchmark
Nakhooda, S., Watson, C., & Schalatek, L. (2015). The global climate finance estimates. Environ. Resour. Econ. 21, 47–73.
architecture. Overseas Development Institute. Tongson Jr, R.R.I., Caelian, M.V., 2024. Economic and social benefits and challenges in
Newell, P., 2025. Towards a more transformative approach to climate finance. Clim. the infrastructure projects implementation in a component city in negros occidental.
Policy 25 (2), 257–268. Tech. Soc. Sci. J. 58, 290.
Ningaye, P., Ketu, I., 2023. Does infrastructure development matter for the shadow Trinh, H.H.. (2022). Smart energy systems, infrastructure financing, and the wider
economy in African countries? Int. Rev. Appl. Econ. 37 (3), 290–310. economy. In Handbook of smart energy systems (pp. 1-23). Cham: Springer
Njangang, H., Padhan, H., Tiwari, A.K., 2024. From aid to resilience: assessing the impact International Publishing.
of climate finance on energy vulnerability in developing countries. Energy Econ. Wassou, N.L., Kelly, A.M., Ndeffo, L.N., 2024. Enhancing population’s health status in
134, 107595. sub-Saharan Africa: does economic complexity matter? Econ. Bull. 44 (4),
Norman, M., Nakhooda, S., 2015. The state of REDD+ finance. Cent. Glob. Dev. Work. 1356–1366.
Pap. (378). Wassou, N.L., Ndeffo, L.N., Kelly, A.M., 2025. Does economic complexity drive health
OCDE. (2022).Engaging citizens in cohesion policy. OECD. 8 mars 2022. 〈https://www. sector investment? Empirical assessment of Sub-Saharan Africa. J. Open Innov.
oecd.org/en/publications/engaging-citizens-in-cohesion-policy_486e5a88-en.html〉. Technol. Mark. Complex., 100588
Oster, E., 2019. Unobservable selection and coefficient stability: theory and evidence. Wooldridge, J.M. (2010). Econometric analysis of cross section and panel data. MIT
J. Bus. Econ. Stat. 37 (2), 187–204. press.
Pauw, P., Mbeva, K., Van Asselt, H., 2019. Subtle differentiation of countries’ Wooldridge, J.M., 2015. Control function methods in applied econometrics. J. Hum.
responsibilities under the Paris agreement. Palgrave Commun. 5 (1). Resour. 50 (2), 420–445.
Pauw, W.P., Moslener, U., Zamarioli, L.H., Amerasinghe, N., Atela, J., Affana, J.P.B., Zen, F., & Saputra, W.. Enhancing Climate-Resilient Infrastructure Development in
Weikmans, R., 2022. Post-2025 climate finance target: how much more and how Indonesia.
much better? Clim. Policy 22 (9-10), 1241–1251. Zhang, K.Q., Chen, H.H., Tang, L.Z., Qiao, S., 2022. Green finance, innovation and the
Pickering, J., Betzold, C., Skovgaard, J., 2017. Managing fragmentation and complexity energy-environment-climate nexus. Front. Environ. Sci. 10, 879681.
in the emerging system of international climate finance. Int. Environ. Agreem. Polit. Zhang, T., Wang, W., An, B., Wei, L., 2023. Enhanced glacial lake activity threatens
Law Econ. 17, 1–16. numerous communities and infrastructure in the third pole. Nat. Commun. 14 (1),
Pigou, A., 1920. The Economics of Welfare. Macmillan, London, England. 8250.
Rinaldi, S.M., Peerenboom, J.P., Kelly, T.K., 2001. Identifying, understanding, and Zhao, C., Dong, K., Wang, K., Nepal, R., 2024. How does artificial intelligence promote
analyzing critical infrastructure interdependencies. IEEE Control Syst. Mag. 21 (6), renewable energy development? The role of climate finance. Energy Econ. 133,
11–25. 107493.
Romer, P.M., 1990. Endogenous technological change. J. Political Econ. 98 (5, Part 2), Zhao, C., Zhai, X., Yuan, M., Liu, Z., 2025. The window of opportunity: estimating the
S71–S102. role of climate finance on urban-rural electricity inequality. J. Environ. Manag. 374,
Roodman, D., 2009. A note on the theme of too many instruments. Oxf. Bull. Econ. Stat. 124153.
71 (1), 135–158. Zhao, J., Zhou, B., Li, X., 2022. Do good intentions bring bad results? Climate finance and
economic risks. Financ. Res. Lett. 48, 103003.

15