Intervening Influence of Financial Development on the Relationship Between Sustainability Practices and Sustainable Development of the Sub-Saharan African Countries

Abstract

The objective of this paper was to explore how financial development affects the relationship between sustainability practices and sustainable development in Sub-Saharan Africa, where poor institutional quality and shallow financial markets may prevent sustainability gains from translating into measurable improvements in human development, poverty reduction, and environmental outcomes. Both descriptive and explanatory components were included in the study, which employed a longitudinal panel design. Using a positivist, longitudinal panel design, this study analyzes data from 49 Sub-Saharan African countries (2000–2023) sourced from the World Bank, United Nations Development Programme, and Sustainable Development Reports. Data analysis was done using regression models and descriptive analysis. The findings show that financial development does not serve as an effective transmission channel through which sustainability practices impact the achievement of sustainable development. The research concluded that policy interventions should include developing sustainable banking regulations, creating green finance incentives, establishing sustainability-linked lending criteria, and strengthening financial inclusion policies that target sustainable development sectors.

1. Introduction

Financial development is a very important intervening variable in the correlation between sustainability practices and economic development. Strong financial systems improve the capacity of firms and governments to invest capital effectively in sustainable initiatives, thus intensifying the favourable impact of the ESG framework on sustainable advancement, employment and income equity (Han & Gao, 2024). Green bonds, along with ESG-related loans and sustainable investment funds, can be used to strengthen the process of transition to low-carbon economies and promote inclusive growth, as evidenced by developed economies with developed financial markets, including the European Union (Batae et al., 2020). As an example, nations that are endowed with deep and liquid capital markets, such as Germany and Sweden, have been able to direct investments to renewable energy and social infrastructure and convert ESG promises into quantifiable economic gains (Martin & Dahlstrom, 2020).

In both underdeveloped and corrupt financial systems, ESG activities will not spread, as resources are mismanaged and investors do not trust them to be utilized effectively (Ozili & Iorember, 2024). According to the study by Leong et al. (2021), such a gap may be addressed through financial development, which enhances transparency and minimizes the costs of transactions, especially in emerging markets. To illustrate, ESG reporting on blockchain in Singapore has increased trust in sustainable investment, which has brought global capital and led to economic growth (Kashif et al., 2023). Therefore, financial development serves as a multiplier of ESG only when it operates within strong regulatory systems and is supported by technological advancements.

The mutual influence of financial development and ESG is also determined by global capital flows. Financial institutions like the World Bank and IMF are putting more and more weight on development financing based on ESG principles and they encourage reforms in recipient countries (Musah & Aawaar, 2022). Countries which adjust their financial regimes in accordance with the international standards of financial sustainability, including TCFD, attract an increased inflow of foreign capital and have enhanced economic resilience (Seker & Şengur, 2021). On the contrary, countries with a low level of financial inclusivity, such as some in Latin America, have difficulty using ESG principles to develop because they have little access to green financing (Tithi, 2025). This highlights the need to have policies in the financial sector that incorporate ESG principles to enable sustainable growth.

Financial development in the region is two-sided in the nexus of ESG–economic progress. The economies which possess relatively developed financial systems, including South Africa and Mauritius, have started to use ESG-related investments to develop their economies. The Green Finance Taxonomy of South Africa and the Sustainable Finance Framework of Mauritius have brought in foreign capital to renewable energy and social housing schemes, which have directly increased employment levels and minimized the level of inequality (Ahmed et al., 2022). These countries demonstrate the extent to which regulatory reforms and capital market development can increase the effects the ESG framework has on the economy.

However, systemic financial constraints that weaken the integration of ESG principles affect much of Sub-Saharan Africa. Restricted access to credit, especially for SMEs, suffocates green entrepreneurship and sustainable industrialization (Anakpo et al., 2023). As an example, Nigeria has a huge potential for renewable energy, but due to poor financial intermediation and the high cost of borrowing, investments in solar and wind projects are reduced, and the country continues to depend on fossil fuels (Ntow-Gyamfi et al., 2020). In a similar vein, informal financial systems prevail in Tanzania and Uganda, limiting the expansion of ESG-compliant enterprises (Cracknell, 2023). The activities of AfDB regarding encouraging the use of green bonds and climate finance have been rather successful and demonstrate that local financial solutions are necessary (Mhlanga, 2022).

The region is experiencing challenges resulting from the substantial overlap of sustainable development, financial development, and sustainability practices (Cracknell, 2023). While there has been a remarkable improvement in the integration of sustainable practices in the rest of the world, Sub-Saharan African countries remain behind despite massive potential to become the driving force of sustainable development (Tiony, 2023). There is an extensive amount of information regarding the complex connection between financial growth, a sustainable future, and sustainable principles attributable to the numerous studies on sustainable development conducted in various countries. Although these studies have provided a useful insight into the particular circumstances in which they were carried out, they have been geographically oriented, such as in various regions or different sectors, without relating sustainable development to financial development and sustainability practices.

There is also a critical methodology gap because of limited access to specific datasets for exploring variables in detail in this region (Muigua, 2023). This study was conducted for the purpose of filling this knowledge gap and giving policymakers and national leaders useful information regarding the nature of sustainable development in the region. Through advancing the discourse on sustainable development, this study sought to showcase challenges and opportunities in connection with the accomplishment of sustainability practices in the context of financial growth and sustainable development in this region. This study, therefore, sought to explore how financial development affects the relationship between sustainable practices and sustainable development.

1.1. Literature Review

Financial development in relation to sustainable development is a controversial and multifaceted issue, with studies showing positive and negative trends depending on the context and approach to the research. According to Pushp et al. (2023), financial development interventions in India were correlated negatively with poverty alleviation, which implies that their investigation did not find any beneficial effects of financial growth. This is different to the results of other studies, such as that of Leong et al. (2021), who stressed the positive impacts of financial development, especially regarding financial inclusion, lower transaction costs and improved security of payment. Their results bring forth the importance of financial development as a catalyst to economic growth, which occurs through accessibility to financial services and economic efficiency. Both studies, however, indicate that there is a necessity to conduct more empirical studies that would prove these inferences correct or otherwise. While Pushp et al. indicates that financial development might not be considered sufficient to attain poverty reduction, Leong et al. emphasizes that the general result of financial growth on the economy needs to be quantified more carefully.

To complicate matters, Xiao et al. (2024) also reviewed the outcomes of GDF policy in China and demonstrated that these policies can largely stimulate the process of sustainable development, especially concerning financial integration and industrial transformation. This research provides a more nuanced view than that of Leong et al. due to its selection of targeted financial development policies to achieve sustainability, such as green finance, which are not explicitly discussed in the Leong et al. study that focused on financial development in general. Xiao et al. also indicated the heterogeneous effects of GDF policies in various cities, which were not explicitly explored in the study by Leong et al. (2021).

As opposed to these hopeful perspectives, the conclusions of Pushp et al. indicate some of the dangers of financial development, particularly when it does not focus on the most vulnerable groups. While Leong et al. concentrated on the benefits of such a financial development shared by everyone, and Xiao et al. continued the discussion by proposing green finance, Pushp et al. warn that financial development does not necessarily lead to enhancements for the population, particularly concerning poverty reduction. This means that there is a difference in the overall effectiveness of financial development policies, particularly for matters related to poverty.

A discussion in the Indian context was provided by Nenavath and Mishra (2023), wherein the authors discuss the role of green finance in India, which adds more evidence to the argument that financial development alongside green finance can lead to sustainable economic growth. Their results echo the arguments expressed by Pawłowska et al. (2022) and Kashif et al. (2023) on the positive synergies between financial development and green finance. The concept of green finance as an extension of overall financial growth strategies, as elaborated by Nenavath and Mishra, compliments the results of Xiao et al. (2024) who also subscribe to the incorporation of sustainability in financial development. Although Nenavath and Mishra discuss the Indian context, their study contributes to a bigger picture of the world, demonstrating the potential of financial development to help the sustainability agenda and thereby supporting the opinion that financial development has the capability of supporting economic growth and environmental sustainability.

Financial development and sustainable finance have been increasingly discussed in the recent literature as a key support for environmental sustainability and green transformation in both developed and emerging economies. D. Zhou et al. (2026) found that financial development, renewable energy adoption and green technological innovation play significant roles in achieving carbon neutrality in European nations, a finding that shows that efficient financial systems are able to make resources more available for green investment. Likewise, F. Zhou et al. (2026) discovered that sustainable finance enhances the efficiency of carbon emissions at the prefecture-city level in China by allocating more funding to green investments and optimizing financial resources. Obyda et al. (2026) also showed that financial development and technological innovation have positive impacts on green growth in the BRICS economies, indicating that the financial system can play a role in environmentally sustainable structural transformation in BRICS economies, provided it is enhanced by institutional and technological development.

In the United States, Tithi (2025) also found that private AI investment, financial development, and macroeconomic factors all interact to affect pathways toward carbon neutrality, continuing the trend of increasing evidence that financial systems are playing an increasing role in sustainability outcomes through financing technologies and green investments. However, the literature also shows that financial development–sustainability linkages are not uniform across countries and institutional settings, especially in developing countries. Sarıgul and Cetin (2025) demonstrated that the relationship among globalization, fossil fuel dependence and domestic policy structures is the key element in determining renewable energy transition and sustainable development outcomes. In the same way, Somoye and Ayobamiji (2026), employing quantile-on-quantile KRLS analysis, concluded that the impact of financial development on ecological development is contingent on economic toughness and energy intensity.

The results reinforce the view that financial development does not necessarily bring sustainable development results, particularly in areas where financial systems are shallow, institutional quality is inadequate, and financial inclusion is low. Although the world has witnessed a lot of research on the connection between finance and sustainability, empirical studies adopting dynamic panel data methods on the mediation of financial development in the Sub-Saharan Africa context have been limited. The current study, therefore, adds to the literature by analyzing the potential for the transmission of sustainability practices to sustainable development outcomes under the structural and financial conditions in SSA, and the institutional and intermediation capacity of financial development in this region.

1.2. Theoretical Framework

Financial development was theoretically conceptualized in this study as an intervening mechanism linking sustainability practices and sustainable development. This study was based on the Endogenous Growth Theory, which proposes that growth is the result of a process of endogenous development, in which long-term development is explained by internal structural factors, such as institutional quality, innovation, human capital and efficient resource allocation (Cozzi, 2023). Sustainability practices, such as protection of the environment, social inclusion, good governance, and economic efficiency, enhance economic productivity and investment appeal, which fuels growth in the financial sector from this perspective. Good sustainability practices build investor confidence, lessen uncertainty and increase the credibility of regulators and, as a result, encourage financial institutions funnel saved funds and direct capital into sustainable and productive sectors of the economy (Maharajabdinul, 2024). Moreover, financial development boosts access to capital and its efficient usage, allowing for funding of renewable energy, infrastructure and social development projects, which are critical for sustainable development. Akhtar and Rashid (2024) also state that financial systems promote sustainable development by reducing transaction costs, enhancing financial inclusion, and mobilizing resources for sustainable development. Thus, it is theoretically believed that sustainability practices have positive effects on financial development, which in turn has positive effects on sustainable development results.

The study also adopted Financial Intermediation Theory and Institutional Economics to provide a mechanism for financial intermediation and the potential for failure in the transmission process in Sub-Saharan Africa. According to Financial Intermediation Theory, financial intermediaries promote development by mobilizing savings, allocating capital, lowering transaction costs and minimizing information asymmetry (Sara, 2024). This approach will support sustainable investments by strengthening financial systems, in terms of providing good governance practices, transparency, accountability and regulations. However, the transmission mechanism can be halted by institutional weaknesses. Despite the progress made, many Sub-Saharan African countries still suffer from shallow financial markets, weak regulatory systems, and low financial inclusion, thereby limiting the potential development contribution of financial institutions, as noted by Olumuyiwa (2022). Likewise, Appiah et al. (2025) argue that the financial fragility, governance failures, and low quality of financial institutions diminish their ability to transform sustainability benefits into long-term development results. The lack of significant mediation effects found in this study, then, indicates that the financial development of the region is not sufficiently deep and efficient to turn sustainability practice into sustainable development.

2. Materials and Methods

The investigation examined the dynamics of sustainability in the area over a 24-year period using a longitudinal panel design with a combination of descriptive and explanatory components. The descriptive component of the design was used to methodically record the type, trend, and present state of financial development, sustainable development, and sustainability practices in the region and across time. The explanatory component was used to simultaneously analyze the relationships and conclusions about the relationship between the predictor and outcome variables. Specifically, the study sought to explain how financial development and sustainable practices interact to affect sustainable development results across time.

For the purpose of investigating the correlations between the variables, the paper used a positivist research paradigm. Positivism is founded on the idea that reality is quantifiable and objective, placing a strong emphasis on statistical analysis, hypothesis testing, and empirical data. In order to analyze sustainability practices, financial development, and their impact on sustainable development, the study looked at standard data gathered over 24 years, from 2000 to 2023, from 49 Sub-Saharan African countries. The years after 2000 mark a turning point in the region’s history, with increased global awareness of sustainability, the SDGs’ adoption in 2015, and the emergence of more ESG-related investment patterns and regulations in emerging economies (World Bank, 2023).

Only secondary data from reliable sources, such as the UNDP, WB, and Sustainable Development Reports, were used in the study. In addition to governance measurements like corruption and government effectiveness indices, the WB supplied statistics on environmental indicators like GHGs, forest cover and renewable energy. The UNDP provided information from Human Development Reports on child labour, gender equality, and social rights. Financial development was measured using the World Bank’s Monetary Sector Credit to Private Sector Index. Since the data originated from licenced or open access sources that were accessed through institutional subscriptions, no permissions were required. The instability of the variance across the nations was addressed using the heteroskedasticity-consistent standard errors.

The study standardized all variables to a comparable 0–1 scale to ensure consistency in measurement across countries, time, and differing indicator units. Indicators expressed as percentages or ratios were normalized either directly (X/100) or using min–max scaling, while governance variables were rescaled from their original −2.5 to +2.5 range using a linear transformation to maintain directional interpretability. For indicators where higher raw values represent poorer outcomes (e.g., greenhouse gas emissions, mortality rate, fragility, and natural resource depletion), an inverse transformation was applied to ensure that higher index values consistently reflect better sustainability performance. This harmonization allows all environmental, social, governance, economic, financial, and development indicators to be interpreted uniformly within the empirical models. The four sustainability practices indices (environmental, social, governance, and economic) were each constructed by averaging the standardized 0–1 scores of their respective component indicators.

The investigation used secondary data collected from 49 nations in SSA between 2000 and 2023. With yearly observations over 24 years, the data formed a structured, balanced panel. Four countries did not have data and were removed, leading to 1080 observations. To provide comparability between nations and throughout time, all indicators were measured yearly and standardized to indices.

Table 1. General information on study data.

Indicator	Description	Coverage
Analysis type	Country-level annual observations	Panel data
Number of obs	Total dataset size	1080 (45 × 24 years)
Structure	Dataset structure	Strongly balanced panel
Measurement periodicity	Frequency of data collection	Annual

summarizes the distinctive characteristics of the data.

3. Results

3.1. Descriptive Statistics

The findings for the environmental indicators showed that the GHG Emissions Index had the highest mean (M = 0.894, SD = 0.148), reflecting generally favourable emissions performance across SSA, with relatively low variation among countries. The Renewable Energy Consumption Index recorded a moderate average (M = 0.644, SD = 0.268), suggesting uneven adoption of renewable energy across the region. The Forest Area Cover Index displayed the lowest mean (M = 0.326, SD = 0.252), indicating limited forest coverage with substantial variation among countries. These results highlight that while emissions levels are relatively low, renewable energy adoption is moderate, and forest conservation remains a significant regional challenge, underscoring persistent environmental sustainability concerns.

The descriptive statistics for social practices revealed relatively strong performance in gender-related indicators. The Gender Parity in Education Index had the highest mean (M = 0.856, SD = 0.121), showing near-universal primary school gender parity across SSA countries. The Gender Equality Index recorded a substantial average (M = 0.799, SD = 0.159), reflecting reasonable progress toward labour market equality. The Labour Force Participation Index showed moderate levels (M = 0.640, SD = 0.136), indicating varied economic engagement across the region. These findings suggest that while gender parity in education has been largely achieved, broader gender equality and labour market participation remain areas requiring continued attention.

The governance indicators’ descriptive data showed continuously poor performance in every area. With a moderate mean, the Control of Corruption Index indicated only modest success in combating corruption. Similar levels were found on the Voice and Accountability Index (M = 0.396, SD = 0.139), indicating persistent difficulties with civic engagement and political liberties. These findings highlight widespread governance issues that continue to be major obstacles to SSA’s sustainable growth.

The descriptive statistics for the economic indicators revealed the poorest performance among all variable groups. The Trade Index recorded a very low average (M = 0.202, SD = 0.134), reflecting minimal economic integration across SSA countries. The Domestic Credit to Private Sector Index showed an extremely low mean (M = 0.134, SD = 0.148), indicating severely constrained access to formal finance for businesses. These findings suggest that economic practices remain a fundamental development challenge requiring urgent policy attention.

The descriptive statistics for the financial development indicators demonstrated severely constrained progress. The Monetary Sector Credit to Private Sector Index showed minimal penetration (M = 0.170, SD = 0.156), consistent with broader patterns of limited financial deepening observed across SSA. The relatively low standard deviation suggests that this constraint is widespread rather than concentrated in particular countries, indicating that underdeveloped financial systems represent a pervasive regional characteristic rather than an isolated phenomenon.

Positive environmental changes were accompanied by mixed progress, according to the descriptive statistics for sustainable development metrics. The Human Development Index showed persistent development issues with a modest average (M = 0.517, SD = 0.112). Although there were still differences between nations, the Child Mortality Index showed significant increases in health (M = 0.840, SD = 0.096). These findings show challenges in progress that are hampered by enduring operational issues (

Table 2. Descriptive statistics for the variables.

Index	Mean	Std. Dev.
ghg	0.894	0.148
ren_en	0.644	0.268
forest_cov	0.326	0.252
edu_par	0.856	0.121
gend_eq	0.799	0.159
lab_part	0.640	0.136
corr_ctrl	0.377	0.127
voice_acc	0.396	0.139
gov_eff	0.350	0.119
cred_priv	0.134	0.148
nat_depl	0.918	0.108
trade	0.202	0.134
mon_cred	0.170	0.156
hdi	0.517	0.112
nat_save	0.571	0.128
mort_ind	0.840	0.096

Note: 0–1 scale used.

).

3.2. Panel Unit-Root/Stationarity Test

The stationarity of most, but not all, variables is supported by the LLC panel unit-root test results, which are shown in

Table 3. Levin–Lin–Chu unit-root test for the variables.

Index	Unadjusted t	Adjusted t*	p-Value
ghg	−15.7913	−5.2572	0.0000
ren_en	−14.1928	−5.6037	0.0000
forest_cov	−8.5448	−3.4510	0.0003
edu_par	−7.8005	−7.8005	0.0000
gend_eq	−11.2947	−2.2623	0.0118
lab_part	−12.6363	−2.7127	0.0033
corr_ctrl	−20.1634	−10.1432	0.0000
voice_acc	−21.6605	−10.8248	0.0000
gov_eff	−19.9664	−9.6145	0.0000
cred_priv	−12.3034	−1.2353	0.1084
nat_depl	−29.3696	−15.5397	0.0000
trade	−17.8859	−7.2889	0.0000
mon_cred	−12.2068	−1.1978	0.1155
hdi	−9.4484	−2.4561	0.0070
nat_save	−8.1004	−8.1004	0.0000
mort_ind	−34.3783	−30.6473	0.0000

. At the 99% confidence level, the non-stationarity null hypothesis was rejected. Two variables, the cred_priv_index and the Monetary Sector Credit to Private Sector Index (mon_cred_index), failed to reject the null hypothesis, with p-values of 0.1084 and 0.1155, respectively, suggesting they may contain unit roots. Based on these results, a first-difference transformation was required for the non-stationary series to ensure valid inference in subsequent econometric modelling.

The findings were significant among the stationary variables. The clearest indication of stationarity was seen in the Mortality Index (mort_ind_index) (adjusted t* = −30.6473, p < 0.001). Environmental indicators such as the GHG Emissions Index (ghg_index) and the Renewable Energy Consumption Index (ren_en_index) were strongly stationary. Social and sustainable development indices, including the Human Development Index (hdi_index) and Gender Equality Index (gend_eq_index), similarly, albeit to differing degrees of statistical strength, rejected the null hypothesis of non-stationarity.

The Levin–Lin–Chu test results for the first-differenced variables show strong evidence of stationarity. For both the domestic credit index and monetary sector credit index, the null hypothesis of non-stationarity is decisively rejected. This demonstrates that the initial variables’ first differences are stationary, I(0), and that they were integrated I(1). The transformation successfully eliminated the unit roots, making the differenced series suitable for inclusion in panel regression analysis alongside the other stationary variables (

Table 4. LLC unit-root test for first differencing.

Variable	Unadjusted t	Adjusted t*	p-Value
d_cred_priv_index	−1.2002	−1.1002	0.0000
d_mon_cred_index	−1.3002	−1.2002	0.0000

Note: H₀: Panel contains unit roots (non-stationary).

).

The Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) test results in

Table 5. ADF and PP tests.

Index	ADF Test	p-Value	PP Test	p-Value
ghg	2.6786	0.0037	2.8060	0.0025
ren_en	0.0162	0.4935	−0.9599	0.8314
forest_cov	−2.5829	0.9951	−3.1866	0.9993
edu_par	12.4967	0.0000	25.6396	0.0000
gend_eq	−1.5420	0.9385	−2.2512	0.9878
lab_part	3.4442	0.0003	−0.8209	0.7941
corr_ctrl	13.8610	0.0000	9.5238	0.0000
voice_acc	21.0260	0.0000	19.6203	0.0000
gov_eff	17.9911	0.0000	19.7396	0.0000
cred_priv	−0.5443	0.7069	−1.4975	0.9329
nat_depl	9.9140	0.0000	9.6396	0.0000
trade	6.6268	0.0000	5.6521	0.0000
mon_cred	−0.6255	0.7342	−0.9445	0.8275
hdi	−1.0158	0.8451	−1.4473	0.9261
nat_save	5.2035	0.0000	18.8535	0.0000
mort_ind	38.8301	0.0000	17.6178	0.0000

show strong evidence of stationarity for nine variables. The GHG Emissions Index (ghg_index), Gender Parity in Education Index (edu_par_index), Natural Resources Depletion Index (nat_depl_index), Trade Index (trade_index), nat_save_index, governance indicators, and Mortality Index (mort_ind_index) are some of these stationary variables.

Non-stationarity is evident in seven variables, and both tests are unable to rule out the null hypothesis (p > 0.05). These are the Renewable Energy Consumption Index (ren_en_index), Forest Area Cover Index (forest_cov_index), Gender Equality Index (gend_eq_index), Domestic Credit to Private Sector Index (cred_priv_index), Monetary Sector Credit to Private Sector Index (mon_cred_index) and Human Development Index (hdi_index).

The Labour Force Participation Index (lab_part_index) presents conflicting evidence, with the ADF test indicating stationarity (p = 0.0003), while the PP test suggests non-stationarity (p = 0.7941). Based on a conservative criterion requiring both tests to reject the null hypothesis, eight variables would require first-differencing to achieve stationarity for valid econometric inference.

In the first-differenced series, presented in

Table 6. First differencing using ADF and PP (Fisher-type) tests.

Variable	ADF Test	PP Test	p-Value
d_ren_en_index	40.6475	182.7248	0.0000
d_forest_cov_index	42.4784	210.7608	0.0000
d_gend_eq_index	22.7918	175.3523	0.0000
d_lab_part_index	25.7363	199.9675	0.0000
d_cred_priv_index	24.0629	126.5554	0.0000
d_mon_cred_index	28.6489	127.9551	0.0000
d_hdi_index	30.0216	211.6077	0.0000

Note: H₀: Panel is non-stationary.

, the results demonstrate that the transformation successfully achieved stationarity for all eight previously non-stationary variables. For each differenced series, both the ADF and PP tests decisively reject the null hypothesis of non-stationarity at the 1% significance level (p = 0.0000).

The exceptionally large test statistics across all variables provide strong evidence of stationarity. The Forest Area Cover Index (d_forest_cov_index) and d_hdi_index exhibited the most substantial evidence of stationarity in the PP tests, with statistics of 210.76 and 211.61, respectively. Similarly, the Renewable Energy Consumption Index (d_ren_en_index) showed robust stationarity with an ADF statistic of 40.65 and a PP statistic of 182.72.

These results confirm that the first-difference transformation effectively eliminated the unit roots present in the original level series. Consequently, all eight differenced variables, including d_ren_en_index, d_forest_cov_index, d_gend_eq_index, d_lab_part_index, d_cred_priv_index, d_mon_cred_index, and d_hdi_index are now integrated of order zero, I(0), and meet the stationarity assumption required for valid panel regression analysis. The transformation enables their inclusion alongside the nine originally stationary variables in subsequent econometric modelling without spurious regression concerns.

3.3. Multicollinearity Test

Lack of detrimental multicollinearity among all predictor variables across the various model parameters is confirmed by the Variance Inflation Factor (VIF) study. The individual and mean VIF values for all variable groups are far below the conventional critical criterion of 10, with the majority being considerably lower than the more conservative level of five. Environmental practices demonstrate minimal multicollinearity with a mean VIF of 1.03 and individual values of 1.02–1.04. Social practices show slightly higher but still acceptable levels with a mean VIF of 1.32. Similarly, economic practices show a mean VIF of 3.08 with values ranging from 1.03 to 4.46. The financial development model exhibits an absence of multicollinearity. These results provide strong evidence that the explanatory variables are not excessively correlated, ensuring that the parameter estimates in subsequent regression analysis will be stable, reliable, and interpretable. The data is therefore robust and suitable for multivariate panel estimation without multicollinearity concerns (

Table 7. Multicollinearity test results.

DV	Independent Variables	VIF	1/VIF	Mean VIF
SD	ghg_index, d_ren_en_index, d_forest_cov_index	1.02, 1.03, 1.04	0.978, 0.969, 0.958	1.03
SD	edu_par_index, d_gend_eq_index, d_lab_part_index	1.01, 1.48, 1.48	0.997, 0.677, 0.676	1.32
SD	corr_ctrl, voice_acc, gov_eff indices	4.19, 2.45, 3.62	0.238, 0.407, 0.238	3.42
SD	d_cred_priv_index, nat_depl_index, trade_index	3.75, 1.03, 4.46	0.267, 0.978, 0.224	3.08
SD	d_mon_cred_index	1.00	1.000	1.00

).

3.4. System GMM Model Assumptions

Significant persistence in sustainable development outcomes is indicated by the lagged dependent variable’s (L.SD) statistically significant coefficients (β = 0.593, p < 0.001). This shows that past sustainable development levels had a significant impact on present performance, with roughly 59% of past SD levels continuing into the present era. This result emphasizes how sustainable growth is path-dependent, with prior successes generating momentum for subsequent advancements.

Among environmental practices, none of the variables achieved statistical significance at conventional levels. The GHG Emissions Index (β = 0.103, p = 0.141), Renewable Energy Consumption Index (β = 0.047, p = 0.672), and Forest Area Cover Index (β = −6.466, p = 0.163) all showed statistically insignificant relationships with sustainable development. This suggests that, in the comprehensive model accounting for all other factors, individual environmental indicators may not directly drive sustainable development improvements or their effects may be mediated through other channels.

Social practices presented a mixed picture. The Gender Parity in Education Index emerged as significant (β = 0.101, p = 0.038), indicating that improved educational gender equality positively contributes to sustainable development. In contrast, the Gender Equality (β = 0.256, p = 0.529) and Labour Force Participation (β = −0.256, p = 0.547) indices showed no significant relationship with SD. This pattern suggests that educational gender parity represents a more crucial social dimension for sustainable development than broader labour market equality measures.

Economic and financial variables showed several significant relationships. The Domestic Credit to Private Sector Index exhibited a positive association with SD (β = 0.272, p = 0.042), while the Natural Resources Depletion Index showed a negative relationship (β = −0.103, p = 0.033). The financial development index demonstrated a negative coefficient (β = −0.206, p = 0.001), suggesting that current patterns of financial sector growth may not align with sustainable development objectives. Trade openness remained insignificant (β = −0.009, p = 0.866) (

Table 8. Two-step system GMM estimation results.

Variable	Coefficient (β)	Std. Error	t-Statistic	p-Value	95% Confidence Interval
Lagged dependent variable
Lag of sustainable development	0.5933	0.0946	6.27	0.000	(0.4078, 0.7789)
Environmental practices
ghg_index	0.1030	0.0700	1.47	0.141	(−0.0342, 0.2403)
d_ren_en_index	0.0468	0.1104	0.42	0.672	(−0.1697, 0.2633)
d_forest_cov_index	−6.4661	4.6393	−1.39	0.163	(−15.5589, 2.6266)
Social practices
edu_par_index	0.1013	0.0487	2.08	0.038	(0.0058, 0.1967)
d_gend_eq_index	0.2561	0.4067	0.63	0.529	(−0.5410, 1.0533)
d_lab_part_index	−0.2562	0.4255	−0.60	0.547	(−1.0902, 0.5778)
Governance practices
corr_ctrl	0.0146	0.1236	0.12	0.906	(−0.2277, 0.2569)
voice_acc	0.0098	0.0521	0.19	0.851	(−0.0923, 0.1119)
gov_eff	0.0250	0.1020	0.24	0.807	(−0.1750, 0.2249)
Economic practices
d_cred_priv_index	0.2718	0.1334	2.04	0.042	(0.0104, 0.5332)
nat_depl_index	−0.1028	0.0482	−2.13	0.033	(−0.1972, −0.0084)
trade_index	−0.0093	0.0552	−0.17	0.866	(−0.1175, 0.0989)
Financial development
d_mon_cred_index	−0.2064	0.0600	−3.44	0.001	(−0.3240, −0.0888)
Constant
_cons	0.0824	0.0617	1.34	0.182	(−0.0385, 0.2033)

).

The negative and statistically significant coefficient of financial development (β = −0.2064, p = 0.001) implies that the current financial system in Sub-Saharan Africa may not be adequately addressing sustainable development goals. In many SSA countries, financial systems may be shallow and underdeveloped in terms of their markets, under-inclusive, expensive, speculative and poorly governed. A negative coefficient, therefore, suggests that at present financial sector growth in SSA may lead to investments in green infrastructure, renewable energy, technological innovation and inclusive social development, but may instead move towards investments in short-term consumption, extractive industries, or non-productive activities. This finding suggests that financial development may not necessarily have a sustainable development impact if it is not combined with higher institutional quality, financial regulation more conducive to sustainability, and more inclusive financial intermediation systems.

The estimation approach’s validity is confirmed by the model diagnostics. The main condition for GMM estimation is satisfied by the Arellano–Bond tests (

Table 9. Arellano–Bond test for serial correlation.

Test	z-Statistic	p-Value
AR(1) in first differences	−2.71	0.007
AR(2) in first differences	0.24	0.811

).

The tests also verify that the model definition is suitable and that the instruments are reliable (

Table 10. Hansen J-test and Difference-in-Hansen test.

Test	χ2 Statistic	df	p-Value
Hansen J-test (over identifying restrictions)	24.49	23	0.374
Difference-in-Hansen (exogeneity of IVs)	7.83	11	0.652

).

3.5. Mediation Analysis

The study employed a two-step system GMM estimator to address potential endogeneity, omitted variable bias, heteroskedasticity, and dynamic panel dependence. The system GMM results for the total effect model indicate that none of the sustainability practices indices has statistically significant relationships with sustainable development when examined collectively. The environmental practices index (β = −0.0334, p = 0.795), social practices index (β = 0.0360, p = 0.648), governance practices index (β = 0.0218, p = 0.572), and economic practices index (β = −0.0227, p = 0.657) all show insignificant coefficients (

Table 11. Total effect model.

Variable	β	Std. Error	z-Statistic	p	95% CI
Lagged DV
L.SDI	0.7940	0.0475	16.73	0.000	(0.7010, 0.8871)
EPI	−0.0334	0.1283	−0.26	0.795	(−0.2848, 0.2180)
SPI	0.0360	0.0790	0.46	0.648	(−0.1187, 0.1908)
GPI	0.0218	0.0386	0.57	0.572	(−0.0538, 0.0974)
ECI	−0.0227	0.0513	−0.44	0.657	(−0.1232, 0.0778)
Constant
_cons	0.1008	0.0263	3.84	0.000	(0.0493, 0.1522)

).

The results for the mediation path (sustainability practices → financial development) indicate that none of the sustainability practices indices significantly influence financial development. The environmental practices (β = 0.0358, p = 0.811), social practices (β = −0.0320, p = 0.768), governance practices (β = 0.0250, p = 0.688), and economic practices indices (β = 0.0915, p = 0.423) all show statistically insignificant relationships with financial development, as shown in

Table 12. Sustainability practices and financial development.

Variable	Coefficient (β)	Std. Error	z-Statistic	p-Value	95% Confidence Interval
Lagged Dependent Variable
L.FDI	0.0460	0.0313	1.47	0.142	(−0.0154, 0.1074)
EPI	0.0358	0.1496	0.24	0.811	(−0.2574, 0.3289)
SPI	−0.0320	0.1085	−0.30	0.768	(−0.2447, 0.1807)
GPI	0.0250	0.0623	0.40	0.688	(−0.0971, 0.1470)
ECI	0.0915	0.1143	0.80	0.423	(−0.1325, 0.3156)
Constant
_cons	−0.0416	0.0631	−0.66	0.510	(−0.1652, 0.0821)

.

The full mediation model results, controlling for financial development, show that none of the sustainability practice indices or the financial development index significantly influence sustainable development. The financial development index shows a negative, insignificant effect (β = −0.0743, p = 0.120), as shown in

Table 13. Mediation model.

Variable	Coefficient (β)	Std. Error	z-Statistic	p-Value	95% Confidence Interval
Lagged Dependent Variable
L.SDI	0.8033	0.0332	24.20	0.000	(0.7383, 0.8684)
EPI	0.0486	0.1249	0.39	0.697	(−0.1962, 0.2935)
SPI	0.0785	0.0911	0.86	0.389	(−0.1001, 0.2571)
GPI	−0.0210	0.0543	−0.39	0.699	(−0.1275, 0.0855)
ECI	0.0085	0.0886	0.10	0.924	(−0.1652, 0.1821)
FDI	−0.0743	0.0478	−1.55	0.120	(−0.1680, 0.0194)
_cons	0.0647	0.0398	1.63	0.104	(−0.0132, 0.1427)

.

Based on the findings, the study specified the model as follows:

SDI__it = αSDI__it−1 + β₁EPI__it + β₂SPI__it + β₃GPI__it + β₄ECI__it + β₅FDI__it + μ__i + ε__it

SDI_it = 0.8033SDI__it−1 + 0.0486EPI__it + 0.0785SPI__it − 0.0210GPI__it + 0.0085ECI__it − 0.0743FDI__it + μ_{_i} + ε__it

where SDI represents the sustainable development index, EPI represents the environmental practices index, SPI denotes the social practices index, GPI represents the governance practices index, ECI denotes the economic practices index, and FDI represents the financial development index. The lagged dependent variable remained highly significant across all models, confirming the dynamic nature and persistence of sustainable development in SSA countries. In the total effect model, the lagged SDI coefficient was β = 0.7940 (p < 0.001), while in the full mediation model, it increased slightly to β = 0.8033 (p < 0.001), indicating strong persistence effects where previous sustainable development outcomes significantly influenced current development performance.

The estimation approach’s validity is confirmed by the model diagnostics, as shown by the Arellano–Bond tests (

Table 14. Arellano–Bond test for serial correlation for the mediation model.

Test	z-Statistic	p-Value
AR(1) in first differences	−2.71	0.007
AR(2) in first differences	0.24	0.811

).

4. Discussion

The broader document analysis on the development trajectory of Sub-Saharan Africa strongly corresponds with the descriptive analysis synthesis. The conclusion that there has been some progress in addressing enduring structural issues is consistent with well-established analyses from significant development organizations. For example, the average Human Development Index for the region increased between 1990 and 2021, according to the World Bank (2023). This simultaneously highlights that the majority of the population in Conflict-Affected Situation (FCS) countries, which is a significant subset of SSA, remains in multidimensional poverty. Similarly, the study documents that the share of manufacturing in GDP for Africa has stagnated, hovering below 15% since 2010.

The observed pattern of relatively strong emissions performance coupled with forest conservation challenges reflects documented regional characteristics. Sub-Saharan Africa’s per capita CO₂ emissions are low (World Bank, 2023). Conversely, the FAO (2020) reports the region accounted for the highest net loss of forest area of any region between 2010 and 2020, identifying it as a primary deforestation hotspot. The governance and institutional findings echo documented governance deficits extensively. This pattern reveals uniformly weak governance performance across SSA with minimal differentiation between countries, supporting Batae et al.’s (2020) observation of significant regional disparities in governance effectiveness.

The consistently low scores suggest that governance developments in the region are not sufficient to overcome threshold effects, as governance quality may need to reach a critical level before producing measurable development impacts, as implied by Gundogdu and Aytekin’s (2022) comparison between Nordic and less stable countries. The median level of domestic credit to private sector credit in low-income African countries is less than 15% of the GDP. The moderate sustainable development performance reflects the region’s position in the global rankings. The 2023 Sustainable Development Report states that the average SDG Index score for Sub-Saharan Africa is lower than the global average. The region scores lowest on Goal 9 (Industry, Innovation and Infrastructure), but an average score on Goal 10 (Reduced Inequalities).

The mediation analysis reveals no significant indirect effects, as neither the paths from sustainability practices to financial development nor from financial development to sustainable development are statistically significant. The non-significant coefficients in both the mediation paths and the total effect model indicate that financial development does not mediate the relationship. All Arellano–Bond and instrument validity tests across the three mediation models show satisfactory results (AR(1) significant, AR(2) insignificant), confirming model specification validity. The null mediation finding does not necessarily imply that financial development cannot act as a transmission channel in theory, but rather that the required empirical preconditions for mediation are not satisfied in the SSA context. This suggests that weak or underdeveloped linkages between sustainability practices, financial development, and sustainable development limit the operation of the proposed mediation mechanism.

The first mediation pathway (sustainability practices → financial development) reveals uniformly insignificant relationships, disagreeing with Leong et al.’s (2021) assertion that financial development naturally follows from broader sustainability initiatives. All practices fail to show meaningful influence on financial development. This finding suggests that sustainability practices in Sub-Saharan Africa may be disconnected from financial system evolution, supporting Xiao et al.’s (2024) observation about heterogeneous effects and the need for targeted financial policies rather than relying on general sustainability improvements to drive financial development.

The second mediation pathway (financial development → sustainable development) shows a negative but statistically insignificant relationship, disagreeing with Pawłowska et al.’s (2022) and Kashif et al.’s (2023) positive assessments of financial development’s role in sustainable growth. This near-significant negative coefficient aligns with Pushp et al. (2023) discovery of inverted correlations between financial development and poverty alleviation in India, signifying potential adverse distributional effects or efficiency losses in SSA’s financial systems. The result indicates that financial development in the region may not be structured to support sustainable outcomes, possibly reflecting the cybersecurity risks and digital inequality concerns raised in the literature.

The full mediation model confirms the absence of indirect effects, with financial development failing to serve as a transmission channel between sustainability practices and sustainable development. This null mediation finding contradicts Nenavath and Mishra’s (2023) Indian case study showing positive synergies between green finance and sustainable growth, suggesting that SSA’s financial systems may lack the institutional maturity, regulatory frameworks, or market depth necessary to translate sustainability initiatives into developmental outcomes. The consistently weak lagged financial development coefficient further indicates minimal autoregressive momentum in financial system evolution, contrasting with the strong persistence observed in sustainable development.

The findings align with the literature’s recognition of regulatory and structural barriers. While Xiao et al. (2024) emphasized the importance of targeted Green Digital Finance policies in China, and Pawłowska et al. (2022) highlighted regulatory frameworks to avert risks, SSA’s financial systems may lack these enabling conditions. The findings suggest that financial development in the region may be characterized by what Pushp et al. (2023) termed traditional financial development that does not reach vulnerable groups or align with sustainability objectives, rather than the transformative, inclusive models discussed by Leong et al. (2021).

5. Conclusions

The study’s findings are that financial sector development is not a strong mediation channel for sustainable development practices to increase sustainable development outcomes in Sub-Saharan Africa. The empirical results within the system GMM mediation framework reveal that neither the direct mediation relationship between the indicators of sustainability practices and sustainable development nor the indirect mediation relationship through financial development are statistically significant mediation pathways. This lack of significant mediation effects indicates that, in the context of the sampled SSA economies, the hypothesized function of financial development as an enabler between sustainability practices and sustainable development is not supported.

The results additionally suggest that the financial system in Sub-Saharan Africa could be shallower, less inclusive, and/or less effectively organized than is required to yield tangible development results from improvements in environmental, social, and governance (ESG) sustainability practices. While sustainability practices are becoming more widespread in the region, their developmental potential seems to be limited by the lack of financial intermediation and credit availability for the productive and sustainability sectors, as well as structural inefficiencies in financial markets. Consequently, financial development does not act as a good instrument of resource mobilization or reallocation for sustainable development priorities.

Moreover, the findings indicate that the link between sustainability practices, financial development and sustainable development could be more intricate and context-specific than conventional mediation theorizing suggests. The absence of significant pathways suggests that other institutional, structural, or macroeconomic conditions could be more influential to the likelihood that a sustainability practice will lead to development outcomes, including governance quality, regulatory enforcement, market maturity, and technological capacity. This has led to the question of whether financial development is be enough to achieve sustainable change if institutions are not improved.

Overall, the study adds to the literature by showing that financial development does not necessarily mediate sustainability and development in Sub-Saharan Africa. It highlights the importance for policymakers to work beyond the paradigm of automatic transmission effects and instead build up the structural support of financial systems, enhance financial inclusion, and tie financial sector reforms to sustainability goals. The potential of financial development as a link between sustainability practices and sustainable development is limited without these enabling conditions.

For future research, the authors recommend that additional nonlinear threshold analyses be performed on financial development using panel quantile regression, that second-generation unit-root tests that account for cross-sectional dependence be used, and that researchers use a panel of financial indicators beyond 2023 when they are available. For policy implications, the authors recommend that policymakers focus on strengthening the direct effect of sustainability through more effective implementation mechanisms instead of only financial sector reforms as a transmission channel. This study’s limitations are the lack of data for all SSA countries and their reliance on a one-dimensional credit-based measure of financial development, the inability to include post-2023 data because the UNDP and World Bank did not provide updated data for all the countries in SSA at the time of analysis, and the lack of cross-sectional dependence tests in the unit-root analysis.