Does Profitability Moderate the Relationship Between the Leverage and Dividend Policy of Manufacturing Firms in Nigeria and South Africa?

Abstract

This study examines the moderating role of profitability in the relationship between leverage and dividend policy in listed manufacturing firms in Nigeria and South Africa. Using a sample of 915 firm-year observations from 2013 to 2022, the analysis employs panel Tobit regression to manage the censored nature of dividend data, with logistic regression applied as a robustness check. The findings reveal a negative association between leverage and dividend payout ratio for Nigerian firms, while this association is less pronounced and statistically insignificant in South Africa, reflecting a more flexible financial environment. Profitability strengthens the leverage–dividend policy relationship in Nigeria, enabling firms to maintain dividends despite high leverage; however, this moderating effect is weaker in South Africa. These results underscore the importance of context-specific financial strategies, recommending that Nigerian policymakers improve access to affordable credit, while South African policymakers focus on sustaining market stability. This study advances the understanding of dividend policy in emerging markets by clarifying how leverage and profitability interact to shape dividend practices.

1. Introduction

Corporate dividend policy remains one of the most debated topics in finance, owing to its intricate relationships with firm performance, leverage, and broader corporate governance frameworks. Dividend policy is a strategic decision that determines the timing, amount, and method of profit distribution to shareholders (Barros et al. 2022, 2023). Beyond being a channel for shareholder value creation, it serves as a signal of financial health while also reflecting inherent tensions in resource allocation between reinvestment for growth and returns to investors (Arhinful and Radmehr 2023). These tensions are particularly pronounced in emerging markets, where economic volatility, access to finance, and regulatory inconsistencies further complicate dividend policy decisions. This complexity is particularly evident in Nigeria and South Africa, which represent contrasting yet complementary contexts for studying dividend policy. Nigeria’s developing financial market and resource-driven economy provide insights into dividend policy in environments with significant financial constraints, while South Africa’s more advanced and well-regulated financial system offers a benchmark for comparison.

Understanding these dynamics raises critical research questions, including the following: How do leverage and profitability influence dividend policy in Nigeria and South Africa? What is the relationship between leverage and dividend policy in listed manufacturing firms in these two countries? And does profitability play a moderating role in this relationship? Addressing these questions is crucial, given the strategic importance of dividend policy in emerging markets.

Existing theories offer varied perspectives on the underlying mechanisms that drive corporate dividend policy. Agency theory suggests that dividends mitigate agency conflicts by reducing excess cash, which might otherwise fund inefficient projects (Jensen and Meckling 1976). This dynamic is especially relevant in highly leveraged firms, where debt holders and shareholders contend for access to the firm’s cash flows. Meanwhile, pecking order theory (Myers and Majluf 1984) emphasises firms’ preference for internal financing over debt, positing that profitable firms rely less on external funding and thus enjoy greater flexibility in distributing dividends when sufficient resources are available. Signalling theory (Ross 1977) further explains that firms use dividends to convey stability and profitability to investors, a tactic particularly beneficial in emerging markets where information asymmetry can cloud investor perceptions.

In Nigeria and South Africa, these theories find nuanced application given the market volatility, regulatory constraints, and governance challenges. Manufacturing sectors in both countries face high leverage and costly access to capital, factors that influence their dividend policies (Adelegan et al. 2021; Ahmad et al. 2018; Bello and Lasisi 2020). South Africa’s financial market, characterised by greater regulatory stability and maturity, contrasts sharply with Nigeria’s, where firms operate under more resource constraints and face higher financial volatility. These differences make the comparison of Nigeria and South Africa particularly compelling, as it highlights how diverse institutional, economic, and governance environments shape dividend policy strategies. Motivated by these dynamics, this study aims to address three objectives: (1) to examine how profitability influences dividend policy, (2) to evaluate the relationship between leverage and dividend policy, and (3) to explore how profitability strengthens the leverage–dividend policy relationship, offering a comparative analysis between Nigeria and South Africa.

Despite the relevance of these theoretical frameworks, a significant gap persists in the empirical literature regarding profitability’s moderating role in the relationship between leverage and dividend policy in emerging markets. Existing studies present mixed findings, with some suggesting that profitability enhances a firm’s ability to balance leverage and dividends, while others report varying results depending on the economic context (Adelegan et al. 2021; Ahmad et al. 2018; Anggoro and Yulianto 2019). This inconsistency highlights the need for further investigation, particularly in regions like Africa, where diverse economic and regulatory environments could influence these dynamics. Addressing this gap, the current study examines how profitability, measured as return on assets (ROA), moderates the leverage–dividend policy relationship in Nigerian and South African manufacturing firms. Specifically, it focuses on the role of profitability in shaping this relationship, with the debt-to-assets ratio (DAR) serving as the primary measure of leverage.

This study addresses three critical gaps in the literature. First, it provides new empirical evidence on the moderating role of profitability in the leverage–dividend policy relationship within emerging markets, focusing specifically on the African context. While previous studies, such as those by Al-Najjar and Kilincarslan (2016) and Komrattanapanya and Suntraruk (2014), have examined profitability’s direct impact on dividend policy, its interaction with leverage—particularly as a stabilising factor in volatile markets—remains underexplored. This study bridges this gap by analysing how profitability, measured by return on assets (ROA), influences dividend behaviour in manufacturing firms operating in Nigeria and South Africa.

Second, this study undertakes a comparative analysis of Nigeria and South Africa, two countries with distinct economic structures, regulatory frameworks, and corporate governance systems. This comparative approach highlights how differences in financial development, market maturity, and institutional settings shape dividend policies (Adelegan et al. 2021; Ahmad et al. 2018). While Nigeria represents a more resource-constrained and regulatory-challenged environment, South Africa offers insights from a relatively more developed financial market. By juxtaposing these two economies, this study reveals the extent to which country-specific factors influence firms’ ability to manage leverage and dividend payouts, adding to the existing literature by Setiawan et al. (2023) and Anggoro and Yulianto (2019).

Third, this study addresses methodological limitations in prior research by employing Tobit regression, which is well suited for handling censored dividend data. In many emerging markets, firms often limit or completely forgo dividend payments due to financial constraints, making standard regression techniques less effective in capturing the underlying relationships. By using Tobit regression alongside logistic regression for robustness checks, as recommended by Al-Najjar and Kilincarslan (2016) and supported by Komrattanapanya and Suntraruk (2014), this study provides a more accurate and reliable analysis of the leverage–dividend policy nexus. These methodological advancements contribute to a deeper understanding of dividend behaviour in contexts where financial constraints are prevalent.

In sum, this study explores the interplay between profitability, leverage, and dividend policy in Nigeria and South Africa, using a robust methodological approach that includes Tobit regression to account for censored data and logistic regression as a robustness check. It contributes to the literature by highlighting the nuances of financial strategy in emerging markets, emphasising the importance of profitability in shaping dividend policies. Furthermore, it offers actionable insights for corporate policymakers and financial strategists in African markets, providing practical strategies for managing leverage and dividend payouts under capital constraints and economic volatility.

The remainder of this paper is organised as follows. Section 2 provides a review of the relevant literature and formulates the hypotheses. Section 3 describes the methodology, covering data sources, sample selection, variable definitions, and econometric models. Section 4 presents the empirical results alongside the robustness checks and discusses the key findings. Finally, Section 5 concludes with a summary of the findings, policy implications, limitations, and recommendations for future research.

2. Literature Review and Hypothesis Development

2.1. Profitability and Dividend Policy

Profitability allows firms to meet operational requirements and distribute returns to shareholders, often leading to increased dividend payouts. The relationship holds particular significance in emerging economies such as Nigeria and South Africa, where unique challenges including market volatility, regulatory frameworks, and financial constraints influence corporate financial decisions.

Agency theory suggests that dividend payouts decrease agency costs by restricting the free cash flow accessible to managers, thereby aligning managerial behaviour with the interests of shareholders. Dabboussi’s (2024) study in Saudi Arabia demonstrates that enhanced profitability enables firms to manage leverage more effectively, thus securing resources for dividends that reduce potential agency conflicts. Profitability allows these firms to fulfil debt obligations and sustain regular dividend payouts, thereby establishing a direct connection between dividend policy and efficient resource allocation to enhance shareholder value.

Pecking order theory posits that profitable firms prioritise retained earnings over external financing for investment funding, thereby enabling more flexible dividend distribution. Empirical studies indicate that this preference contributes to dividend stability across various markets. Fama and French (2002) observed that profitable firms uphold stable dividend policies, indicating that profitability offers financial flexibility. DeAngelo et al. (2006) found that a high level of retained earnings in relation to equity facilitates stable dividend payments, which in turn enhances financial stability. This phenomenon is evident in emerging markets: Amidu and Abor (2006) demonstrated that Ghanaian firms with high ROA regularly paid dividends to indicate financial stability, even amid competitive conditions. Similarly, Shyam-Sunder and Myers (1994) noted that profitable firms tend to distribute dividends due to their diminished dependence on external capital.

Signalling theory posits that dividends convey financial strength to investors, a crucial factor in markets characterised by significant information asymmetry. In these contexts, profitable firms frequently utilise dividend payments as a signal of stability. Rafique (2012) noted that profitable firms in emerging markets continue to distribute dividends despite underdeveloped financial conditions, indicating that such payments bolster investor confidence. Louziri and Oubal’s (2022) study on the Casablanca Stock Exchange indicates that profitable firms maintained dividend payments even in adverse conditions, highlighting dividends as a marker of resilience. Lotto’s (2020) study in Tanzania revealed that enhanced profitability resulted in higher dividends and strengthened investor confidence. Rochmah and Ardianto (2020) highlighted the significance of free cash flow in facilitating advantageous dividend policies, particularly in volatile markets. Based on this review, the first research hypothesis is constructed as follows:

H1. 

There is a positive association between profitability and dividend policy in listed manufacturing firms in Nigeria and South Africa.

2.2. Leverage and Dividend Policy

The interplay between leverage and dividend policy typically illustrates a balance aimed at reducing agency costs while addressing cash flow limitations. High leverage diminishes the excess cash accessible to managers, thereby constraining the likelihood of inefficient expenditures and aligning managerial behaviour with shareholder interests. Al-Najjar and Kilincarslan (2016) illustrate that in emerging markets, firms with high leverage prioritise debt obligations over dividend distributions to mitigate agency conflicts, reinforcing the idea that dividend payments may be reduced when debt servicing is required.

Additionally, firms typically favour financing investments via retained earnings instead of external sources, adhering to a financing hierarchy to reduce the costs associated with external funding. According to Frank and Goyal (2009), this preference indicates that firms with elevated leverage levels typically possess limited resources available for dividend distribution. Highly leveraged firms tend to decrease dividend payouts to conserve internal funds for debt obligations, consistent with the pecking order theory, which posits that firms distribute dividends only when surplus cash is present.

Evidence from Rahmawati and Narsa (2020) regarding Indonesian firms in the LQ45 index further supports this relationship. The findings indicate that leverage negatively impacts dividend policy, as firms prioritise debt repayment to ensure financial stability. The findings indicate that high leverage restricts dividend policy, especially in emerging markets characterised by elevated external financing costs, thereby reinforcing the inclination of leveraged firms to reduce dividend distributions.

In situations characterised by information asymmetry, particularly in emerging markets, dividends function as an indicator of financial stability. Firms with moderate leverage may utilise dividends to signal financial resilience to investors, as noted by Utami and Inanga (2011) in Indonesia. Le et al. (2019) report that Vietnamese state-invested enterprises utilise dividends to indicate strong future earnings, despite elevated leverage levels. This illustrates how dividends can convey a firm’s stability and earnings potential, even in the presence of significant debt obligations.

The evidence from emerging markets indicates that high leverage typically compels firms to prioritise debt servicing over dividend distributions; however, dividends can still serve a strategic function in signalling stability when profitability permits. Considering these insights, we propose the subsequent hypothesis:

H2. 

There is a negative association between leverage and dividend policy in listed manufacturing firms in Nigeria and South Africa.

2.3. Profitability, Leverage, and Dividend Policy

Extensive research has analysed the individual relationships among profitability, leverage, and dividend policy; however, the moderating role of profitability between leverage and dividend policy is insufficiently explored, especially in emerging markets such as Nigeria and South Africa. The economic instability and governance challenges in these markets create a distinct context for analysing the impact of profitability on financial decisions. This research examines the relationship between profitability and its influence on the effect of leverage on dividend policy.

Empirical evidence supports the moderating role of profitability in the relationship between leverage and dividend policy. Al-Najjar and Kilincarslan (2016) found that profitable firms in emerging markets continue to distribute dividends even in the presence of high leverage ratios. Fosu (2013) observed that South African firms with higher profitability efficiently manage financial leverage and sustain dividend distributions. Similarly, Le et al. (2019) showed that profitable Vietnamese firms maintain dividend distributions even in the presence of high debt levels. The findings suggest that profitability may mitigate the constraining impact of debt on dividend policy; however, additional research is required to validate these dynamics in various contexts. Based on these findings, we propose the final hypothesis of this study as follows:

H3. 

Profitability strengthens the relationship between leverage and dividend policy in listed manufacturing firms in Nigeria and South Africa.

3. Methodology

3.1. Research Sample

The research sample consists of 915 firm-year observations of listed manufacturing firms in Nigeria and South Africa, covering a 10-year period from 2013 to 2022. The sample was carefully selected using a purposive sampling technique to ensure that only firms with complete and reliable financial data were included. This approach was essential for maintaining the integrity and robustness of the analysis, given the specific research focus on dividend policy, profitability, and leverage.

First, we only considered firms that have consistently listed on the Nigerian Exchange Group (NGX) or the Johannesburg Stock Exchange (JSE) throughout the study period. This ensured that the data reflected a continuous operational presence and minimised disruptions that could arise from firms entering or exiting the market. Second, this study focused exclusively on the manufacturing sector to provide consistency in the analysis and to account for sector-specific characteristics that may influence financial decisions. The manufacturing sector was chosen due to its significant contribution to economic activities in both countries and its relevance in leverage and dividend policy studies. Finally, firms were required to have published annual reports containing comprehensive financial information, including key metrics related to profitability, leverage, and dividend distributions. This criterion ensured the availability of consistent and detailed financial data across the sample.

The data were collected from a variety of reputable sources, including the official websites of the stock exchanges, firm annual reports, and recognised financial databases. These sources provided the necessary depth and accuracy for this study, facilitating a thorough and data-driven investigation into the relationships among the key variables.

3.2. Definition of Variables

Table 1. Definitions of variables.

Variable	Type	Measurement	Definition	References
Dividend Ratio (DPO)	Dependent	Ratio of total dividends paid to net income	Reflects the proportion of earnings distributed to shareholders. Used as a proxy for dividend policy.	Damodaran (2015); Rozeff (1982); Al-Najjar and Kilincarslan (2016)
DPO_binary	Dependent	Binary (1 if dividends paid, 0 otherwise)	Indicates if a firm paid dividends in a given year, enabling analysis of the likelihood of dividend payments.	Abdulkadir et al. (2016)
Debt-to-Asset Ratio (DAR)	Independent	Ratio of total debt to total assets	Represents the leverage by indicating the extent to which assets are financed by debt, reflecting leverage levels.	Jensen et al. (1992); Myers (1977)
Return on Assets (ROA)	Moderator	Ratio of net income to total assets	Measures a firm’s profitability and operational efficiency, reflecting asset use for profit generation.	Akpadaka et al. (2024); Rasheed et al. (2023); Rehman and Takumi (2012)
Firm Size (FSIZE)	Control	Natural logarithm of total assets	Controls for the impact of firm size, as larger firms tend to have stable earnings and higher likelihood of dividend payment.	Holder et al. (1998); Nizar Al-Malkawi (2007)
Institutional Ownership (INSTOWN)	Control	Proportion of shares owned by institutional investors	Reflects corporate governance influences, as institutional ownership often relates to preference for consistent dividends.	Shleifer and Vishny (1997)
Asset Structure (AS)	Control	Ratio of fixed assets to total assets	Controls for the asset base, indicating financial flexibility, with higher fixed assets potentially impacting leverage/dividends.	Priyan et al. (2023)
Current Ratio (CR)	Control	Ratio of current assets to current liabilities	A liquidity measure assessing the firm’s short-term financial health, influencing its capacity to pay dividends.	Akbar and Nurita (2022)
Revenue Growth (RG)	Control	Percentage change in revenue	Reflects firm growth rate, with high-growth firms more likely to retain earnings for expansion, impacting dividend policy.	Coulton and Ruddock (2011); Fama and French (2001)
Country Dummy (CTRY_DUM)	Control	Binary (0 = Nigeria, 1 = South Africa)	Accounts for country-specific effects, capturing differences in financial market, regulatory, and economic conditions.	Cox and Schechter (2019); Yip and Tsang (2007)

Note. Authors’ compilation.

presents a summary of the variables employed in this study, encompassing their classifications, measurement techniques, and definitions. This table delineates the fundamental framework for examining the interrelations among dividend policy, leverage, profitability, and additional control variables within the research context.

3.3. Research Models

This study uses six regression models to examine the relationships between leverage, profitability, and dividend policy, as well as how profitability interacts with leverage and dividend policy. We use panel Tobit regression as the primary model and complement it with a logistic regression model. These models are selected to appropriately address the characteristics of the data and the research objectives.

3.3.1. Panel Tobit Regression Model

The Tobit model is appropriate due to the censoring of the dependent variable, DPO. Firms may exhibit a dividend payout ratio of zero in instances where dividends are not distributed. This establishes a lower limit of zero, rendering ordinary least squares regression inappropriate because of possible biases and inefficiencies in parameter estimation. The Tobit model addresses censoring, facilitating more precise estimation of the relationships between dependent and independent variables (Nizar Al-Malkawi 2007; Singhania and Gupta 2012; Warganegara et al. 2020).

The panel Tobit regression model is specified as follows:

DPO_it = β₀ + β₁ROA_it + β₂DAR_it + β₃(ROA_it × DAR_it) + β₄FSIZE_it + β₅INSTOWN_it + β₆AS_it + β₇CR_it + β₈RG_it + β₉CTRY_DUM_it + u_it

(1)

In this model, DPO_it is the dependent variable measuring the dividend policy of firm i at time t, proxied by DPO. The explanatory variable measuring the leverage of firm i at time t is given by DAR_it. ROA_it is the moderating variable, and its interaction with the leverage variable (ROA_it × DAR_it) is included to capture its moderating effect on the relationship between leverage and dividend policy. Control variables describe the characteristics of the firm at time t, and they are firm size (FSIZE_it), asset structure (AS_it), liquidity (CR_it), revenue growth (RG_it), institutional ownership (INSTOWN_it), and country dummy (CTRY_DUM_it). Finally, u_it is the random error term for firm i at time t, accounting for unobserved firm-specific effects and idiosyncratic errors (Le et al. 2019; Le et al. 2019; Nguyen 2024).

3.3.2. Logistic Regression Model

The research employs a logistic regression model as a robustness check to examine the determinants of dividend policy, using the binary dependent variable DPO_binary as outlined in Table 1. This model examines the likelihood of dividend payments in connection with leverage, profitability, and the moderating effect of profitability. The logistic regression model is formulated as follows:

logit(P(DPO_binary_i = 1)) = α + β₁DAR_i + β₂ROA_i + β₃(DAR_i × ROA_i) + γ₁FSIZE_i + γ₂INSTOWN_i + γ₃AS_i + γ₄CR_i + γ₅RG_i + γ₆ CTRY_DUM_i + ε_i

(2)

In this model, logit (P(DPO_binary_i = 1)) denotes the probability that firm i will pay dividends. The coefficients β₁, β₂, and β₃ show the effects of the DAR, the ROA, and how they interact with each other without taking the time effect into account. This regression model also incorporates the same set of control variables with that of the Tobit regression.

By using this logistic regression model as a robustness check, this study ensures that the findings from the panel Tobit regression are consistent and that the relationships between leverage, profitability, and dividend policy hold true even when analysed through a different methodological approach. This dual approach enhances the credibility and reliability of this study’s conclusions, providing a comprehensive understanding of dividend behaviour across varying financial and market contexts (Barth and Clinch 2009; Petersen 2009; Wooldridge 2016).

4. Results and Discussion

4.1. Descriptive Statistics

Table 2. Combined summary statistics of Nigeria and South Africa.

Variable	Observations	M	SD	Min	Max
DPO	915	0.3508	1.9801	−33.0271	42.5646
DPO_binary	915	0.0492	0.2164	0	1
ROA	915	0.0391	0.1028	−0.5801	0.5768
DAR	915	0.5730	0.2870	0.0998	3.3710
FSIZE	915	15.7354	2.2005	6.9470	21.2047
INSTOWN	915	0.4477	0.2793	0	0.9600
AS	915	0.3759	0.2131	0.0113	0.9578
CR	915	1.5861	1.3614	0.0239	19.2508
RG	915	0.2327	1.3187	−0.9504	20.3942

Note. Authors’ compilation.

presents the combined summary statistics of the variables for manufacturing firms in Nigeria and South Africa. The average DPO was 0.35, indicating a modest level of payouts relative to firm profits with substantial variation between firms (min = −33.03; max = 42.56). The binary representation of DPO (DPO_binary), which takes a value of 1 for DPO greater than 1 and 0 otherwise, had a mean of 0.049, highlighting that only a small proportion of firms distributed substantial dividends. ROA averaged 3.91%, ranging from −58.01% to 57.68%, suggesting significant variation in firm profitability. The DAR had a mean of 57.30%, indicating that firms in both countries used substantial leverage in financing their assets. The average firm size (FSIZE) was 15.74, while institutional ownership (INSTOWN) was around 44.77%. The asset structure (AS), current ratio (CR), and revenue growth (RG) also displayed significant variation across firms, reflecting the diversity in firm operations and financial health in these emerging markets.

Table 3. Summary statistics of Nigeria.

Variable	Observations	M	SD	Min	Max
DPO	473	0.4781	2.2424	−3.9808	42.5646
DPO_binary	473	0.0571	0.2322	0	1
ROA	473	0.0350	0.1127	−0.5801	0.5396
DAR	473	0.6300	0.3353	0.1242	3.3710
FSIZE	473	16.1766	2.3279	6.9470	21.2047
INSTOWN	473	0.5021	0.2841	0	0.9600
AS	473	0.4488	0.2279	0.0123	0.9578
CR	473	1.3217	1.3969	0.0239	19.2508
RG	473	0.2230	1.3093	−0.9422	20.3942

Note. Authors’ compilation.

and

Table 4. Summary statistics of South Africa.

Variable	Observations	M	SD	Min	Max
DPO	442	0.2145	1.6456	−33.0271	4.6313
DPO_binary	442	0.0407	0.1979	0	1
ROA	442	0.0436	0.0909	−0.4196	0.5768
DAR	442	0.5120	0.2078	0.0998	2.9863
FSIZE	442	15.2632	1.9501	9.8684	19.4350
INSTOWN	442	0.3894	0.2620	0	0.9500
AS	442	0.2979	0.1635	0.0113	0.7182
CR	442	1.8690	1.2640	0.2172	11.4527
RG	442	0.2430	1.3301	−0.9504	16.4460

Note. Authors’ compilation.

present the summary statistics for Nigerian and South African firms, respectively. Nigerian firms, on average, had a higher DPO of 0.48 compared to 0.21 for South African firms, indicating more frequent or substantial dividend payments among Nigerian firms. The binary representation, DPO_binary, was also slightly higher in Nigeria (M = 0.057) compared to South Africa (M = 0.041), which further supports the trend of more generous dividend payments by Nigerian firms. Nigerian firms also showed higher leverage, with a mean DAR of 63.00%, compared to 51.20% in South Africa. This suggests that Nigerian firms relied more on debt financing. Additionally, Nigerian firms had a larger average firm size and higher institutional ownership compared to South African firms. However, South African firms displayed a higher current ratio (CR) of 1.87, indicating stronger liquidity positions on average, and a slightly higher average revenue growth (RG).

The descriptive statistics highlight that Nigerian firms tend to have higher dividend payouts, larger firm sizes, and more leverage than South African firms. However, South African firms exhibit better liquidity and revenue growth. These differences are indicative of diverse financial environments and corporate behaviours across the two countries, providing a basis for exploring how these characteristics affect dividend policy.

4.2. Correlation Analysis

Table 5. Combined correlation matrix of Nigeria and South Africa.

Variable	DPO	DPO_binary	ROA	DAR	FSIZE	INSTOWN	AS	CR	RG
DPO	1.0000
DPO_binary	0.3264	1.0000
ROA	0.0596	0.1654	1.0000
DAR	0.0046	−0.0267	−0.4091	1.0000
FSIZE	0.1305	0.1156	0.2614	−0.1279	1.0000
INSTOWN	0.0108	0.0053	−0.0201	−0.0657	0.0745	1.0000
AS	0.0445	−0.0434	−0.0787	0.1461	0.1602	0.1926	1.0000
CR	−0.0822	0.0282	0.1433	−0.3510	−0.1679	−0.0405	−0.2442	1.0000
RG	0.0099	0.0123	0.0183	0.0701	−0.0017	−0.0467	0.0054	0.0081	1.0000

Note. Authors’ compilation.

provides the correlation results for the combined dataset, which includes all sampled manufacturing firms from Nigeria and South Africa. The findings show several notable relationships. DPO and DPO_binary exhibit a moderate positive correlation of 0.326, suggesting that firms with higher dividend payouts are more likely to distribute significant dividends in absolute terms. ROA has a weak positive relationship with DPO (r = 0.0596), implying that more profitable firms tend to distribute higher dividends, although this relationship is not strong. The DAR does not have a strong relationship with the DPO (r = 0.0046), which suggests that leverage does not have a strong association with DPO across the whole sample.

FSIZE shows a weak positive correlation with DPO (r = 0.1305), indicating that larger firms are slightly more likely to pay dividends, which is consistent with the notion that larger firms typically have more stable cash flows, enabling them to sustain dividend payouts. The CR is negatively correlated with DPO (r = −0.0822), suggesting that firms with higher liquidity ratios tend to pay lower dividends, possibly due to a preference for maintaining internal reserves. INSTOWN and AS demonstrate minimal correlation with DPO, indicating that these factors may not directly influence dividend behaviour in the pooled sample of firms.

The country-specific correlation analyses (see

Table 6. Correlation matrix for Nigeria.

Variable	DPO	DPO_binary	ROA	DAR	FSIZE	INSTOWN	AS	CR	RG
DPO	1.0000
DPO_binary	0.4110	1.0000
ROA	0.0736	0.1936	1.0000
DAR	−0.0458	−0.0255	−0.5326	1.0000
FSIZE	0.1305	0.1362	0.4064	−0.3596	1.0000
INSTOWN	0.0119	0.0184	0.0721	−0.1764	0.1975	1.0000
AS	0.0121	−0.1012	−0.1208	0.1006	0.0735	0.1341	1.0000
CR	−0.0131	−0.0187	0.1352	−0.2728	−0.0932	0.0100	−0.2802	1.0000
RG	0.0223	0.0330	−0.0638	0.1702	0.0275	−0.0977	0.0314	−0.0572	1.0000

Note. Authors’ compilation.

for Nigeria and

Table 7. Correlation matrix for South Africa.

Variable	DPO	DPO_binary	ROA	DAR	FSIZE	INSTOWN	AS	CR	RG
DPO	1.0000
DPO_binary	0.1797	1.0000
ROA	0.0428	0.1263	1.0000
DAR	0.0765	−0.0563	−0.1560	1.0000
FSIZE	0.1011	0.0712	0.0707	0.1905	1.0000
INSTOWN	−0.0257	−0.0306	−0.1327	0.0034	−0.1922	1.0000
AS	0.0434	0.0094	0.0283	0.0313	0.1334	0.1320	1.0000
CR	−0.1638	0.1133	0.1422	−0.4341	−0.1862	−0.0128	−0.0415	1.0000
RG	−0.0063	−0.0125	0.1250	−0.0859	−0.0349	0.0127	−0.0245	0.0813	1.0000

Note. Authors’ compilation.

for South Africa) reveal differences in the relationships between variables, suggesting that the dividend behaviours may be influenced by country-specific factors such as financial regulations, market environments, and economic stability. In Table 5, the correlation results for Nigerian firms show a stronger positive relationship between DPO and DPO_binary (r = 0.4110) compared to the combined sample, suggesting that Nigerian firms that tend to pay dividends are more likely to make substantial payouts. The relationship between ROA and DPO in Nigeria (r = 0.0736) remains weak, indicating that profitability does not have a significant influence on dividend payouts in this country. Interestingly, the correlation between DAR and DPO in Nigeria is negative (r = −0.0458), implying that more leveraged firms tend to distribute lower dividends. This finding may indicate risk aversion in dividend policies among highly leveraged Nigerian firms.

In Table 7, the correlation results for South African firms highlight some differences compared to Nigeria. Specifically, the correlation between DPO and DAR is positive (r = 0.0765), suggesting that more leveraged South African firms are slightly more inclined to distribute dividends. This finding contrasts with the negative correlation found in Nigerian firms and could reflect differing attitudes towards risk management or varying access to external financing between the two countries. Additionally, the relationship between firm size and DPO is weaker in South Africa (r = 0.1011) compared to Nigeria, suggesting that firm size may play a lesser role in determining dividend payments in South African firms.

4.3. Regression Analysis Results

In this section, we present the findings of the regression analyses that aim to evaluate the impact of profitability, leverage, and control variables on the dividend policies of listed manufacturing firms in Nigeria and South Africa. As shown in

Table 8. Compact summary of regression results for panel Tobit and logistic models.

Variable	Combined_XTT	NGR_XTT	SA_XTT	Combined_Logit	NGR_Logit	SA_Logit
ROA	−0.487	−12.857	2.482 ***	6.018	−34.700 ***	21.345 ***
	(2.548)	(8.022)	(0.742)	(4.584)	(8.947)	(5.529)
DAR	−1.303 **	−2.922 **	0.043	−0.213	−5.114 ***	1.064
	(0.589)	(1.242)	(0.185)	(1.304)	(1.682)	(2.359)
DAR_ROA	7.923	29.575 **	−0.037	1.573	69.530 ***	−25.550 ***
	(4.123)	(13.753)	(1.215)	(7.884)	(15.538)	(9.673)
FSIZE	0.345 ***	0.550 ***	0.092 ***	0.321 ***	0.267 **	0.424 **
	(0.060)	(0.116)	(0.015)	(0.093)	(0.104)	(0.185)
INSTOWN	−0.085	−0.443	0.096	0.121	0.483	0.132
	(0.331)	(0.626)	(0.104)	(0.541)	(0.716)	(1.039)
AS	−0.167	−0.548	0.051	−1.829 **	−4.811 ***	0.091
	(0.555)	(0.958)	(0.163)	(0.888)	(1.561)	(1.494)
CR	0.056	0.040	0.083 ***	0.131	−1.168 **	0.276
	(0.065)	(0.115)	(0.023)	(0.091)	(0.507)	(0.153)
RG	−0.096	−0.066	−0.099 ***	0.015	0.120	−1.029
	(0.087)	(0.173)	(0.034)	(0.099)	(0.115)	(0.664)
CTRY_DUM	0.114			−0.166
	(0.256)			(0.382)
Intercept	−5.286 ***	−7.650 ***	−1.559 ***	−8.153 ***	−1.700	−11.613 ***
	(1.099)	(2.064)	(0.275)	(1.856)	(2.495)	(3.332)
sigma_u	0.827 ***	1.129 ***	0.000 ***
	(0.128)	(0.251)	(0.000)
sigma_e	1.957 ***	2.709 ***	0.500 ***
	(0.059)	(0.115)	(0.022)
rho	0.151	0.148	0.000
	(0.041)	(0.057)	(0.000)
Number of Observations	914	473	441	914	473	441
Rank of VCE	12	11	11	10	9	9
AIC	2824.192	1604.875	619.625	338.064	178.503	148.115
Pseudo R2				0.113	0.225	0.135
F-Statistic	34.980	22.540	15.782
Prob > Chi2	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000

Note. Authors’ compilation. ***: p < 0.01; **: p < 0.05; no asterisk: not significant (standard errors are provided in parentheses).

, the regression models further explore the moderating role of profitability in the relationship between leverage and dividend policies.

4.4. Discussion

4.4.1. Profitability on Dividend Policy

The first hypothesis (H1) posited a positive relationship between profitability and dividend policy in listed manufacturing firms in Nigeria and South Africa. This section examines whether profitability influences firms’ decisions to distribute dividends, which is critical in emerging markets where financial signalling plays a significant role.

The regression analysis for the combined sample of Nigerian and South African firms (Table 8) reveals a negative but statistically insignificant relationship between ROA and DPO (coefficient = −0.487, p > 0.05). This suggests that, across both markets, firms with higher profitability do not necessarily increase their dividend payouts. This finding offers support to the pecking order theory, which posits that firms prefer to use internal funds for reinvestment rather than distributing them as dividends. Conversely, this finding contradicts signalling theory, which suggests that profitable firms use dividends to signal financial strength and stability to the market. Additionally, the finding challenges the Agency theory viewpoint, which posits that dividends serve as a mechanism to reduce agency conflicts by distributing excess cash to shareholders, thereby limiting managerial discretion.

For the Nigerian segment, the results reveal an insignificant negative relationship between ROA and DPO (coefficient = −12.857, p = 0.109, as shown in Table 8). Given this statistical insignificance, it is difficult to definitively attribute the observed relationship to the pecking order theory. While the negative direction of the relationship aligns with pecking order expectations—suggesting that firms may prefer to retain earnings over distributing them—the lack of significance means that Nigerian firms may not consistently prioritise retained earnings for growth as theorised. This outcome partly aligns with Safiq and Yulianti (2023), who reported a significant negative relationship between ROA and dividend policy. In contrast, empirical research in other emerging markets, including studies by Amidu and Abor (2006) in Ghana and Arko et al. (2014) in sub-Saharan African nations, has revealed a significant positive correlation between profitability and dividend payout. This suggests that more profitable firms in these areas tend to distribute higher dividends.

Conversely, South African manufacturing firms demonstrated a significant and positive association between ROA and DPO (coefficient = 2.482, p < 0.01, as presented in Table 8). This positive correlation aligns with signalling theory. In South Africa’s more developed financial environment, profitable firms may sustain dividends to enhance investor confidence, aligning with findings by Lotto (2020) in Tanzania.

4.4.2. Leverage on Dividend Policy

The significant negative relationship between the DAR and DPO in the combined sample (coefficient = −1.303, p < 0.05) aligns well with existing theory, lending support to Hypothesis 2 (H2), which posits a negative relationship between DAR and dividend policy in listed manufacturing firms in Nigeria and South Africa. The findings are consistent with the agency cost framework established by Jensen and Meckling (1976), who argue that agency conflicts can emerge when managers have discretionary control over free cash flow, potentially leading to inefficient or self-serving investments. Jensen (1986) further elaborates that high leverage reduces free cash flow due to the burden of debt servicing, thereby limiting the opportunity for managerial opportunism. Consequently, highly leveraged firms are inclined to adopt conservative dividend policies, prioritising debt repayment over shareholder distributions to align managerial incentives with the interests of both shareholders and creditors.

This theoretical explanation is especially relevant for Nigerian firms, where the negative impact of DAR on DPO is even more pronounced (coefficient = −2.922, p < 0.05), strongly supporting H2. The financial environment in Nigeria, characterised by high borrowing costs and limited access to capital markets, demands stringent financial discipline. Here, the data suggest that highly leveraged Nigerian firms restrict dividend payouts to preserve cash for debt obligations, minimising agency conflicts and reflecting the practical necessity of efficient resource management. This behaviour corroborates empirical evidence from studies like Al-Najjar and Kilincarslan (2016), which emphasise the importance of debt management in emerging markets.

In contrast, the results for South African firms, where the relationship between DAR and DPO is insignificant (coefficient = 0.043, p > 0.05), challenge the generalisability of Jensen’s (1986) agency cost theory. The more developed and flexible financial markets in South Africa appear to mitigate the need for stringent financial discipline. This reduced agency concern may explain why high leverage does not significantly affect dividend policy, suggesting that South African firms face fewer constraints in managing debt and dividend payouts. The presence of better governance mechanisms and access to diverse financing options in South Africa likely play a crucial role, allowing firms to maintain a more balanced approach.

Therefore, while the findings for Nigeria support H2 and are consistent with the agency cost perspective, the results for South Africa indicate a more complex dynamic. In this context, other factors, such as signalling and financial market flexibility, may exert a more significant influence, reducing the pressure of agency concerns. This divergence underscores the importance of understanding market-specific conditions and the availability of alternative financial and governance mechanisms in shaping the leverage–dividend policy relationship.

4.4.3. Moderation of the Leverage and Dividend Policy Relationship by Profitability

This section tests Hypothesis 3 (H3), which posits that profitability positively moderates the relationship between leverage and dividend policy in listed manufacturing firms in Nigeria and South Africa. The combined model results reveal that the interaction term, DAR_ROA, has a positive coefficient of 7.923 but is not statistically significant (p > 0.05; Table 8). This suggests that, overall, profitability does not have a consistent or robust moderating effect on the leverage–dividend policy relationship when considering both Nigerian and South African firms together. Consequently, it appears that firms, on average, do not heavily rely on profitability to manage the trade-off between debt obligations and dividend payouts in a unified context.

However, a more nuanced understanding emerges from the segmented analyses, which provide stronger support for H3. In the Nigerian segment (NGR_XTT), the interaction term, DAR_ROA, is highly significant and positive, with a coefficient of 29.575 (p < 0.05). This strong moderation effect indicates that profitability enhances the ability of Nigerian firms to manage high leverage while sustaining dividend distributions. It aligns with signalling theory, which suggests that profitable firms use dividends as a signal of financial stability, even in the face of substantial debt. This finding emphasises the crucial role of profitability in Nigeria’s challenging financial environment, characterised by high borrowing costs and limited access to external financing. The results support H3, demonstrating that higher profitability enables Nigerian firms to meet their debt obligations while still rewarding shareholders—mitigating the adverse effects of leverage on dividend policy.

Conversely, the results for the South African segment (SA_XTT) show a negative coefficient of −0.505 for DAR_ROA, which is not statistically significant. This suggests that profitability does not significantly moderate the relationship between leverage and dividend policy in South Africa and may even have a slight weakening effect. The lack of significance implies that South African firms, operating in more developed and flexible financial markets, do not rely as heavily on internal profitability to maintain dividend payments. Instead, these firms may have more options for external financing, reducing the necessity for profitability to buffer the impact of high leverage on dividend decisions. This observation partially aligns with the pecking order theory, which highlights the prioritisation of internal over external financing.

These results underscore the contrasting roles of profitability as a moderating factor in emerging markets. In Nigeria, where financial constraints are more severe, profitability plays a critical role in enabling firms to manage leverage effectively while maintaining dividend payouts, which supports H3. This outcome illustrates the reliance on internal financial strength to balance debt and dividends in a constrained environment. In contrast, in South Africa, where firms benefit from more developed financial markets and access to diverse funding sources, the moderating role of profitability is less pronounced, reflecting a strategic difference influenced by market conditions and external financing options.

4.4.4. Effect of Control Variables on Dividend Policy

Table 8 also shows control variable analysis results, which help explain how firm characteristics affect DPO. Firm size positively correlates with DPO across all models. The coefficient for firm size is 0.345 (p < 0.01) in the combined model, 0.550 in Nigeria, and 0.092 in South Africa. This consistent significance suggests that larger firms pay higher dividends due to their financial stability and steady cash flow generation, which supports regular dividend distributions. Meanwhile, institutional ownership has no significant effect on DPO in any model, with coefficients of −0.085 in the combined model, −0.443 in Nigeria, and −0.096 in South Africa. This suggests that institutional ownership does not influence dividend policy in these markets. Similarly, no significant relationship exists between asset structure and DPO in all models, with coefficients of −0.167 in the combined model, −0.548 in Nigeria, and −0.051 in South Africa. It appears that a firm’s asset composition does not influence dividend payouts in the contexts studied.

However, the current ratio shows a significant positive association with DPO in the South African segment only, with a coefficient of 0.083 (p < 0.01). Since they can distribute surplus cash without risking operational funds, South African firms with higher liquidity are more likely to pay dividends. The current ratio is not significant in the combined and Nigerian models, suggesting liquidity does not consistently influence dividend policy across markets. Furthermore, revenue growth has a substantial adverse relationship with DPO in South Africa, with a coefficient of −0.099 (p < 0.01), suggesting firms with higher revenue growth may prioritise reinvestment over dividend distribution. RG is not significant in the combined or Nigerian models, indicating that revenue growth influences dividend policy differently across markets.

Finally, with a coefficient of 0.114, the country dummy variable in the combined model indicates no statistically significant country-specific effects on dividend policy for Nigeria and South Africa. This result suggests that nationality alone does not play a decisive role in shaping dividend policy decisions in these two countries. Instead, it points to the possibility that firm-level factors, such as leverage, profitability, and firm size, as well as broader macroeconomic conditions, are more critical determinants of dividend policy. The absence of a significant country effect highlights the convergence of certain financial practices across these emerging markets, possibly driven by globalisation, regional economic integration, or similar corporate governance frameworks. It also underscores the importance of understanding firm-specific and sector-specific dynamics when analysing dividend policy rather than relying solely on geographical distinctions.

4.5. Robustness Checks

To ensure the robustness of the main findings, logistic regression models were conducted using a binary form of the DPO (DPO_binary) as the dependent variable, as presented in Table 8. This section discusses the robustness results, focusing on the relationship between profitability, leverage, and dividend policy, analysed separately for the combined sample and the Nigerian and South African contexts.

In the combined logistic regression model, ROA remains positive but statistically insignificant, with a coefficient of 6.018. This indicates that, overall, profitability does not have a substantial direct impact on the probability of dividend payments when considering all firms together. The DAR remains negative and significant at the 1% level, with a lower coefficient of −0.213, suggesting that leverage continues to reduce the likelihood of dividend payments. The interaction term (DAR_ROA) is positive but remains statistically insignificant, with a coefficient of 1.573, indicating that the moderating effect of profitability on the leverage–dividend relationship is not strong in the combined model.

In the Nigerian segment, ROA remains negative and is now statistically significant at the 1% level, with a large coefficient of −34.700. This finding reinforces the negative relationship between profitability and the likelihood of paying dividends, suggesting that highly profitable firms in Nigeria prefer to reinvest earnings rather than distribute them as dividends, consistent with the pecking order theory. The DAR also remains negative and significant at the 1% level, with a reduced coefficient of −5.114, indicating that leverage continues to have a strong, adverse effect on dividend policy.

The interaction term (DAR_ROA) in Nigeria remains positive and is now significant at the 1% level with an amplified coefficient of 69.530. This result highlights the strong moderating effect of profitability, suggesting that profitable Nigerian firms are better able to manage high levels of debt while still maintaining dividend payments, supporting the idea that profitability can alleviate the financial constraints posed by leverage.

In the South African context, ROA remains positive and statistically significant at the 1% level, with a coefficient of 21.345. This suggests that profitable firms in South Africa are more likely to pay dividends, aligning with the signalling theory, where dividends are used to indicate financial strength to investors. In contrast, the DAR remains positive but statistically insignificant, with an increased coefficient of 1.064, indicating that leverage does not significantly influence dividend policy in South Africa.

The interaction term (DAR_ROA) in South Africa remains negative and becomes statistically significant at the 1% level, with a coefficient of −25.550. This finding suggests that the positive relationship between leverage and dividend policy is weakened by profitability, implying that highly leveraged but profitable firms in South Africa may prioritise financial stability over dividend payments, reflecting a more conservative approach to managing financial resources.

The logistic regression models demonstrate overall significance, as indicated by the prob > chi-squared values of 0.0000 across all models. The Nigerian model has the highest pseudo-R-squared value (0.225), suggesting that it explains dividend policy decisions more effectively than the combined model (0.113) and the South African model (0.135). Additionally, the Akaike information criterion values are lower for the country-specific models, suggesting a better model fit in these contexts.

Robustness checks confirm the key findings and reveal important nuances. In Nigeria, ROA remains negatively associated with dividend payments, now with stronger statistical significance, emphasising the preference for reinvestment over dividend distribution among profitable firms. Leverage continues to exert a negative influence on dividend policy, but the moderating effect of profitability is pronounced, allowing some firms to sustain dividend payouts despite high debt levels. In South Africa, profitability positively influences the dividend payout ratio, and the interaction between leverage and profitability becomes significantly negative, highlighting a more conservative approach to managing dividends.

5. Conclusions

This study provides a rigorous and comprehensive analysis of the factors influencing dividend policy among listed manufacturing firms in Nigeria and South Africa, focusing on the role of leverage and profitability. This study identifies distinct patterns and complexities in dividend policy management across these two economies. The findings reveal notable differences in how firms in these countries respond to financial leverage and profitability in shaping their dividend policies. In Nigerian firms, the stronger negative relationship between the debt-to-assets ratio (DAR) and the dividend payout ratio (DPO) highlights the acute financial constraints they experience. These constraints force firms to prioritise repaying debt over distributing dividends. Conversely, South African firms exhibit a less pronounced impact of leverage on dividend policy, reflecting a more developed financial market environment that offers greater flexibility in balancing leverage and shareholder returns.

Furthermore, this study demonstrates that profitability, measured by return on assets (ROA), plays a significant moderating role in the relationship between leverage and dividend policy in Nigerian firms. This indicates that higher profitability provides financial stability, enabling firms to sustain dividend payments even under high leverage. However, in South Africa, the moderating effect of profitability is weaker, suggesting that firms in a more advanced financial system rely less on internal profitability to support dividend distributions. These findings underscore the critical role of economic and institutional contexts in shaping corporate financial strategies in emerging markets.

This study also incorporates a robustness check through logistic regression analysis to examine the likelihood of dividend payments. This analysis reinforces the reliability of the results obtained from the Tobit regression model, confirming that the relationships among leverage, profitability, and dividend policy remain consistent across different methodological approaches. This methodological rigor enhances the credibility and robustness of this study’s conclusions, making its insights valuable for both theory and practice.

Policy recommendations derived from this research are crucial for financial regulators and corporate managers in both Nigeria and South Africa. For Nigerian policymakers, this study emphasises the importance of improving access to affordable credit facilities to alleviate the financial pressures on highly leveraged firms and enable more balanced dividend policies. In South Africa, policymakers should focus on maintaining financial market stability and ensuring firms have access to diverse financing options to support strategic growth and shareholder value. Additionally, corporate managers in both countries should align their dividend strategies with their specific financial and operational contexts, particularly by leveraging profitability to effectively balance debt obligations and shareholder returns.

Despite similarities in the manufacturing sectors of Nigeria and South Africa, the operational and financial strategies differ significantly. Nigerian firms face greater financial constraints, requiring them to carefully manage their resources to meet both debt and equity stakeholder expectations. In contrast, South African firms benefit from a more favourable financial environment, allowing for more flexible dividend policies. This study, therefore, emphasises that dividend policy formulation should be tailored to the unique financial, economic, and governance conditions of each country. These nuanced insights contribute to a better understanding of how financial environments shape corporate decisions in emerging markets.

However, this study has certain limitations that open avenues for future research. The focus on Nigeria and South Africa, while providing a rich comparative framework, may limit the generalisability of the findings to other emerging markets. The unique economic and financial structures of these countries highlight important dynamics, but extending the analysis to other African nations or developing economies could provide broader insights. Furthermore, the analysis does not explicitly address endogeneity issues such as omitted variable bias or reverse causality, which may influence the estimated relationships. Future research could employ advanced econometric techniques, such as dynamic panel models or the generalised method of moments (GMM), to address these issues and strengthen causal inferences.

This study’s emphasis on manufacturing firms offers valuable sector-specific insights but limits the applicability of the findings to other industries. Investigating dividend policies across diverse sectors, such as technology, services, or natural resources, would provide a more comprehensive understanding of the determinants and dynamics of dividend decisions. Additionally, incorporating country-level variables, such as inflation, GDP growth, and financial market development indices, could provide deeper insights into how macroeconomic factors interact with firm-level determinants to shape dividend policy.

Lastly, the economic context of the study period, characterised by fluctuating market conditions and economic shocks, may affect the robustness of the conclusions over time. Incorporating year dummies or expanding the study period to capture longer-term trends could enhance the reliability of the findings. Moreover, future research could explore whether similar dynamics exist in other regions of Africa or among countries with varying levels of financial market maturity. Addressing these limitations and extending the scope of the analysis could significantly enrich the understanding of dividend policy in emerging markets.

Despite these limitations, this research provides valuable insights into how leverage and profitability influence dividend policy in Nigeria and South Africa. It highlights the importance of tailored financial strategies and offers actionable recommendations for policymakers and corporate leaders. By addressing the limitations, future research can refine and extend these insights, offering a more comprehensive understanding of dividend policies in diverse economic and sectoral contexts.