Banks currently face fierce competition, both locally and internationally, because of the contemporary era's rapid technological advancements, particularly in the areas of information and communication technology, economic liberalization, privatization, and globalization.  To overcome severe competition, it needs to achieve a competitive advantage (CA) over others (Awwad & Qtaishat, 2023) and intellectual capital (IC) is increasingly recognized as a key driver of competitive advantage and financial performance (Mondal & Ghosh, 2012). Intellectual capital, which includes human capital, structural capital, and relational capital, plays a crucial role in enhancing efficiency, innovation, and customer trust (Barak & Sharma, 2024). 

Traditional financial indicators alone are no longer sufficient to capture the true drivers of performance; thus, understanding the contribution of IC to financial outcomes such as Return on Equity (ROE) has become a pressing concern for academics, regulators, and practitioners. The banking industry in Bangladesh provides a unique context to explore this relationship. As an emerging economy with a growing financial sector, Bangladesh's commercial banks are under increasing pressure to strengthen their performance, ensure stability, and enhance investor confidence. At the same time, banks face dynamic challenges such as rapid technological innovation, heightened competition, changing regulatory frameworks, and unforeseen shocks such as the COVID-19 pandemic. These developments have further underscored the importance of intangible resources, particularly intellectual capital, in maintaining resilience and sustaining growth. However, despite its significance, there is still a lack of clear understanding and empirical evidence on the direct and indirect impacts of intellectual capital on the financial performance of banking companies (Maditinos, 2011). The problem lies in the difficulty of measuring and managing intellectual capital effectively within the banking sector (Petty & Guthrie, 2000). The problem addressed in this research paper is the lack of empirical evidence on the impact of intellectual capital on the financial performance of private commercial banks in Bangladesh although intellectual capital is recognized as an important driver of organizational performance, particularly in the knowledge-intensive industries (Hasan and Karim, 2022; Faruq et al., 2023).  

According to Ting et al. (2020), research has shown mixed findings about the significance of the relationship between intellectual capital and the company's financial success, making it difficult to determine the specific nature of the connection. For instance, while some researchers stated that there was a relationship (Chen et al., 2005; Clarke et al., 2011), others argued that there was none (Chan, 2009). The present study will help to shed light on the controversy surrounding the nature of the connection between financial performance proxied by return on equity (ROA) and intellectual capital by answering the following question: What is the impact of Knowledgeable assets (intellectual Capital) on return on equity (ROE) of commercial banks in Bangladesh?

Intellectual Capital and Bank Performance

Intellectual capital (IC) broadly consists of human capital (skills, knowledge, experience of employees), structural capital (organizational processes, databases, systems), relational or customer capital, and capital employed efficiency. The Value-Added Intellectual Coefficient (VAIC™) is a commonly used framework to operationalize IC into components such as Human Capital Efficiency (HCE), Structural Capital Efficiency (SCE), and Capital Employed Efficiency (CEE). Several studies in banking contexts have found that IC is positively associated with financial performance. In Bangladesh Mollah & Rouf, (2022) find that HCE and CEE have statistically significant positive impact on bank performance (ROE and ROA), whereas SCE on its own is not always significant. Nabi, Gao, Rahman et al. (2020) (Bangladesh) report a positive relationship between VAIC and performance; among VAIC components, CEE often plays a strong role. Siraj-Ud-Dowla et al. examine non-financial as well as banking firms in Bangladesh and confirm that overall IC, HCE, and CEE enhance ROA and ROE. In Saudi Arabia, recent work finds that HCE is especially decisive for ROE (and NPM), with SCE and CEE also contributing Sayed & Nefzi, (2024). While intellectual capital (IC) is often associated with value creation, several studies have identified its negative or insignificant impact on firm profitability. One explanation is that investments in human and structural capital require significant financial resources, which may reduce short-term returns (Al-Musali & Ismail, 2014). Excessive spending on employee training, knowledge management systems, or technology upgrades may not yield immediate benefits, thereby exerting downward pressure on profitability (Kamath, 2015). Moreover, poor management of intellectual resources can lead to inefficiencies, knowledge redundancy, and higher operating costs, diminishing financial performance (Smriti & Das, 2018; Hossain et al., 2024). In emerging economies, where financial markets and innovation ecosystems are less developed, and intellectual capital may fail to translate into competitive advantage, resulting in a negative association with profitability (Singh et al., 2016). Thus, despite its strategic importance, intellectual capital can adversely affect firm profitability when not efficiently managed.

H1: Intellectual Capital is positively associated with banks performance

Firm size and Profitability

The relationship between firm size and profitability has been widely studied, but findings remain mixed. Several studies suggest that larger firms tend to achieve higher profitability (Hossain, 2019; Hossain & Saif, 2019; Hossain et al., 2018) due to economies of scale, stronger market power, and easier access to finance, which lower costs and enhance returns (Lee, 2009; Majumdar, 1997). For instance, larger firms can spread fixed costs across greater output and negotiate better terms with suppliers and creditors, resulting in higher return on assets (ROA) and return on equity (ROE). However, other research highlights that increasing firm size may generate diseconomies of scale, higher agency costs, and managerial inefficiencies that erode profitability, especially in mature or bureaucratic organizations (Shepherd & Wiklund, 2009; Goddard et al., 2005). Evidence of a non-linear relationship has also emerged, showing that profitability improves with firm size up to a threshold, after which further growth reduces performance (Hossain et al., 2024; Amato & Burson, 2007; Serrasqueiro & Nunes, 2008). These contrasting perspectives indicate that the effect of firm size on profitability is contingent on industry structure, governance quality, and the stage of firm development. So, we assert that 

H2: Firm size is positively associated with ROE of banks

Firm Age and Profitability

Firm age is often argued to have a positive association with profitability, as older firms typically benefit from accumulated experience, established reputations, and stronger market positioning (Hossain, 2019). Over time, firms develop more efficient processes, improve managerial expertise, and cultivate stable relationships with customers and stakeholders, which can enhance financial outcomes. In the banking sector, age may provide an additional advantage in terms of credibility, customer trust, and regulatory compliance, all of which contribute to sustainable profitability. Empirical studies support this view. For example, Coad et al. (2013) highlight that older firms may demonstrate more stable and improved performance due to learning effects and survival of the fittest mechanisms. Similarly, Stinchcombe, (2000) argues that organizational survival over time reduces “liability of newness,” enabling older firms to achieve efficiency and financial success. In the context of banking, Bektas, (2014) found that bank age positively influences profitability, as longer operational histories help build customer loyalty and risk management capabilities. Furthermore, Pervan et al. (2017) provided evidence from Croatian banks that older firms often record higher profitability due to experience and established networks. On the other hand, firm age is negatively related to bank performance. For instance, Majumdar, (1997) observed a negative or insignificant relationship between firm age and performance, noting that older firms may suffer from bureaucratic inefficiencies and rigidities that offset the benefits of experience. Similarly, Coad et al. (2013) argue that firm age alone is not a reliable determinant of profitability, as market dynamics and managerial efficiency often play more significant roles. In the context of banking, Pervan et al. (2014) found that firm age had an insignificant effect on bank profitability in selected transition countries, reinforcing the idea that operational efficiency and external market factors outweigh age as performance drivers. Based on this reasoning and prior empirical evidence, the following hypothesis can be developed:

H3: Firm age has a positive and significant association with profitability

COVID-19 and Profitability

The COVID-19 pandemic has had a profound negative impact on the financial performance of firms and banks, though the severity varied across industries and regions. Empirical evidence shows that firms in contact-intensive sectors such as tourism, airlines, and retail suffered sharp declines in profitability and market value, while financially constrained and highly leveraged firms were particularly vulnerable (Shen et al., 2020; Ding et al., 2021). Banks experienced declining returns on assets (ROA) and equity (ROE) due to increased credit risk, compressed net interest margins, and rising loan loss provisions, although the scale of deterioration was mitigated by policy interventions such as loan moratoria and government guarantees (Demirgüç-Kunt et al., 2021; Beck & Keil, 2022). At the same time, firms with stronger liquidity positions, digital readiness, and sustainability performance demonstrated greater resilience and faster recovery, highlighting the role of financial flexi-bility and intangible resources in crisis management (Amore et al., 2022; Albuquerque et al., 2020). Overall, the literature suggests that while COVID-19 significantly disrupted firm and bank performance, precrisis fundamentals and policy responses shaped the extent of the impact.

H4: The COVID-19 period is negatively associated with ROE for banks

This study follows a quantitative and longitudinal research approach to attain the purpose of the project. The listed commercial banks in the Dhaka Stock Exchange are the population of our study and in 2016, there were 30 listed commercial banks (now 35). The necessary data were collected from the listed commercial banks of Bangladesh during 2016-2023. As the data of this study reflect both time and firm inclusion, so panel data statistical analysis was considered, and Strata Software 15.0 was used. To determine which of the three models (pooled regression, fixed effect, and random effect) would be most appropriate for analysing cross-sectional time-series data, researchers conducted several techniques to assess the relationship between the study variables. These procedures included the following:

1. To analyse data trends, means, standard deviations, and value maxima and minima were employed.
2. Comparisons between the pooled regression model and the panel data model were made using the Breusch and Pagan Lagrange multiplier test.
3. The effectiveness of the random effects model and the fixed effects model was compared using the Hausman test.

Table 1: Measurement of the Variable of the study.

Diagnostic (Specification) tests 

Additionally, the researchers conducted diagnostic (specification) tests. This is to confirm the model's validity and make sure the cross-sectional time-series data complies with the following assumptions of the linear regression model:

1. To confirm that there is no correlation between the dependent variable's value in the current period (t) and its value in the prior periods, autocorrelation or serial correlation was used.
2. The test of the normal distribution of errors was conducted. This is to guarantee a normal distribution of the estimation errors.
3. To make sure that two or more independent variables do not have a strong linear relationship, a multicollinearity test was performed.
4. To make sure that the variance in the regression errors is homogenous (constant), a heteroskedasticity test was conducted.

Explanation of MVAIC components

Regression model: 

Model -1: Y(ROE) = α +β1MVAIC+ β2F_Size + β3F_Age + β4COVID- 19+ ε

Descriptive statistics 

Table 2 presents the descriptive statistics for the variables used in the study. 

Table 2: Showing the descriptive statistics of the study.

Value-Added Intellectual Coefficient (MVAIC), a measure of intellectual capital efficiency, has a mean of 3.68 with a standard deviation of 1.20, indicating moderate variation across the sample banks. Firm size (F_Size), measured in logarithmic terms, has a mean of 11.34 and a low standard deviation (0.38), suggesting that most banks in the sample are of relatively similar size Firm age (F_Age) ranges widely from 8 to 52 years, with an average of approximately 24 years, reflecting a mix of relatively young and mature banks. The COVID-19 dummy variable shows a mean of 0.25, implying that roughly 25% of the observations correspond to the COVID-19 period. The dependent variable, return on equity (ROE), averages 16.7% with a standard deviation of 12%, while the minimum and maximum values indicate some banks experienced negative profitability, reflecting heterogeneity in financial performance. Overall, these descriptive statistics highlight variation across both firm-specific characteristics and performance measures, which provide a basis for further regression analysis.

Test of Multicollinearity and Autocorrelation

Table 3 presents the results of multicollinearity and autocorrelation tests for Model 1, where return on equity (ROE) is the dependent variable.

Table 3: Test of Multicollinearity and Autocorrelation.

Multicollinearity: The Variance Inflation Factor (VIF) and Tolerance statistics are used to assess multicollinearity among the independent variables. Generally, a VIF above 10 or a tolerance below 0.1 indicates problematic multicollinearity. In this model, the highest VIF is 4.347 for MVAIC, and the lowest tolerance is 0.230, which are well within acceptable limits. Other variables, such as logTA (VIF = 1.649), Age (VIF = 1.744), and Covid_19 (VIF = 1.033), also show low multicollinearity. This indicates that the independent variables are not highly correlated, and the regression estimates are reliable.

Autocorrelation: The Durbin-Watson (DW) statistic is 1.507. Since DW values near 2 suggest no first-order autocorrelation, this value indicates that there is no serious autocorrelation in the residuals. Although slightly below 2, it is within a range that is generally considered acceptable for cross-sectional data or panel data with small time dimensions.

The impact of Intellectual capital on return on equity (ROE)

This section presents the relationship between intellectual capital and returns on equity (ROE). The results from Table 3 are shown as follows.

Examining Random Effects: The Lagrange Multiplier Test (Pooled OLS and Random Effects)

Table 4 presents the Bruesch-Pagan Lagrange Multiplier Test. It shows that the null hypothesis is rejected as the chi-square value (x2) is 181.63, Probability= 0.000, indicating that the variance across the companies is different (Not equal to zero). It suggests that the random effects model is more valid than Pooled OLS model.

Examining Fixed Effect Model: Hausman Specification Test

Hausman specification test is also done to determine the suitable model between the random effects model and fixed effects model. It is seen that individual effects were not significantly correlated with other regressor in the model (chi X2  =-7.26; Probability = 0.1230). The null hypothesis is accepted and indicates random effects model is more appropriate than the fixed effects model (Table 4).

Regression results

Across all three models, MVAIC has a positive and statistically significant coefficient (Pooled OLS = 0.0639***, Random Effects = 0.0795***, Fixed Effects = 0.0826***). This indicates that higher intellectual capital efficiency is associated with higher profitability in banks. The significance at the 1% level demonstrates a robust relationship between MVAIC and ROE, consistent with prior research suggesting that efficient management of human, structural, and relational capital enhances financial performance.

This result highlights that banks with more effective utilization of intellectual capital-comprising human, structural, and relational capital, tend to achieve higher profitability for their shareholders. The 1% significance level suggests that the probability of this relationship occurring by chance is extremely low, lending high confidence to the result. Prior studies reinforce this finding. For example, Chen et al. (2005) demonstrated a positive and significant association between intellectual capital and market and financial performance in Taiwanese listed firms. Similarly, Maji and Goswami, (2017) found that intellectual capital efficiency significantly enhances profitability and competitive advantage in Indian banks. More recently, Smriti and Das, (2018) confirmed that MVAIC has a strong positive effect on financial performance indicators, including ROE, in the Indian banking sector. These studies suggest that intellectual capital is a strategic asset that enables banks to innovate, manage risks effectively, and leverage knowledge resources for sustained profitability. 

Table 4: Regression Analysis: MVAIC and ROE.

Robust t-statistics/ Z statistics are reported in the parentheses

***; *** and * indicate statistically significance at the 1%, 5% and 10% level

In this context, the robust link between MVAIC and ROE underscores the growing importance of intangible assets in driving shareholder returns, particularly in knowledge-intensive industries like banking. Firm Size (F_Size): The coefficient for firm size is positive in all models but statistically weak. In the Pooled OLS model, it is marginally significant at the 10% level (0.0375*), while in the panel models, it is not statistically significant. This suggests that size may have a limited or inconsistent effect on bank profitability in the sample. The effect of firm age is negligible across all models and is not statistically significant, implying that the number of years a bank has been operating does not strongly influence ROE in this context except in Pooled OLS model. It suggests that the longevity of a bank does not necessarily translate into higher profitability. While older firms may benefit from accumulated experience, established customer bases, and reputational capital, these advantages may not always yield superior financial performance in highly competitive and regulated banking environments. Several studies align with this view, showing that firm age has little or no direct effect on profitability ( Majumdar, 1997;  Coad et al., 2013; Pervan et al., 2014; Hossain & Saif, 2019; Ratan et al., 2020). The only exception in your analysis-the Pooled OLS model showing some effect-could be due to model specification and not accounting for unobserved heterogeneity across banks, which fixed and random effects models handle more effectively. Hence, consistent with prior literature, the negligible role of firm age suggests that profitability in banks depends more on strategic, structural, and regulatory factors than on the sheer number of years in operation. The COVID-19 dummy variable shows a negative coefficient across all models, though it is not statistically significant. This suggests that while the pandemic may have exerted downward pressure on bank profitability, the effect is not strong enough to be conclusive in this sample.

It may be inferred from the test findings and discussions in the preceding section that there is a statistically significant influence of ICP (MVAIC) on the company's financial performance over the eight years of observations, from 2016 to 2023. The results are in line with the previous research conducted by Mondal, A., & Ghosh, S. K., 2012; Alowaimer et al., 2025). Firm size is positively and significantly connected with financial performance proxied by return on equity. It denotes that older companies have in competitive advantage in earning revenues compared to younger companies (Lee, 2009; Majumdar, 1997). Theoretically, intellectual capital is not merely a passive resource but an active driver of competitive advantage and superior financial performance. Integrating IC into strategic management practices enables firms to innovate, adapt, and thrive in competitive markets. The positive impact of IC on financial performance is supported by multiple theoretical frameworks, which highlight the strategic role of knowledge, relationships, and organizational processes. The findings of this study connect prominent theories like the resource-based view (RBV) of a firm, knowledge-based theory, stakeholder theory, signaling theory, and dynamic capabilities theory. The practical implications of the relationship between Intellectual Capital (IC) and Financial Performance (FP) provide valuable insights for managers, investors, policymakers, and organizations. Understanding these implications helps firms leverage their intangible assets to achieve better financial outcomes. Organizations should invest in employee development, training, and retention programs because skilled and motivated employees (human capital) directly enhance productivity and profitability. 

This study also suffered from some limitations. For instance, existing studies predominantly rely on annual reports and financial statements, which may not fully capture the qualitative aspects of IC, such as employee expertise, innovation culture, and leadership quality. The reliance on secondary data also raises concerns about accuracy and completeness. The influence of regulatory changes, economic conditions, competition, and technological advancements is often overlooked, limiting the practical implications of research findings. Although the MVAIC model is an improvement over VAIC, it still simplifies intellectual capital into a few components (HCE, SCE, RCE, CEE). It does not account for factors like customer satisfaction, digital transformation, or knowledge-sharing culture. Most research is confined to the banking sector, missing comparative insights that could be drawn from related industries such as insurance, fintech, or microfinance institutions. 

Practically, Intellectual Capital is a critical driver of financial performance that needs active management, investment, and integration into business strategy. Firms that harness their intellectual assets-people, processes, and relationships - will achieve better financial outcomes, higher innovation, and sustainable growth.

Conceptualization: M.S.H.: and A.A.S.; Methodo-logy: A.A.S.:  and D.C.; Software: M.S.H.: and M.E.I.; Validation: D.C.; Formal Analysis: M.S.H.; Investigation and Resources:  D.C.; Data Curation: A.A.S.; Writing–Original Draft Preparation: M.S.H.; Writing–Review & Editing: M.E.I.: and D.C.; Visualization: M.E.I., Supervision: M.S.H.: and A.A.S. Authors have read and agreed to the published version of the manuscript.

Authors received funding from University of Rajshahi under the annual development Programme in Financial Year 2023-2024.

The researchers express sincere gratitude to the Faculty of Business Studies, University of Rajshahi, for providing financial support under the Annual Development Programme (ADP) in the 2023–2024 Financial Year. This generous funding greatly facilitated the successful completion of the research project by enabling necessary data collection, analysis, and related activities. The researcher deeply appreciates the faculty's continued commitment to promoting academic research and scholarly excellence.

The authors declare no conflict of interest.