The findings concern a sector that provides savings and credit services to communities and micro, small and medium-sized enterprises, particularly where access to conventional banking remains limited. The article reports that 220 Indonesian rural banks failed between 2016 and mid-2023, even as the sector expanded in customers, third-party funds, credit distribution and total assets. Reported causes of failure included credit misuse, non-performing loans, misuse of customer deposits, embezzlement, collateral irregularities, violations of lending limits and liquidity problems. The study therefore examined sustainability in terms of economic, environmental and social performance, while representing governance through transparency, accountability, responsibility, independence, fairness and entrepreneurial culture. Sustainable performance was measured with a six-point questionnaire scale rather than solely through financial ratios or institutional survival.

Information technology adoption had a positive and significant association with good corporate governance, with a standardized path coefficient of 0.130 and a p-value of 0.016. Its direct association with sustainable performance was positive but did not meet the conventional 5% significance threshold: the coefficient was 0.092 and the p-value was 0.059, making the direct result statistically inconclusive rather than evidence of a negative relationship. The indirect association through governance was significant, with a coefficient of 0.026 and a p-value of 0.044, while the total effect was also significant at 0.118 with a p-value of 0.011. This pattern was consistent with indirect-only mediation in the estimated model, meaning that the technology–performance relationship was more evident through governance than through a statistically detectable direct pathway. The mediated effect was nevertheless modest, and the technology-adoption model accounted for little variation in the governance construct itself.

Internal control was the strongest direct predictor among the organizational capabilities examined. Its coefficient for sustainable performance was 0.325, with a p-value below 0.001, and its relationship with good corporate governance was also positive and significant, at 0.260 with a p-value below 0.001. The internal-control measure comprised control activities, information and communication, and monitoring, which the article links with risk detection, compliance, accountability and protection of organizational assets. Its indirect association with sustainable performance through governance was significant at a coefficient of 0.052 and a p-value of 0.001. Because both the direct and indirect paths were significant, the results were consistent with partial mediation. The total effect was 0.377, combining the direct coefficient of 0.325 with the governance-mediated contribution. Statistical strength did not imply an outsized practical influence, however: the effect size for internal control on sustainable performance was 0.078, within the study’s small-effect range.

Human capital capability followed a different pathway. Knowledge, experience and skills were positively associated with sustainable performance, with a coefficient of 0.133 and a p-value of 0.031, but the effect size was only 0.013. The capability did not have a statistically significant association with good corporate governance: its estimated coefficient was 0.051, with a p-value of 0.288 and an effect size of 0.002. Its indirect effect through governance was also non-significant, at a coefficient of 0.010 and a p-value of 0.301, while its total effect on sustainable performance was significant at 0.143 with a p-value of 0.018. The results suggest that employee capabilities may support credit assessment, risk management, customer service and problem-solving without automatically changing formal governance structures. They do not establish that governance is unnecessary for mobilizing human capital, only that the governance construct measured in this study did not significantly transmit that relationship.

Good corporate governance was itself positively associated with sustainable performance, with a coefficient of 0.200 and a p-value below 0.001, and an effect size of 0.031. In the model, governance combined transparency, accountability, responsibility, independence, fairness and entrepreneurial culture. Entrepreneurial culture extended the measure beyond a compliance-only perspective by including initiative, opportunity recognition, innovation and responsible risk-taking, although it remained a dimension of governance rather than a separate construct. The positive governance–performance association was conditional on environmental dynamism: the interaction between governance and dynamism had a coefficient of 0.123 and a p-value of 0.004, indicating that the estimated governance–performance relationship was stronger as market, technological and regulatory conditions became more rapidly changing or unpredictable. The interaction effect was small, with an effect size of 0.018, and the study did not report simple slopes at specified dynamism levels.

The model provided a partial explanation of the outcomes. Information technology adoption, internal control and human capital capability together accounted for 5.2% of the variance in good corporate governance, with an R-squared value of 0.052, which the authors classified as weak explanatory power. Adding the three capabilities, governance and environmental dynamism explained 29.1% of the variance in sustainable performance, also classified as weak under the study’s adopted benchmarks. The model’s predictive relevance was small for governance and moderate for sustainable performance, but no independent holdout or alternative out-of-sample prediction procedure was reported. The low governance R-squared indicates that important influences were omitted, potentially including bank size, organizational age, location, ownership, resources and regulatory characteristics. These variables were described in the sample but were not included as controls, so their independent effects cannot be inferred.

The evidence came from an electronic survey of Indonesian rural banks that had adopted information technology, including conventional and Islamic institutions. Of 1,325 targeted banks, the researchers received 526 completed responses and retained 436 after screening; one senior-management representative answered for each institution. Most participants were directors, and 94.5% of the participating banks had operated for more than a decade. Java represented 60.1% of the sample, while 91.5% of the institutions were conventional and 8.5% were Islamic. The questionnaire was pretested with 35 managers, after which three sustainable-performance indicators with loadings below 0.70 were removed, leaving 63 indicators. In the main sample, first-order loadings ranged from 0.784 to 0.967, and the higher-order measurement model met the specified reliability, validity and collinearity criteria. A Harman single-factor test found that the first component explained 27.303% of indicator variance, below the conventional 50% screening threshold, although the authors cautioned that this diagnostic cannot rule out common-method variance.

Several limitations constrain how far the findings can be generalized. Recruitment through institutional contacts and professional networks, together with voluntary participation, may have introduced coverage and self-selection effects, and the absence of comparable data for nonresponding banks prevented a formal test for non-response bias. Reliance on a single respondent and self-reported measures also allowed for single-informant bias. No multi-group analysis tested whether relationships differed between conventional and Islamic banks or among regulatory classifications, partly because the Islamic-bank subgroup contained only 37 observations. The results therefore apply most directly to the pooled sample and do not show that the same relationships operate equally across all bank types. The absence of simple-slope estimates also limits the detail of the conditional governance result, while the cross-sectional design prevents claims about whether the observed relationships persist over time.

For rural banks, the results offer a more evidence-based division of organizational roles than the idea that any single investment can deliver sustainability. Information technology was associated with improved governance and showed a modest performance association through that pathway, but it had no statistically significant direct association at the 5% level. Internal control had both direct and governance-mediated associations with sustainable performance, making it the strongest direct capability in the model, although its practical effect remained small. Human capital was associated directly rather than significantly through governance. Governance was relevant in its own right and was more strongly associated with sustainable performance in more dynamic environments, although the interaction was also small. The observed pattern therefore favors complementarity among digital systems, controls, employee expertise and governance, rather than treating any one factor as sufficient. These findings come from one national survey and do not prove that installing a particular system or practice will cause an individual bank to become sustainable.

Subject of Research: Organizational capabilities, governance and sustainable performance in Indonesian rural banks

Article Title: Governing for sustainability in a dynamic environment: the roles of corporate governance, organizational capabilities and entrepreneurial culture in Indonesian rural banks

Article References: Atmoko, G. D. P., Sasmoko, Maulana, A. E., & Rahim, R. K. (2026). Governing for sustainability in a dynamic environment: the roles of corporate governance, organizational capabilities and entrepreneurial culture in Indonesian rural banks. Frontiers in Sustainability, 7, Article 1983486. https://doi.org/10.3389/frsus.2026.1983486

Image Credits: AI Generated

DOI: 10.3389/frsus.2026.1983486

Keywords: Indonesian rural banks, sustainable performance, corporate governance, internal control, information technology, human capital, environmental dynamism, PLS-SEM, financial inclusion, Governing, sustainability, dynamic