This study employs panel data from 32 Indonesian provinces over the period 2017–2023. Two provinces, DKI Jakarta and DI Yogyakarta, are excluded as they exhibit substantially higher HDI and ICT levels compared to other provinces, indicating potential high-leverage observations that may disproportionately influence the estimation results. This exclusion is applied to obtain more representative estimates for the majority of provinces, and robustness checks including all provinces yield consistent directional results. The dependent variable is the inclusive economic growth indicator (IG), obtained from pillar 2 data on a scale of 1 to 10. This index comprises sub-pillars of income equality and poverty alleviation, derived from Ministry of National Development Planning (2025). The independent variable is the Information and Communication Technology Development Index (ICT-DI). The ICT-DI data was sourced from Statistics Indonesia annual publication (Statistics Indonesia, 2019-2024). The HDI data derived from the Statistics Indonesia annual reports (2024).

Model selection follows standard panel data procedures, including the Chow, Hausman, and Lagrange Multiplier (LM) tests, followed by relevant classical assumption tests. The Chow test is utilized to choose the appropriate model in panel data regression by comparing the common effect and fixed effect models. The Hausman test evaluates the relationship between individual effects and independent variables in the model. The Lagrange multiplier test is utilized to ascertain the more effective model between random effects and common effects models. If the selected model is categorized as Ordinary Least Squares (OLS), such as the fixed effect model or the common effect model, the classical assumption tests performed are limited to the heteroscedasticity test and the multicollinearity test. When utilizing the Generalized Least Squares (GLS) model, including the random effects model, the standard assumption tests involve the residual normality test and the multicollinearity test.

The empirical model is constructed based on endogenous growth theory and human capital theory, which emphasize the role of technological progress and human capability in driving inclusive economic outcomes. From a theoretical perspective, digitalization enhances productivity and access to economic opportunities, thereby promoting inclusive growth. However, the extent of this impact depends on the level of human development, which determines a region’s capacity to adopt and effectively utilize digital technologies. From an empirical perspective, this relationship is operationalized by specifying ICTDI as the main explanatory variable and HDI as a moderating variable, allowing the estimation of both direct and interaction effects.

The baseline model examining the impact of digitalization on inclusive economic growth is specified as follows:

(1)

The model is estimated using panel data approaches, including the Common Effects Model (CEM), Fixed Effects Model (FEM), and Random Effects Model (REM). Model selection is based on the Chow test, Hausman test, and Lagrange Multiplier test.

To examine the moderating role of HDI, a moderated regression model is employed as follows:

(2)

Where:

= Inclusive economic growth in region period

= ICT development index in region period

= Human development index in region period

= constant

= regression coefficients

= error term

To reduce potential multicollinearity arising from the interaction term, all independent variables are mean-centered prior to constructing the interaction variable. This procedure ensures more stable coefficient estimates and facilitates interpretation of the moderating effect.

This study's findings reveal that digitalization has a positive and significant effect on inclusive growth. This finding is consistent with endogenous growth theory (Romer, 1990), which emphasizes that technological progress enhances productivity and expands economic opportunities. This suggests that digitalization can act as a mechanism to promote social and economic inclusion ((Basnayake et al., 2024); (Ren et al., 2022)). Empirical research demonstrates that digital finance, digital governance, and ICT infrastructure correlate with improvements in inclusive growth metrics ((Daud & Ahmad, 2023); (Li et al., 2024); (Wang et al., 2023)). This suggests that digitalization improves inclusivity by reducing barriers to access and enabling broader participation in economic activities. A significant focus of digitalization is the transformation of the public sector. Digital public administration is increasingly associated with improved equitable and sustainable development outcomes, such as better budget management and increased accessibility of services (Kwilinski et al., 2023). Moreover, digital finance and e-governance can boost income generation, improve fiscal efficiency, and promote educational achievement, all of which are essential components of inclusive growth (Wang et al., 2023). Digital inclusive finance operates through intermediary processes, encompassing human capital development and innovative activities (Yang et al., 2024). These findings are consistent with the results of this study, which show that ICT development significantly enhances inclusive growth across Indonesian provinces by expanding access to economic opportunities and public services.

The findings of (Nchake & Shuaibu, 2022) show that ICT infrastructure has a positive and significant impact on inclusive growth in Africa, with an average increase of 0.4% to 0.7%, using robust empirical analysis covering 46 countries from 2000 to 2019. Internet access, broadband, and mobile connectivity have shown benefits across all economic strata, especially for the underprivileged (Kouladoum, 2023). Furthermore, energy, transportation, and digital infrastructures have a redistributive effect, diminishing wealth inequalities (Mutiiria et al., 2020). In contrast, (Ofori et al., 2024) illustrate that digital infrastructure does not significantly contribute to inclusive growth, highlighting the importance of supportive institutions and additional investments. An agreement exists regarding the necessity of extensive infrastructure development to effectively realize the potential of digitalization ((Kouladoum, 2023); (Nchake & Shuaibu, 2022)). These findings from other developing regions provide important context for understanding the Indonesian case, where similar patterns of uneven digital benefits are observed. This aligns with the findings of this study, where the positive impact of digitalization is evident, but its effectiveness varies depending on regional conditions, particularly differences in human development levels.

However, the positive effect of digitalization on inclusive growth observed in the aggregate analysis may mask important heterogeneity across regions with different levels of human development. The literature has increasingly highlighted the importance of a comprehensive approach in assessing inclusive growth. (Sharma, 2019) emphasizes that inclusive economic growth must involve raising incomes, expanding access to quality employment, and improving essential services such as education and healthcare. In this regard, indices such as the HDI are especially relevant as a metric of regional capacity to adopt and utilize the benefits of digitalization.

This study's findings on moderation regression analysis show that the interaction between digitalization and HDI has a negative and significant effect on inclusive growth. This indicates that higher levels of HDI weaken the inclusivity-enhancing effect of digitalization, confirming that its impact is not uniform across regions with different levels of human development. Importantly, this negative moderating effect is particularly evident in provinces with higher HDI levels, where digitalization does not necessarily translate into more inclusive outcomes. Instead, the benefits of digitalization tend to be concentrated among specific groups, particularly those with higher education and digital skills.

This finding highlights that higher human development does not automatically guarantee equitable distribution of digital benefits, but may instead intensify inequality under certain conditions. This result can be explained by skill-biased technological change, where digitalization disproportionately benefits individuals with higher education and digital skills. As a result, the gains from digital transformation are concentrated among more skilled groups, while lower-skilled individuals face limited access to digital opportunities. In high-HDI regions, digitalization also accelerates labor market transformation, including automation and job restructuring, which reduces employment opportunities for routine and low-skilled workers. Consequently, this leads to widening income disparities, unequal access to employment, and limited participation of vulnerable groups in digital economic activities, thereby weakening inclusive growth.

The evidence from previous studies highlights a complex and at times non-linear relationship. Multiple studies suggest that although digitalization may initially obstruct inclusive outcomes, its long-term benefits become apparent once essential thresholds of infrastructure and human capital are attained. (Kwilinski et al., 2023) discovered a U-shaped relationship between government digitalization and inclusive growth, while (Xiang et al., 2022) found an inverted U-shaped correlation between the rise of the digital economy and low-carbon, inclusive growth in China. These findings indicate diminishing returns or potential negative externalities at high levels of digitalization, such as increased digital inequality or technological unemployment. The role of governance remains a topic of discussion: while effective governance can promote digitalization (Daud & Ahmad, 2023), certain data suggests it may hinder its effectiveness under particular circumstances (Nguyen, 2021). The varied results emphasize the importance of contextually informed and adaptable strategies.

A substantial segment of the research highlights the importance of understanding initial conditions and contextual factors in evaluating the effects of digitization on equitable growth. (Wiston & Edgar, 2019) argue that innovation-driven growth requires adequate institutional preconditions and human capital competencies to mitigate future inequalities. Research by the (Bank, 2025), indicates that effective digitalization promoting equitable growth consists of three fundamental components: digital connectivity, digital skills, and inclusive digital public services. In developing countries, deficiencies in digital skills and inclusive digital services can intensify social and economic disparities.

(Kou, 2024) indicates that the effect of digitalization on income equality may be affected by the labor market. They suggest that automation and digitalization could replace conventional employment, especially in informal or labor-intensive sectors, unless accompanied by reskilling and the improvement of digital competences. This insight highlights that the capacity of a population to adapt to digital transformation reflected in their human development level may fundamentally shape whether digitalization promotes or obstructs inclusive growth.

To better understand the negative moderation effect, subgroup analysis was conducted by categorizing provinces based on their HDI levels relative to the national average. This analysis reveals contrasting dynamics in how digitalization affects inclusive growth across different human development contexts.

The findings of the subgroup regression indicate that in provinces with a 2023 HDI below the national average, digitalization has a positive and significant effect on inclusive growth. This suggests that digitalization can act as a mechanism to promote social and economic inclusion in regions with relatively underdeveloped human capital. A study by (Ha & Chuah, 2023) in Southeast Asia reveals that regions with higher urbanization experience more immediate benefits from digital transformation compared to rural regions characterized by insufficient infrastructure and limited educational opportunities. This finding aligns with research in Indonesia that reveals inequalities among provinces in the utilization of digital technologies ((Adiningsih et al., 2019); (Ariansyah et al., 2019); (Kartiasih et al., 2023)).

According to (Bai et al., 2021), digital technology facilitates the active participation of MSMEs and informal sector workers in the economy through digital payments, especially mobile money. Moreover, technology leapfrogging allows underdeveloped countries to rapidly improve their technological capabilities by bypassing traditional development stages and directly adopting innovative advancements (Tan et al., 2018). This concept is particularly relevant to provinces with a HDI below the national average, where digitalization fosters inclusive growth by opening new pathways for economic participation.

(Matli & Wamba, 2023) highlight that the lack of traditional infrastructure renders digital technology a more effective for remote work means of promoting economic activities. For instance, instead of investing in physical bank branches, digital financial services can swiftly connect with communities that were previously unbanked, thus improving access to financial services and fostering economic participation.

Additionally, digitalization has transformed workforce productivity by facilitating greater access to online training and educational materials, allowing employees to augment their skill sets ((Dao et al., 2025); (Gegenfurtner et al., 2020)). Moreover, community-level digitalization, especially in rural areas, has the potential to strengthen the resilience and independence of local communities. According to (Manoby et al., 2021), initiatives aimed at digitalizing villages play a crucial role in promoting inclusive development by improving human capital within the grassroots framework.

(Wargadinata & Tendean, 2024) observed that the process of digital transformation is still facing challenges associated with regional disparities, especially in terms of digital literacy and the readiness of infrastructure. Numerous communities in undeveloped regions continue to depend on traditional services due to their limited capacity or access to utilize digital technology effectively. This highlights the importance of a digital strategy that is based on community readiness to ensure that technology implementation is inclusive.

Conversely, in provinces with HDI exceeding the national average, digitalization has a negative and significant effect on inclusive growth. This indicates that in high-HDI regions, digitalization may reinforce existing inequalities rather than promote inclusivity. Specifically, this effect is reflected in widening income disparities, reduced employment opportunities for low-skilled workers, and limited participation of vulnerable groups in digital economic activities.

This can be explained by the nature of digitalization in more developed regions, where technological adoption tends to favor highly skilled individuals, leading to a digital divide. At the same time, automation and efficiency gains may reduce demand for routine labor, thereby limiting job opportunities for lower-income populations. In such contexts, digitalization no longer expands inclusion but instead concentrates economic benefits among already advantaged groups. This mechanism is consistent with the empirical finding of this study, where the negative interaction term between ICT-DI and HDI indicates that higher human development levels are associated with a weaker inclusivity effect of digitalization.

In public policy discussions, digitalization is perceived as a means to improve inclusion in service delivery. Digital governance promotes increased citizen engagement through e-participation and e-services, hence improving accountability and transparency (United Nations Department of Economic and Social Affairs, 2024). However, these benefits are only experienced in regions with sufficient digital literacy and internet connectivity, which explains why high HDI provinces may not automatically translate digital advancement into inclusive outcomes, as the benefits tend to concentrate among already advantaged groups, potentially widening rather than narrowing socioeconomic gaps.