Digital Transformation and Blockchain Ecosystem Readiness as Drivers of ESG Performance: Evidence from European Union Countries

The literature review is organized into four thematic subsections, followed by a bibliometric analysis, linking the theoretical, empirical, and methodological evidence to the research questions and empirical design of the study.

2.1 Digital Transformation and environmental, social, and governance performance: Direct effects and transmission mechanisms

Recent literature generally supports the view that digital transformation is associated with stronger ESG performance through lower information costs, greater transparency, green innovation, and more efficient resource use. However, the convergence of positive findings should not be taken as evidence of a universal relationship. Many studies rely on samples of Chinese listed companies, text-based measures of digitalization, and ESG ratings from agencies that use different methodologies. The apparent consistency of the results may therefore partly reflect similarities in samples, study periods, and measurement approaches rather than the relationship between digital transformation and ESG alone [1-7].

Studies examining mediating mechanisms indicate that technology does not generate sustainable outcomes when introduced independently of organizational capabilities. Green learning, technological competencies, human capital, internal control, innovation, and supply chain coordination influence whether digital investments are associated with improvements in ESG performance. Evidence from systems engineering research similarly indicates that sustainability objectives need to be incorporated into organizational requirements and product development processes rather than treated as an outcome of technological change alone [8]. This shifts attention from the scale of technology investment to the way digital capabilities are integrated into decision-making and governance processes. The evidence suggests that stronger outcomes are more likely when digitalization is accompanied by absorptive capacity, consistent regulatory pressure, and effective institutional structures [9-17].

The literature, however, does not support an unconditional positive relationship. Several studies report nonlinear patterns, differences across the E, S, and G dimensions, and adverse environmental consequences associated with the expansion of digital activity. These findings question the assumption that gains in digital efficiency necessarily translate into greater sustainability, particularly where digital infrastructure depends on energy-intensive resources or where institutional capacity is limited. Meta-analytic evidence likewise reports a positive average relationship, but with substantial heterogeneity across technologies, sectors, and regulatory contexts [18-23].

At the macro level, digital transformation is connected to ESG performance through innovation systems, digital finance, and public governance. The strength of this relationship, nevertheless, varies with the levels of economic development, financial constraints, and institutional quality. The central question is therefore not simply whether digital technologies can support ESG performance, but under what conditions, for which economies, and at what costs. Addressing this question requires cross-country analysis that distinguishes digital capacity from technology use and examines whether the relationship remains robust after structural and macroeconomic conditions are taken into account [24-29].

2.2 Blockchain ecosystem readiness, transparency and environmental, social, and governance

The literature presents blockchain as an infrastructure that can strengthen the traceability, verification, and reliability of sustainability-related data. From this perspective, blockchain may support ESG reporting, green finance, and supply chain monitoring. Much of the existing research, however, is based on sector-specific cases, systematic reviews, or conceptual frameworks, while comparative country-level evidence remains limited. This makes it important to distinguish the technological potential of blockchain from empirically demonstrated outcomes [30-33].

A related strand of the literature suggests that blockchain readiness depends on more than just digital infrastructure. Data quality, interoperability, ICT expertise, organizational capacity, funding, and regulatory clarity also shape the conditions under which blockchain applications can be developed and used. Studies of the public sector and sustainability management indicate that adoption is more likely when technological, institutional, and collaborative conditions are jointly present. Yet, the relative importance of these factors remains unclear, and initial adoption is sometimes treated as equivalent to broader ecosystem readiness [34-37].

In the ESG context, blockchain may help reduce information asymmetry, strengthen carbon accountability, and improve verification processes. An immutable ledger, however, does not ensure that the information entered into the system is accurate, nor does it guarantee that relevant actors will disclose sensitive data. Its contribution therefore depends on reporting standards, audit quality, and coordination across the value chain. The relationship between blockchain and ESG performance is consequently conditional rather than automatic [38-42].

A particularly important unresolved issue is the assumption that greater technical transparency necessarily leads to stronger governance. Blockchain systems may reproduce existing power asymmetries, exclude actors with limited digital capabilities, and introduce additional energy and organizational costs. BERI should therefore be interpreted as a measure of enabling conditions rather than evidence of sustainable outcomes. This distinction supports empirical testing across countries, examining technological readiness alongside institutional quality and ESG performance [43, 44].

2.3 Digital-green transition, sustainable development and community context

The dual digital and green transition is widely discussed in the literature as a means of aligning technological modernization with sustainability objectives. European evidence indicates that digitalization can improve energy efficiency, productivity, and innovation, although these benefits are unevenly distributed. Economies with more developed infrastructure, stronger human capital, and effective institutions tend to benefit more, whereas countries facing persistent digital divides may fall further behind. This uneven pattern challenges the assumption that digital and green transitions necessarily progress in parallel [45-49].

At the enterprise and sector levels, the evidence is more mixed. Digitalization can help small and medium-sized enterprises reduce costs, use resources more efficiently, and develop more sustainable business models. These gains, however, depend on organizational capacity, access to finance, and the ability to absorb new technologies. When digital transformation is treated mainly as a technical replacement without corresponding changes in governance, it may increase platform dependence, energy consumption, and the exclusion of smaller actors. Such differences help explain why empirical findings vary across enterprise size, sector, and level of digital maturity [50-54]. Sector-specific evidence also shows that the sustainability implications of technological development cannot be assessed through efficiency or cost indicators alone. In multimodal transport, for example, route evaluation changes when carbon emissions are considered together with transport time and economic cost [55].

The social and institutional dimensions remain less firmly established than the economic effects. Some studies associate digital transformation with greater social inclusion, improved access to services, and stronger public-sector capacity. Others show that territorial inequalities, skill shortages, and weak governance may concentrate the benefits among already advantaged groups. Digital public administration can improve transparency and service delivery, but its effectiveness depends on citizen trust, institutional design, and communities’ practical ability to access and use new systems [56-61].

The digital transition should thus be evaluated not only through national performance indicators but also in terms of its implications for local development, social inclusion, and environmental governance. The community development literature indicates that technological capacities generate greater value when they operate alongside responsive institutions, social participation, and sustainable resource management. An important unresolved question is whether national digital policies translate into equitable outcomes across territories. Addressing this issue requires research that connects macro-level indicators with communities’ access to technology, public services, and economic opportunities [62-64]. This study approaches this gap from a macro-institutional perspective. Although the country, rather than the local community, is the empirical unit of analysis, national digital capacity, technological infrastructure, and governance quality constitute the environment in which communities access digital services, economic opportunities, and sustainability initiatives. The country-level analysis is therefore interpreted as identifying structural conditions relevant to community development rather than as directly measuring community-level outcomes.

2.4 Composite measurement, cross-country heterogeneity and macroeconomic fundamentals

Measuring ESG performance, digital transformation, and blockchain readiness requires composite indicators, as each concept comprises multiple dimensions that a single variable cannot adequately capture. European studies show that aggregating indicators facilitates cross-country comparison and the identification of development profiles. However, the resulting scores depend on the choice of criteria, normalization procedure, and weighting scheme. Rankings should therefore be interpreted as relative outcomes of a specific analytical framework rather than as absolute measures of performance [65-67].

The literature also documents substantial heterogeneity across European countries. Digital clusters, convergence clubs, and differences in country rankings indicate that progress does not follow a common trajectory, even within the European Union's shared regulatory framework. Some studies, however, rely on single-year observations or period averages, which makes it difficult to distinguish persistent structural improvement from short-term variation. Examining changes in rankings over time is therefore as important as assessing a country’s position in a given year [68-70].

A further methodological issue concerns the aggregation of ESG dimensions. Some approaches allow full compensation, meaning that strong performance in one dimension can offset weakness in another. This assumption is problematic because improvements in economic or governance conditions do not necessarily compensate for deficiencies in environmental or social performance. Evidence on ESG investments, market value, and well-being suggests that the E, S, and G pillars generate different outcomes and should be treated as interconnected rather than interchangeable dimensions [71-73].

Broader macroeconomic conditions also shape composite index performance. Investment, financial development, inflation, income, and capital formation affect countries’ capacity to finance digital infrastructure and the sustainable transition. At the same time, these factors may influence both ESG performance and the explanatory variables, increasing the risk of omitted-variable bias when they are excluded from the analysis. Combining Entropy–TOPSIS with panel models that include macroeconomic controls therefore provides a broader analytical basis than country rankings considered in isolation [74-76].

2.5 Bibliometric analysis

The bibliometric analysis was conducted on the final corpus of studies selected through the PRISMA approach. Of the 140 scientific papers identified during the search and screening process, 70 were retained for the final analysis, corresponding to an inclusion rate of 50%. The corpus covers the period 2022–2026 and reflects growing scholarly interest in the relationship between digital transformation, blockchain ecosystem readiness, and ESG performance. The number of publications increased from four in 2022 and nine in 2023 to 22 in 2024 and 25 in 2025, with 10 papers published in 2026. Within the analyzed corpus, this pattern indicates a marked increase in research activity after 2023. The number of publications for 2026 should nevertheless be interpreted with caution, as the data for that year cover only the period up to the time of the research.

The corpus comprises 61 journals or scientific sources, 20 publishers, and 205 unique authors. An average of 3.1 authors per document and a collaboration index of 3.26 indicate that the literature is predominantly collaborative. By publisher, Elsevier accounts for nine papers, Taylor & Francis for eight, Acadlore Publishing for seven, and MDPI and Springer Nature for six each.

The thematic distribution shows that 25 papers examine digital transformation and ESG performance, 15 address blockchain readiness and ESG governance, and 18 focus on the digital-green transition and the community context. A further 12 papers deal with composite measurement, cross-country heterogeneity, and macroeconomic factors. This distribution reflects the multidisciplinary nature of the field, combining technological, institutional, financial, and environmental perspectives.

For the network analysis, binary co-occurrence of standardized terms in the topics field was computed with a minimum threshold of 2 occurrences. Before constructing the network, synonymous terms and closely related terminological variants were harmonized to reduce artificial fragmentation across clusters. The final network contains 22 terms and 51 links. Node size represents the frequency with which a term occurs, line thickness indicates the strength of co-occurrence, and the distance between nodes reflects thematic proximity. Figure 1 shows a thematic structure organized around three interconnected communities. The network has a density of 0.221 and an average degree of 4.64 links per node, indicating a moderate degree of thematic integration alongside clearly identifiable research sub-directions.

tu_pian_1.png

The red cluster represents the dominant thematic axis of the literature, in which digital transformation is closely linked to ESG performance, green innovation, carbon emissions, and environmental regulation. The central position and larger size of the terms digital transformation and ESG performance indicate that they function as core nodes linking organizational outcomes with broader sustainability debates. The green cluster centers on blockchain, transparency, governance, and sustainable finance. Its more specialized position suggests that blockchain research is concentrated primarily on information verification, traceability, and accountability, while its direct connection with macroeconomic ESG performance remains less developed.

The blue cluster includes sustainable development, the European Union, the SDGs, digital readiness, the dual transition, and composite indicators. The links across these clusters suggest that technology does not operate independently in producing sustainable outcomes, but is closely connected with institutional capacities, public policies, and measurement frameworks.

The more peripheral position of terms related to digital readiness, emissions, and composite indicators points to a limited body of cross-country research that simultaneously considers digital transformation, blockchain readiness, and ESG performance. This gap provides the basis for constructing and empirically testing the DTI, BERI, and ESG indices across European Union countries. The bibliometric findings should, however, be interpreted within the boundaries of the search strategy, the period covered, and the selected corpus. In addition, incomplete data on cited references limited the possibility of conducting co-citation and bibliographic linkage analyses.

4.1 Summary of Entropy–TOPSIS results

The ESG, DTI and BERI indices were calculated separately for each year for the 27 EU countries over the period 2016–2024. As shown in Table 3 and Figure 2, the average ESG increased from 0.3685 in 2016 to 0.3818 in 2024, peaking at 0.4062 in 2019. The DTI was relatively stable, changing from 0.4959 to 0.5052, while the BERI showed the strongest increase, from 0.1190 to 0.4761. The gap between the maximum and minimum values narrowed for all three indices, suggesting relative convergence, although differences between countries remained significant. Since the Entropy weights were calculated annually, the results reflect changes in performance and the discriminatory power of the indicators. Therefore, the indices are interpreted as annual relative positions within the EU-27.

Table 3. Annual ESG Performance Index results for the EU-27, 2016–2024

Note: Authors' calculations based on Eurostat data and the Entropy–TOPSIS method.

tu_pian_2.png

4.2 ESG Performance Index results

The Entropy–TOPSIS results show moderate fluctuations in ESG performance in the EU-27 countries over the period 2016–2024. As shown in Table 3, the annual average of the index increased from 0.3685 in 2016 to a maximum of 0.4062 in 2019. After this period, a decline was recorded, with the lowest average value of 0.3570 in 2021, followed by a gradual recovery to 0.3818 in 2024. Compared to the initial year, the average in 2024 was higher by 0.0133 index points, or about 3.61%.

The Netherlands topped the ranking during 2016–2020, Denmark in 2021 and 2024, and Sweden in 2022–2023. The change in the leading countries shows that the relative advantage in ESG performance was not constant. At the bottom of the ranking, Romania was the worst-performing country in seven out of nine years, while Cyprus and Hungary were ranked last in 2018 and 2022, respectively.

The gap between the maximum and minimum score narrowed from 0.3970 in 2016 to 0.3515 in 2024, suggesting relative convergence, but not elimination of differences between countries. The average Entropy weights show the dominance of the environmental dimension with 45.74%, followed by governance with 29.01% and the social dimension with 25.25%.

4.3 Digital Transformation Index results

The Entropy–TOPSIS results show that the average level of digital transformation in the EU-27 remained relatively stable over the period 2016–2024 (Table 4). The annual average DTI changed from 0.4959 in 2016 to 0.5052 in 2024, marking an overall increase of around 1.88%. The highest average value was recorded in 2019, at 0.5195, while the lowest level was in 2022, at 0.4705. The recovery over 2023–2024 shows that digital transformation did not follow a completely linear trajectory (Figure 3).

Finland ranked first in eight of the nine years, while Sweden held the top position only in 2019. This pattern is consistent with Finland’s relative strength in ICT human capital, digital connectivity, and the use of online services. Romania remained at the bottom of the ranking throughout the period, although its score increased from 0.0352 to 0.0841. The difference between the highest and lowest scores declined from 0.8691 in 2016 to 0.7821 in 2024, indicating some narrowing of the digital gap, although substantial cross-country differences persisted. These results should be interpreted with caution, given that 55.97% of the country-year observations contained at least one imputed value.

Table 4. Annual Digital Transformation Index (DTI) results for the EU-27, 2016–2024

Note: Authors' calculations based on Eurostat data and the Entropy–TOPSIS method.

tu_pian_3.png

4.4 Blockchain Ecosystem Readiness Index results

The Entropy–TOPSIS results show that the BERI exhibits the strongest temporal variation among the three indices constructed in this study. As reported in Table 5 and Figure 4, the EU-27 annual average increased from 0.1190 in 2016 to 0.4761 in 2024, indicating a substantial expansion in the conditions supporting blockchain ecosystem development. The increase was gradual through 2021, when the average reached 0.3510, followed by a slight decline to 0.3411 in 2022. The index then rose markedly in 2023 and 2024, reaching its highest levels over the study period.

Slovenia held the top position from 2016 to 2022, while Sweden and Denmark ranked first in 2023 and 2024, respectively. Greece remained at the bottom of the ranking throughout all nine years, although its score increased from 0.0288 to 0.2337, indicating gradual improvement from a very low initial level. The gap between the highest- and lowest-performing countries narrowed from 0.8773 in 2016 to 0.4948 in 2024, indicating some degree of relative convergence. The average Entropy weights show that digital infrastructure accounted for 50.87% of the index, followed by enterprise digital adoption at 34.59% and digital human capital at 14.54%. These weights indicate that broadband coverage, cloud computing use, and enterprise digital activities contributed most to cross-country differences in blockchain ecosystem readiness within the EU.

Table 5. Annual Blockchain Ecosystem Readiness Index (BERI) results for the EU-27, 2016–2024

Note: Authors' calculations based on Eurostat data and the Entropy–TOPSIS method.

tu_pian_4.png

4.5 Comparing the three indices and testing H1–H2

The comparative results show different development patterns for the three indices. ESG and DTI showed limited increases between 2016 and 2024, by 0.0133 and 0.0093 points, respectively, while BERI increased by 0.3571 points. At the same time, the gap between the maximum and minimum scores narrowed for all three indices, suggesting relative convergence, but not elimination of heterogeneity across countries. The Kruskal–Wallis test revealed statistically significant differences across countries for ESG, DTI, and BERI (p < 0.001). These results, together with the significant differences in scores and the persistence of countries at the edges of the rankings, support H1.

The correlations of the rankings between 2016 and 2024 show relatively high stability for ESG and DTI, but greater mobility for BERI (Table 6). The average absolute change was 3.33 positions for ESG, 2.89 for DTI, and 6.52 for BERI. Ireland recorded the largest improvements in ESG, Poland in DTI, and Croatia in BERI, while the largest declines were recorded for France and Bulgaria, respectively. Therefore, H2 is supported, although the changes in rankings are more moderate for ESG and DTI and significantly more pronounced for BERI.

Table 6. Comparative TOPSIS results and cross-country rank dynamics

Notes: *** p < 0.001; *p < 0.05. Positive change indicates improved ranking. Authors' calculations based on annual Entropy–TOPSIS results for EU-27, 2016–2024.

4.6 Descriptive statistics and correlations

Table 7 presents the descriptive statistics and Pearson correlation matrix for the panel of 243 country-year observations. The ESG Performance Index has a mean of 0.3812, while the DTI and BERI have means of 0.4986 and 0.2915, respectively. The relatively higher standard deviation of the DTI (0.1984) indicates considerable heterogeneity in the level of digital transformation across European Union countries, while the ESG and BERI show more moderate variation. The correlation analysis reveals a positive relationship between ESG and DTI (r = 0.531), suggesting that countries with more advanced digital transformation tend to have better ESG performance.

Table 7. Panel A: Descriptive statistics and Panel B: Pearson correlation matrix

The correlation between ESG and BERI is weaker (r = 0.214), suggesting that blockchain ecosystem readiness captures a distinct dimension of institutional and technological development. Among the control variables, ESG is positively correlated with GDPpc (r = 0.472) and gross fixed investment (r = 0.266), while it is negatively correlated with INF (r = –0.174) and UNEMP (r = –0.247). The correlation between DTI and BERI is also relatively low (r = 0.287), indicating that the two indices capture complementary rather than identical dimensions of technological development. The strongest correlation among the explanatory variables is observed between DTI and GDPpc (r = 0.596). However, the VIF analysis does not indicate multicollinearity, as the maximum VIF is 2.91, well below the 5 threshold. All variables are therefore included simultaneously in the econometric models without compromising the stability of the estimates.

4.7 Diagnostic tests and model selection

The results of the diagnostic tests, presented in Table 8, confirm that the panel structure should be taken into account in the econometric estimation. The F-test for country fixed effects and the joint test for year effects reject the respective null hypotheses (p < 0.001), indicating systematic differences between countries and time periods that cannot be neglected in the modeling.

Table 8. Panel diagnostic tests

Note: Authors' calculations based on the EU-27 panel, 2016–2024.

Similarly, the Breusch–Pagan Lagrange Multiplier test rejects the appropriateness of the pooled OLS model (p < 0.001), confirming the presence of unobserved heterogeneity in the panel. The Hausman test rejects the null hypothesis (p = 0.012), indicating that the fixed-effects model is more appropriate than the random-effects model. For this reason, all econometric models are estimated using Two-Way Fixed Effects, which simultaneously controls for country-invariant heterogeneity and time-joint effects.

The results of further diagnostic tests indicate the presence of heteroscedasticity (Modified Wald), serial autocorrelation (Wooldridge) and cross-sectional dependence (Pesaran CD), typical characteristics of macroeconomic panels with European Union countries. To ensure reliable statistical inference, the coefficients are reported with Driscoll–Kraay standard errors, which are robust to heteroscedasticity, autocorrelation and cross-sectional dependence. The multicollinearity analysis does not reveal any serious problems. The average VIF value is 1.82, while the maximum value is 2.91, significantly below the recommended threshold of 5. This confirms that the explanatory variables can be included simultaneously in the model without compromising the stability of the estimates.

4.8 Econometric results and testing of H3–H5 

Table 9 presents the results of three Two-Way Fixed Effects models, estimated with fixed effects of countries and years. In line with the diagnostic tests presented in the previous section, statistical inference relies on Driscoll–Kraay standard errors, which are robust to heteroscedasticity, serial autocorrelation, and cross-sectional dependence. All models include the same set of macroeconomic control variables to isolate the impact of digital transformation and blockchain ecosystem readiness on ESG performance.

Table 9. Two-Way Fixed Effects estimates of ESG performance (Driscoll–Kraay standard errors)

The results of Model 1 show that the DTI is positively and statistically significantly associated with ESG performance (β = 0.1187, DK SE = 0.0439, p = 0.0069, 95% CI [0.032, 0.205]). This relationship remains significant after controlling for macroeconomic conditions and country- and year-specific effects, indicating that higher levels of digital transformation are associated with stronger ESG performance.

In Model 2, the BERI is also positively and statistically significantly associated with ESG performance (β = 0.0714, DK SE = 0.0305, p = 0.0192, 95% CI [0.011, 0.131]). In Model 3, both DTI (β = 0.1016, DK SE = 0.0404, p = 0.0119, 95% CI [0.022, 0.181]) and BERI (β = 0.0588, DK SE = 0.0282, p = 0.0371, 95% CI [0.004, 0.114]) remain positive and statistically significant. The moderate decline in coefficient magnitudes relative to the separate specifications indicates some shared variation between the two indices, while each retains distinct explanatory information for ESG performance.

Among the control variables, GDPpc, GFCF, and LP are positively and statistically significantly associated with ESG performance across the model specifications. FDI remains statistically insignificant in all three specifications (p > 0.10), indicating that no systematic within-country association between FDI and ESG performance is identified after accounting for the other covariates and country and year fixed effects.

Model 3 provides the strongest overall empirical fit, with the highest Within R² (0.223) and Adjusted Within R² (0.191), together with the lowest AIC (−964.37) and BIC (−812.16). The joint Wald test (χ² = 18.47, p < 0.001) also shows that DTI and BERI are jointly statistically significant. Taken together, these results provide empirical support for H3, H4, and H5.

4.9 Robustness checks

To assess the robustness of the baseline estimates, the combined model was subjected to five sensitivity checks: (i) DTI and BERI were lagged by one year to examine temporal robustness; (ii) 2020–2021 were excluded to assess sensitivity to the COVID-19 shock; (iii) macroeconomic controls were winsorized at the 1st and 99th percentiles to limit the influence of extreme observations; (iv) BERI was reconstructed using an alternative specification to assess sensitivity to index operationalization; and (v) a leave-one-country-out analysis sequentially excluded each EU member state to examine whether the results were driven by any single country. All regression-based specifications retain the macroeconomic controls, country and year fixed effects, and Driscoll–Kraay standard errors. The baseline Model 3 estimates (DTI: β = 0.1016, p = 0.014; BERI: β = 0.0588, p = 0.039) serve as the benchmark for comparison.

The one-year-lag specification preserves both the positive sign and statistical significance of DTI (β = 0.1094, p = 0.018) and BERI (β = 0.0617, p = 0.047), with coefficient magnitudes remaining close to the baseline estimates. This indicates temporal stability in the observed associations. The remaining sensitivity checks yield similarly stable results. Excluding 2020–2021 leaves both DTI (β = 0.0968, p = 0.020) and BERI (β = 0.0561, p = 0.044) positive and significant, indicating that the baseline findings are not driven by the pandemic period. Winsorizing the macroeconomic controls at the 1st and 99th percentiles yields coefficients that remain close to the baseline estimates (DTI: β = 0.0995, p = 0.012; BERI: β = 0.0574, p = 0.035), indicating that extreme observations do not materially influence the results. The alternative BERI specification produces a similar pattern, with both DTI (β = 0.1038, p = 0.011) and BERI (β = 0.0546, p = 0.042) retaining positive and statistically significant associations with ESG performance. This suggests that the main findings are not dependent on the baseline operationalization of blockchain ecosystem readiness.

The leave-one-country-out analysis further confirms coefficient stability. Sequentially excluding each of the 27 EU countries yields DTI coefficients ranging from 0.0872 to 0.1169 and BERI coefficients from 0.0478 to 0.0685, with both coefficients remaining positive in all 27 re-estimations. This indicates that no single country drives the baseline associations, and the coefficients remain stable in both sign and magnitude across the sample. Overall, the robustness checks demonstrate substantial stability in the sign and magnitude of the DTI and BERI coefficients. Both variables remain positive and statistically significant across the alternative regression specifications reported in Table 10, while the leave-one-country-out analysis confirms positive coefficient stability across all 27 re-estimations. These results indicate that the main findings are not materially driven by temporal specification, the pandemic period, extreme macroeconomic observations, BERI operationalization, or any single EU country, providing additional robustness support for H3–H5.

Table 10. Robustness and sensitivity tests

4.10 Summary of hypothesis testing

Table 11 summarizes the results of testing the five study hypotheses. The Entropy–TOPSIS analysis identified significant differences in ESG performance among European Union countries over the period 2016–2024, supporting Hypothesis H1. The analysis of ranking dynamics also revealed significant year-to-year changes, indicating that ESG performance is not static but varies over time alongside economic, institutional, and technological developments. These findings support Hypothesis H2.

Table 11. Summary of hypothesis testing

The results of the Two-Way Fixed Effects models, estimated with Driscoll–Kraay standard errors, show that the DTI has a positive and statistically significant impact on ESG performance across all key specifications and robustness analyses. Similarly, the BERI presents a positive and statistically significant effect, confirming that the development of blockchain capacities contributes to improving ESG performance.

The combined model, which includes both DTI and BERI, presents the highest explanatory power, the most favorable values of the information criteria (AIC and BIC), and statistical significance of the joint Wald test. Robustness checks confirm the stability of the coefficients across all alternative specifications, proving that the results do not depend on the analysis period, extreme values, or alternative operationalization of the BERI index. Overall, the empirical evidence fully supports the conceptual model of the study and confirms that digital transformation and blockchain ecosystem readiness constitute two complementary mechanisms for improving ESG performance in European Union countries. Consequently, all five hypotheses of the study are accepted.

This study examined the relationship between digital transformation, blockchain ecosystem readiness, and ESG performance in 27 European Union countries over the period 2016–2024, integrating an approach based on composite indices with panel econometric analyses. The combination of the Entropy–TOPSIS method with Two-Way Fixed Effects models enabled not only the objective measurement of multidimensional performance, but also the identification of factors related to its changes over time and across countries. In this way, the study provides an analytical framework for assessing how national technological and institutional capacities interact with sustainability performance and, indirectly, shape the enabling conditions for inclusive and sustainable community development.

The main scientific contribution of this study lies in integrating transparent multidimensional measurement with longitudinal panel econometrics and in distinguishing digital transformation from blockchain ecosystem readiness as related but analytically separate national capacities. By operationalizing these dimensions through composite indices and testing their independent and joint associations with ESG performance, the study provides a comparative framework for assessing technological readiness and sustainability across the EU-27. Furthermore, the study demonstrates that using composite indices grounded in objective data can provide a more comprehensive assessment of complex phenomena than analyzing individual indicators. From a practical perspective, the results highlight the importance of aligning digital transformation policies with national sustainable development strategies. Public institutions should continue investing in digital infrastructure, expanding e-services, strengthening digital skills, and developing regulatory frameworks that support innovation and the responsible use of blockchain technologies. Companies, in turn, should treat digitalization and blockchain-related capabilities as part of their longer-term sustainability strategies, particularly in ESG reporting, risk management, transparency, and decision-making. Territorial disparities in digital access and technological capacity also require greater policy attention. National digital and sustainability strategies should therefore include mechanisms that enable local communities, especially those in less digitally advanced regions, to access e-services, develop digital skills, participate in technology-enabled economic activities, and benefit from more transparent and accountable governance systems.

This study has several limitations. The analysis is confined to EU countries over the period 2016–2024, and the availability of comparable Eurostat indicators constrains the operationalization of blockchain ecosystem readiness. Composite indices cannot capture every dimension of digital and institutional development, while country-level data do not directly reflect community-level outcomes or territorial disparities. In addition, the static panel framework identifies conditional associations rather than causal effects. Future research could extend the analysis to longer periods and non-EU economies, apply dynamic panel models or stronger causal identification strategies, and use regional or NUTS-level data to examine how digital and blockchain readiness relates to ESG performance and sustainable community development.