Impacts of Digital Entrepreneurial Ecosystems on Sustainable Development: Insights from Latin America

Abstract

Digital Entrepreneurial Ecosystems (DEEs) are transforming the economic landscape through their integration of digital technologies, offering new opportunities for innovation and growth. This study explores the impact of DEEs on sustainable development, focusing specifically on Latin America. As DEEs continue to evolve, understanding their influence on economic, environmental, and social sustainability becomes crucial, particularly in a region characterized by significant developmental challenges. Utilizing a data panel from two different periods of analysis, from 2013 to 2017 and from 2018 to 2022, within the adapted DEE framework provided by the Global Entrepreneurship Development Institute (GEDI), we employ Principal Component Regression (PCR), Partial Least Squares Regression (PLSR), and fuzzy-set Qualitative Comparative Analysis (fsQCA 3.0) to analyze DEE components across 14 Latin American countries. These countries may not have the full spectrum of digital capabilities, yet they are still able to harness the digital elements they do possess effectively. This suggests that even partial digitalization, when strategically utilized, can lead to substantial gains in sustainable development. Additionally, Networking, Digital Protection, and Digital Tech Transfer are DEE components that present a higher magnitude in social, environmental, and economic development in Latin American countries. This study not only contributes to a deeper understanding of a DEE’s role in fostering sustainable development, but it also offers actionable insights for policymakers and entrepreneurs to leverage DEEs for broader societal benefits. The implications of the findings present perspectives under the existing literature, and the conclusion shows recommendations for future research and strategy development.

1. Introduction

Digital Entrepreneurial Ecosystems (DEEs) are becoming increasingly important in the modern economic landscape, driven by the proliferation of digital technologies that transform traditional business models and entrepreneurial activities [1,2]. Understanding DEEs is crucial as the fluidity of entrepreneurial processes and results driven by digitalization means that different DEEs may yield varied outcomes [3]. Some authors [4,5] offer a broader perspective, seeing DEEs as regional combinations of social, political, economic, and cultural elements supporting innovative startups through digital technologies. This broader definition still blurs the distinction between DEEs and traditional entrepreneurial ecosystems by focusing primarily on opportunity generation. Elia et al. (2020) further refine the concept, describing DEEs as self-organizing communities leveraging digital services and tools throughout the entrepreneurial process [6]. Nonetheless, DEEs are not always self-organizing and sometimes require specific governance mechanisms. In our understanding, DEEs stand for complex, dynamic systems of diverse actors utilizing digital technologies for value co-creation, supported by a digital infrastructure that enables resource access, governance, and territorial boundary spanning [1,7]. This definition highlights the central role of digital technologies and dynamic interactions, emphasizing the unique aspects of DEEs compared to traditional entrepreneurial ecosystems [1,4,6,8]. However, different typology proposes multiple core value propositions as a basis for assessing DEEs. Given the variation in governance autonomy and collaboration levels within DEEs, management practices and strategies must be tailored accordingly [1]. Despite the growing interest in DEEs, there is still a need to understand their impact on sustainable development growth comprehensively.

Particularly in the Latin America region, achieving the sustainable development goals (SDGs) requires tackling issues and digital challenges to deal with economic inequality, environmental degradation, and social exclusion. For instance, SDG Goal 1 (No Poverty) and Goal 10 (Reduced Inequality) are particularly relevant given the high levels of income disparity and marginalized communities in many Latin American countries [9]. Environmental sustainability is also a critical concern, as seen in Goal 13 (Climate Action) and Goal 15 (Life on Land), with the Amazon rainforest and other vital ecosystems under threat from deforestation and climate change [10]. Furthermore, Goal 8 (Decent Work and Economic Growth) emphasizes the need for inclusive economic development, which is crucial for a region characterized by informal labor markets and unemployment [11].

In this context, a DEE is potentially pivotal in advancing the SDGs by fostering innovation and economic growth in a digitally connected world. These digital ecosystems, which encompass a network of startups, investors, educational institutions, and support services, are instrumental in addressing SDGs such as Goal 8 (Decent Work and Economic Growth) and Goal 9 (Industry, Innovation, and Infrastructure) by driving technological advancements and creating new business opportunities [12]. Moreover, the integration of digital technologies within DEEs supports Goal 7 (Affordable and Clean Energy) and Goal 13 (Climate Action) by promoting the development and adoption of sustainable energy solutions and climate-friendly innovations [13]. By leveraging digital tools and platforms, entrepreneurs are empowered to scale solutions that tackle pressing global challenges, ultimately contributing to achieving a more inclusive and sustainable global economy [14].

Therefore, this study aims to address critical gaps in the literature by posing the research question: ‘To what extent do DEEs component indicators impact sustainable economic, environmental, and social development in Latin American countries’? We will explore this question by examining the diverse levels at which DEEs influence sustainable development outcomes. Our analysis involves 14 countries in Latin America, utilizing research methods such as Principal Component Regression (PCR), Partial Least Squares Regression (PLSR), and fuzzy-set Qualitative Comparative Analysis (fsQCA). The choice of these 14 Latin American countries is due to limitations in data availability for others. Countries such as Antigua and Barbuda, Belize, Haiti, Trinidad, and Tobago, among others, are out of selection because their data on key DEE metrics are outdated or unavailable. Including these nations would have introduced bias and compromised the analysis. The selected 14 countries represent a balance between geographic diversity and reliable data, allowing for a robust examination of a DEE’s impact on sustainable development. We used a data panel from the 2013 to 2022 period with 39 DEE component indicators adapted within the DEE framework presented by the Global Entrepreneurship Development Institute (GEDI) [15]. This methodological approach is particularly relevant for understanding the unique challenges and opportunities faced by Latin American countries in harnessing DEEs for sustainable development.

The findings reveal that DEEs have significantly evolved in advancing sustainable development by fostering innovative digital solutions. This evolution is particularly prominent in the realms of social and economic sustainable development, where DEEs have been instrumental in driving growth and reducing inequalities. Moreover, this study highlights that networking and digital technology transfer are key DEE components that exhibit the most significant impact across sustainable development outcomes (social, environmental, and economic) in Latin American countries. These metrics facilitate the flow of knowledge, resources, and technological advancements, thereby enhancing the capabilities of entrepreneurs to address pressing regional challenges. By leveraging these strengths, DEEs not only contribute to achieving the SDGs but also promote a more inclusive and resilient economic landscape across the region.

The contribution of this study lies in its comprehensive analysis of the sustainability implications of DEEs. By the end of this paper, readers will gain a deeper understanding of how DEEs have contributed to sustainable development across sustainable economic, environmental, and social development. This study reveals the mechanisms through which DEEs drive sustainable development in emergent economies and offers insights into how policymakers and entrepreneurs can harness DEEs to achieve sustainable growth. Such knowledge is critical for designing effective strategies that leverage digital technologies for broader societal benefits, especially in the context of Latin America.

The remainder of this paper is as follows: Section 2 outlines the research methods, detailing the approaches and techniques used to investigate the research question. Section 3 presents the results, highlighting key findings on the impact of DEEs on sustainable development. Section 4 discusses the implications of these findings, integrating them with the existing literature and theories. Finally, Section 5 concludes the paper, summarizing the main contributions and suggesting directions for future research.

2. Digital Entrepreneurship Ecosystems: Perspectives on Sustainable Development in Latin America

The potential of DEEs to affect sustainable development in Latin America stems from the ability to address deep-rooted structural challenges specific to this region, particularly in relation to social inclusion, innovation diffusion, and institutional resilience [16,17,18]. Latin America faces unique socioeconomic disparities, characterized by large informal economies, unequal access to education, and a historical lag in technological adoption [19]. DEEs, by leveraging digital platforms, can offer scalable, innovative solutions to these systemic problems. One critical perspective is that DEEs enable a decentralized approach to economic development. By utilizing digital tools, entrepreneurs can bypass traditional gatekeepers of capital and infrastructure, accessing markets and resources that were previously out of reach [20]. This decentralization fosters greater economic diversification, allowing smaller communities and marginalized groups to participate in entrepreneurial activities that drive local economic growth. This mechanism is particularly important in rural and underserved urban areas, where traditional entrepreneurial ecosystems may struggle to gain traction due to infrastructure deficits or limited institutional support [21,22].

Moreover, DEEs have the potential to foster institutional innovation in a region where governance weaknesses often hinder development progress. By integrating digital technologies with entrepreneurial practices, DEEs can improve transparency, enhance accountability, and reduce corruption [23,24,25]. And these factors are critical for sustainable development. In the context of Latin America, digitalization can support stronger governance frameworks, particularly in sectors such as energy, agriculture, and health, where digital solutions can introduce efficiencies and increase public trust in institutions [26,27,28].

Another crucial perspective is the role of DEEs in facilitating knowledge transfer and collaboration across borders [29,30,31,32]. Digital platforms enable Latin American entrepreneurs to connect with global networks of expertise, resources, and technologies, allowing for the importation and adaptation of sustainable practices from more developed entrepreneurial ecosystems [33]. On the other hand, international connectivity is essential for addressing regional challenges like climate change, resource depletion, and social inequality [34]. By participating in DEEs, Latin Americans can contribute to and benefit from shared innovations, enhancing their ability to meet local sustainable development goals.

Additionally, a DEE’s focus on fostering a culture of innovation and risk-taking could contribute to reshaping the entrepreneurial mindset in Latin America. Historically, the region has often been risk-averse due to economic instability and uncertainty [9,19,35]. However, the digitalization of entrepreneurial processes lowers barriers to entry, encourages experimentation, and allows for the rapid iteration of business models [36]. This cultural shift towards embracing digital entrepreneurship could accelerate the region’s transition towards more sustainable economic models, as entrepreneurs become more willing to invest in long-term, environmentally friendly ventures.

3. Methods

3.1. Sample

Our sample consists of data from 14 Latin American countries spanning the years 2013 to 2022. Regarding the countries typically recognized by World Bank Group as part of Latin America, spanning Central America, South America, and the Caribbean [37], the exclusion of certain nations in this study is primarily due to limitations in the availability, completeness, and timeliness of the required data. Specifically, countries like Antigua and Barbuda; Belize; Dominica; Dominican Republic; El Salvador; Grenada; Guyana; Haiti; Honduras; Jamaica; Nicaragua; St. Kitts and Nevis; St. Lucia; Suriname; and Trinidad and Tobago were out of selection due to either outdated or entirely unavailable data on the key metrics necessary to measure DEEs. Including these countries without reliable and up-to-date data would have introduced significant bias into the analysis, potentially undermining the integrity of the findings. Moreover, for the period under review (2013–2022), several datasets relevant to these nations are either incomplete or published irregularly, which poses challenges in maintaining a consistent temporal framework for comparative analysis across all selected countries. The choice of the 14 countries reflects a balance between geographic diversity and data reliability, allowing for a robust examination of DEE impacts on sustainable development in the region. The unbalanced panel dataset draws from various historical sources related to sustainable development goals (SDGs) and DEE indicators.

Table 1. Countries’ indexes in SDG and DPE scores.

Countries	SDG		Spillover		DPE
	Rank	Score	Rank	Score	Score
Argentina	47	74.40	13	96.83	30.36
Bolivia	90	68.08	15	96.76	34.00
Brazil	52	73.78	22	96.05	31.24
Chile	32	77.82	-	-	40.60
Colombia	74	70.30	-	-	28.01
Costa Rica	59	72.88	-	-	-
Ecuador	75	70.14	43	94.94	21.31
Guatemala	128	59.41	102	87.33	14.99
Mexico	80	69.28	-	-	29.43
Panama	84	69.09	126	72.98	27.97
Paraguay	91	68.02	51	94.57	15.60
Peru	64	71.88	49	94.68	23.63
Uruguay	34	77.09	68	92.72	36.34
Venezuela, RB	122	62.46	80	90.66	-

Source. United Nations (UN) and the Global Entrepreneurship and Development Institute (GEDI). Legend: sustainable development goals (SDGs); Digital Platform Economy Index (DPE).

shows the 2022 rankings of these countries in terms of their SDG scores and their 2021 Digital Platform Economy Index (DPE) standings.

Countries were ranked based on their overall SDG score, reflecting their progress towards achieving all 17 SDGs, with a score of 100 indicating complete achievement. Additionally, the spillover score evaluates the positive or negative impacts of a country’s actions on other nations’ abilities to achieve SDGs. The Spillover Index measures these impacts across three dimensions, environmental and social effects related to trade, economy and finance, and security, with higher scores indicating more positive and fewer negative spillovers. The DPE Index, a composite measure, includes 11 pillars assessing the integration of digital and entrepreneurial ecosystems across countries. The choice of these 14 countries was by the fact that data for other Latin American countries either did not exist or were outdated, making these 14 countries the most viable option for a comprehensive analysis.

3.2. Data and Measures

Our DEE component indicators are based on the model presented by the Global Entrepreneurship and Development Institute (GEDI) and come from various data sources, acknowledging the limitations in measuring DEEs, which include national heterogeneity in available information due to differences in cyber structure and innovations [38]. The transition from a managed to a platform economy in the 21st century is encapsulated in Niall Ferguson’s (2019) work, highlighting the global reach of platform-based ecosystems facilitated by digital technology, contrasting the regional focus of entrepreneurial ecosystems [8,39,40]. Sussan and Acs (2017) introduced the DEE to address gaps in entrepreneurship conceptualization, integrating users, agents, digital technology, and institutions, which Song (2019) later refined to include multi-sided platforms [8,41]. Recently, Szerb et al. (2022) presented a comprehensive analysis of a DEE framework including four main pillars: Digital User Citizenship, Digital Technology Entrepreneurship, Digital Multi-sided Platforms, and Digital Technology Infrastructure. This DEE framework aims to address gaps in understanding the role of digitalization in modern economies, highlighting different digital components as presented in

Table 2. DEE component descriptions.

DEE Pillars	Components	Descriptions
Digital Technology Infrastructure	Digital Access	It refers to the availability of digital infrastructure such as computers and the Internet for citizens.
	Digital Freedom	It reflects the extent to which a government and its institutions allow the development of digital infrastructure.
	Digital Protection	It encompasses the extent to which laws and regulations safeguard users from piracy and cybercrime.
Digital Multi-sided Platform	Digital Finance	It refers to different facets of finance that depend on digital technologies to facilitate online financial transactions and services.
	Networking	It is an externality where the value of a product or service increases with the number of users.
Digital Technology Entrepreneurship	Digital Adoption	It reflects the fundamental ability of entrepreneurial agents to utilize digital technologies.
	Digital Tech Absorption	It assesses how well entrepreneurial agents are able to integrate and utilize existing digital technologies.
	Digital Tech Transfer	It highlights the capacity to disseminate digital technologies.
Digital UserCitizenship	Digital Literacy	It refers to the ability of entrepreneurs to use computers, digital infrastructure, and digital platforms effectively.
	Digital Openness	It refers to how effectively a country’s institutions promote access to and utilization of digital infrastructure.
	Digital Rights	It pertains to the human and legal rights that enable citizens to use digital infrastructure to ensure privacy protection.

[15]. Digital User Citizenship addresses explicit legitimation and implicit social norms that enable user participation in a digital society. Digital Technology Infrastructure focuses on the coordination and governance needed for establishing institutional standards related to digital technology. Digital Multi-sided Platforms serve as intermediaries for transactions and knowledge exchange, facilitating entrepreneurial innovation and value creation. Digital Technology Entrepreneurship includes third-party agents engaging in experimentation, entrepreneurial innovation, and value creation using digital technologies [8,15,41,42].

In our research, we acknowledge the six-year average time lag present in the DEE framework as presented by the GEDI. The most recent Digital Platform Economy Index, released in 2020, utilizes data from the years 2016–2017. This considerable time lag highlights the need for more current indicators and datasets to accurately reflect the rapidly evolving dynamics of digital ecosystems. Therefore, our study emphasizes the importance of incorporating more recent data sources to ensure the relevance and applicability of the findings in understanding the current state of DEEs in Latin America. We utilized data up to 2022, adapting the indicators from other reliable sources. The DEE framework, as defined in Szerb et al.’s (2022) scheme [15], serves as independent variables, measured by the share of digitalization in each component within a country for a specific year. The dependent variable in the panel is defined according to the categorization of the SDGs in the previous literature [43,44,45,46], which relates to economic, environmental, and societal sustainable development. The control variable in the regression models was the year. We compared each country’s scores for each five-year interval with their income levels. These variables control the interaction between national digital dynamics and SDG trajectories, thereby providing comprehensive coverage of the conditioning factors related to digitalization, entrepreneurship, and sustainability capacity at the macro level.

Table 3. DEE component indicators.

DEE Components	Indicators	Source
Digital Access (DAC)	1. Active mobile-broadband subscriptions/100 pop 2. Fixed broadband Internet subscriptions/100 pop 3. ICT * access 4. Population covered by at least a 3G mobile network (%)	International Telecommunication Union World Bank Group
Digital Adoption (DAD)	5. Computer software spending 6. Creating electronic presentations with software 7. Making calls using VoIP or messaging app	International Telecommunication Union World Bank Group
Digital Finance (DFI)	8. Credit card penetration (% of adults) 9. Debit card penetration (% of adults) 10. Online banking penetration (% of population)	Statista (from Internation Monetary Fund, World Bank, United Nations, and Eurostat)
Digital Freedom (DFR)	11. Language accessibility of top ranked apps 12. Obstacles to access 13. Top-level domains (TLDs) per person 14. Violations of user rights	Freedom House GSMA
Digital Protection (DPR)	15. Corruption perception index 16. Mobile connectivity index—cybersecurity index 17. Software piracy rate	GSMA Statista Transparency International
Digital Literacy (DLI)	18. Basic skills 19. Individuals using Internet 20. Knowledge creation	GSMA International Telecom Union World Bank Group
Digital Openness (DOP)	21. Households with a computer at home (%) 22. Households with Internet access at home (%) 23. ICT * regulatory tracker 24. Limits on content	Freedom House International Telecom Union
Digital Rights (DRI)	25. Freedom of expression and alternative sources of data 26. Online e-participation 27. Property rights 28. Rule of law	Property Rights Alliance V-Dem Core World Bank Group
Digital Tech Absorption (DTA)	29. ICT * services exports, % total trade 30. ICT * services imports, % total trade 31. ICT * use	World Bank Group
Digital Tech Transfer (DTT)	32. Knowledge and technology output index 33. R & D * transfer 34. Share of medium- and high-tech activities in manufacturing export index 35. University–industry R & D * collaboration	Global Entrepreneurship Monitor United Nations World Bank Group
Networking (DNE)	36. Mobile social media penetration 37. Network coverage 38. Network performance 39. Participating in social networks	GSMA International Telecom Union

Note: * Information Communication Technology (ICT); Research and Development (R & D).

presents the structure of variables in the model of analysis [8,15,41,42].

The framework of sustainable development outcomes is derived and aligned with similar frameworks discussed in the literature [43,45,46]. The SDGs address key aspects of the economy, society, and the environment. Consequently, various categorizations have emerged in the literature to classify them into multiple dimensions [43,44,45,46,47]. In this research, we adopt the two-level categorization framework suggested by Wu et al. and Palomares et al. [43,45]. The first level divides the SDGs into three primary dimensions essential for achieving sustainability: economic, social, and environmental. These dimensions are as follows:

• Social dimension: Focuses on sustainable development through equity, welfare, and community well-being. Social dimensions are in two subcategories: (1) social development, encompassing sustainable communities, peace, justice, and global partnerships; and (2) equity in education, employment, and gender, among other factors.
• Environmental dimension: Emphasizes environmental preservation and the sustainable management of key resources. Environmental dimensions are also in two subcategories: (1) resource management, which includes water, clean energy, responsible consumption, and production; and (2) protection of natural ecosystems, addressing climate, land, and marine environments.
• Economic dimension: Targets sustainable economic development, with a focus on individual prosperity and societal well-being. This dimension includes two subcategories, as well: (1) life essentials, such as poverty alleviation, food security, and healthcare; and (2) economic and technological development, which includes growth, sustainable industrialization, and innovation.

This categorization in our analysis is outlined in

Table 4. SDG framework of indicators.

SDG Performance	Description	Indicators
Sustainable Social Development	It considers sustainable development in terms of community welfare, prosperity, and equality	SDG1: No Poverty SDG2: No Hunger SDG3: Good Health and Well-Being SDG4: Quality Education SDG5: Gender Equality SDG10: Reduced Inequalities SDG16: Peace, Justice, and Strong Institutions SDG17: Partnerships for the Goals
Sustainable Environmental Development	It involves safeguarding and preserving the environment, alongside the sustainable management of resources	SDG6: Clean Water and Sanitation SDG7: Affordable and Clean Energy SDG13: Climate Action SDG14: Life Below Water SDG15: Life on Land
Sustainable Economic Development	It encompasses sustainability and individual welfare, focusing on overall prosperity	SDG8: Decent Work and Economic Growth SDG9: Industry, Innovation, and Infrastructure SDG11: Sustainable Cities and Communities SDG12: Responsible Consumption and Production

[43,44,45,46,48].

3.3. Model Specifications

The impact of structural shifts related to DEE components on SDGs explores various analytical approaches. We use panel data analysis with PCR (Principal Component Regression) and PLRS (Partial Least Squares Regression), selected for their unique strengths in addressing real-world challenges [49,50]. In data transformations, we added a constant 2 to the logarithm to deal with missing values across indicators in the log-transformed variables. Additionally, fs/QCA was applied as calibration procedures to examine how the configurational model among DEE components evolved over time in countries at different stages of development on SDGs [51].

3.3.1. Regression Model Specifications

PCR and PLRS are particularly valuable in dealing with explanatory variables and multicollinearity. Both techniques address multicollinearity effectively: PCR creates orthogonal components, while PLRS forms components that maximize covariance with the dependent variable. They are compatible with panel data, but their full potential is when combined to leverage the data structure. Integrating these methods into fixed or random effects models reduces the instability of estimated coefficients, enhances the precision of estimates, explains data variability more effectively, simplifies analysis, and makes models more interpretable and less prone to overfitting [52,53].

The primary benefits of this approach in panel data analysis include (1) dimensionality reduction, crucial for models with variables, with PCR validating the structure of the 11 adapted variables; (2) handling multicollinearity, which poses significant limitations for other methods; and (3) controlling for unobserved heterogeneity. In fixed effects models, this involves controlling for constant unobserved variations over time within each unit (e.g., country), thereby removing bias from omitted time-invariant variables. In random effects models, differences between units are random and uncorrelated with the explanatory variables.

PCR, in particular, combines the strengths of principal component analysis (PCA) with linear regression, making it a powerful tool for high-dimensional data analysis, especially when the number of observations is smaller than the number of predictor variables. PCR works by constructing a few principal components and using them as predictors in a regression model [49]. Mathematically, PCR is as follows:

$$\mathrm{Y}=\mathrm{X}\mathrm{B}+\xi$$

(1)

where Y is the response variable, X is the observed predictor matrix, B is the matrix of regression coefficients, and ξ is the vector of residual errors. The solution for multiple linear regression is by

$$\widehat{\mathrm{B}}={\left({\mathrm{X}}^{\prime }\mathrm{X}\right)}^{-1}{\mathrm{X}}^{\prime }\mathrm{Y}$$

(2)

In PCR, the elimination of multicollinearity in the predictor variables is by extracting orthogonal predictors through PCA on X, and then performing regression on Y using a subset of the resulting components of X. The decomposition of X through PCA is

$$\mathrm{X}={\mathrm{U}\mathrm{S}\mathrm{V}}^{\prime }+\mathsf{\xi }$$

(3)

$$\mathrm{U}=\mathrm{X}\mathrm{V}$$

(4)

where U is the matrix of scores, S is the diagonal matrix of singular values, and V is the matrix of loadings. The multiple linear regression is as follows:

$$\mathrm{Y}=\mathrm{U}\mathrm{B}+\mathsf{\xi }$$

(5)

The solution for regression in this context is

$$\widehat{\mathrm{B}}={\left({\mathrm{U}}^{\prime }\mathrm{U}\right)}^{-1}{\mathrm{U}}^{\prime }\mathrm{Y}$$

(6)

PLSR is an enhanced version of Principal Component Regression (PCR), designed to improve predictive performance by emphasizing predictor variables most strongly related to the response variable. In PLSR, the outer relation for X, similar to PCA, is as follows:

$$\mathrm{X}={\mathrm{U}\mathrm{S}\mathrm{V}}^{\prime }+\mathsf{\xi }$$

(7)

$$U=XV$$

(8)

Similarly, Y is as follows:

$$\mathrm{Y}=\mathrm{T}\mathrm{S}{\mathrm{P}}^{\prime }+{\mathsf{\xi }}^{\mathrm{*}}$$

(9)

$$\mathrm{T}=\mathrm{Y}\mathrm{P}$$

(10)

where T is the matrix of Y scores, S is the diagonal matrix of singular values, and P is the matrix of Y loadings. The description of the inner relation between X and Y uses the X scores and Y scores. The model for this relationship is

$$\widehat{\mathrm{t}}={\mathrm{b}}_{\mathrm{i}}{\mathrm{u}}_{\mathrm{i}}$$

(11)

The primary distinction between PLSR and PCR is that PLSR incorporates linear combinations of both predictor and response variables into the model. This inclusion results in superior predictive performance and greater robustness when handling noisy data [49,54].

Throughout these regression analyses, the variance explanation represented as ‘X’ is an indicator considered in each independent variable, where higher values indicate that the model fits the data better for that independent variable. Independent variables, which explain 100% of the variance, are highly informative. Moreover, the cross-validation represented as CV determines how well the model generalizes to new data. In other words, cross-validation typically refers to the cross-validated prediction error, which helps to understand how well the model generalizes the data. A lower cross-validation value suggests better predictive performance. Composing the indicators analysis, higher values of magnitude (MG) suggest that the independent variable has a higher influence on the dependent variable, which may indicate greater impact or significance. In other words, MG analyzes the impact of the predictor on the outcome. Higher values suggest that the component has higher metrics of the outcome variable, which may indicate greater impact or significance [52,55].

To evaluate the performance of models, Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-square are metrics used in regression tasks, particularly. MAE evaluates prediction accuracy by calculating the average of the absolute differences between predicted values and actual results, making it easy to understand. MSE, on the other hand, determines the average of the squared differences between predicted and actual values, which places greater emphasis on larger discrepancies. RMSE stands for Root Mean Square Error. It is a standard way to measure the error of a model in predicting quantitative data. RMSE is the square root of the average of squared differences between predicted and actual values. It provides a measure of how well a model’s predictions match the observed data, with lower values indicating better accuracy. Additionally, the R-squared coefficient, or coefficient of determination, gauges how well a regression model fits the observed data, providing insight into the proportion of variance explained by the model. It indicates the proportion of the variance in the dependent variable explained by the independent variables, ranging from 0 to 1. An R-square of 0 means the model explains none of the variance, while an R-square of 1 means it explains all of the variance. While R-square shows the proportion of variance explained by the model, it does not indicate the accuracy or reliability of the predictions. It is possible to have a high R-square value but still have significant prediction errors. Therefore, it is crucial to consider additional evaluation metrics alongside R-square when assessing a regression model’s performance [56].

Finally, to ensure the robustness of PCR and PLSR models, bootstrapping techniques provide estimates of bias and standard errors (SEs), thereby offering a comprehensive evaluation of the stability and reliability of the parameter estimates. If the bias is significant, it suggests that the original estimate might be misleading or not fully accurate. It is important to address sources of bias, which could be due to model misspecification or sampling issues. SEs help to assess the precision of the original estimate. If the SE is large, the parameter estimates are less stable, raising concerns about the reliability of the results. In summary, small bias and low SEs in bootstrap statistics ensure that PCR and PLSR models are robust and reliable [55,56,57]. Additionally, Elastic Net Regression can be employed as an additional robustness check, particularly useful for assessing the stability of the model under different regularization conditions. Elastic Net combines L1 (Lasso Regression) and L2 (Ridge Regression) regularization, which helps to handle multicollinearity and manage model complexity by penalizing large coefficients and potentially setting some to zero [58]. By comparing Elastic Net estimates to those from PCR and PLSR, we can further validate the robustness of our models. Elastic Net Regression offers regularization advantages and feature selection not inherently present in PCR and PLSR. The coefficients from Elastic Net are directly interpretable and compared to those from PCR and PLSR to assess differences in the importance and effect of predictors. This comparison ensures that multicollinearity or overfitting influences the results and provides additional insight into the stability of the predictor effects. Furthermore, in Elastic Net, the focus is on the magnitude and sign of the coefficients to understand the impact of predictors. The coefficients’ size and direction are more informative than traditional significance tests in this context. Regularized models aim to find a subset of relevant predictors and balance model complexity with prediction accuracy. We used RStudio commands to run the bootstrapping and Elastic Net Regression.

3.3.2. Configurational Model Specification

We employ fs/QCA to analyze DEE configurations related to SDGs. We generated pooled cross-sectional and time-series data for 14 countries, with each observation covering two periods of analysis, a five-year average from 2013 to 2017 and from 2018 to 2022, to smooth the values reflecting digital advancements. This method enables us to examine which combinations of DEE elements are associated with sustainable economic, environmental, and social development. We conduct fs/QCA analyses for two periods (2013–2017 and 2018–2022) to identify changes in DEE profiles across countries with varying levels of SDGs (refer to Section 3.2). As a case-oriented approach, fs/QCA addresses the cross-national diversity in DEE components and accounts for heterogeneous effects among different country groups [51]. This approach allows us to investigate whether growth depends on multiple causal profiles conforming to these configurations [59,60]. We specify the configurational model as follows: fs/QCA requires the calibration of conditions (variables). After normalizing the data, we added a value of 0.001 to all scores of 0.5 to circumvent theoretical and methodological challenges associated with analyzing sets with 0.5 scores [51,59]. Following current practices [61,62], the calibration process in QCA allows us to benchmark each condition against specific values and identify strong and weak conditions related to DEEs on SDGs. We used fsQCA 3.0 software to run this analysis [63].

4. Results

4.1. Regression Models

The analysis of the PCR and PLRS results for sustainable social development highlights the varying strengths of different predictors in explaining the outcome (see

Table 5. Regression results on sustainable social development.

Models	PCR			PLRS
DEECs	CV	X	MG	CV	X	MG
DAC	0.3070	61.19	53.65	0.2816	60.48	60.71
DAD	0.2664	76.02	66.89	0.2406	75.29	72.54
DFI	0.2673	86.07	66.93	0.2235	80.62	78.62
DFR	0.2563	91.96	70.02	0.2002	83.93	84.63
DLI	0.2419	95.55	73.34	0.1799	89.00	87.67
DOP	0.2519	97.44	73.54	0.1740	96.06	88.58
DPR	0.2533	98.45	75.26	0.1689	97.34	89.21
DRI	0.1975	99.02	84.65	0.1671	98.65	89.40
DTA	0.1949	99.44	85.31	0.1672	99.32	89.48
DTT	0.1666	99.75	89.06	0.1672	99.67	89.48
DNE	0.1602	100.00	89.48	0.1673	100.00	89.48

Note: N = 140. Legend: Digital Entrepreneurship Ecosystem Components (DEECs), Digital Access (DAC), Digital Adoption (DAD), Digital Finance (DFI), Digital Freedom (DFR), Digital Literacy (DLI), Digital Openness (DOP), Digital Protection (DPR), Digital Rights (DRI), Digital Tech Absorption (DTA), Digital Tech Transfer (DTT), Networking (DNE), cross-validation (CV), variance explanation (X), and magnitude (MG).

). Both methods demonstrate that the predictors have a substantial influence on sustainable social development, with consistently high CV (cross-validated variance explanation) values across the board, suggesting robust model performance. Specifically, PCR and PLRS show high X (variance explained by the predictors) and MG (magnitude of the predictors’ impact) for most indicators, underscoring their significant contribution. For instance, Networking (DNE) shows the highest magnitude (MG = 89.48) and full variance explanation (CV = 100.00) in both methods, indicating its critical role in sustainable social development. On the other hand, Digital Access (DAC), while still important, has the lowest MG (53.65 in PCR and 60.71 in PLRS), reflecting a lesser impact. Across all predictors, PLRS consistently shows slightly higher MG values compared to PCR, suggesting that PLRS might capture a more nuanced relationship between the predictors and sustainable social development. Overall, predictors like Digital Tech Transfer (DTT), Digital Rights (DRI), and Digital Freedom (DFR) emerge as particularly strong contributors, with high X and MG values, indicating their vital roles in promoting sustainable social development. Appendix A presents the descriptive statistics of variables.

The regression analysis using PCR and PLRS methods for sustainable environmental development reveals notable differences in the impact of various DEE components (see

Table 6. Regression results on sustainable environmental development.

Models	PCR			PLRS
DEECs	CV	X	MG	CV	X	MG
DAC	0.229	61.19	48.92	0.2291	60.29	57.24
DAD	0.192	76.02	64.15	0.1929	75.19	71.10
DFI	0.178	86.07	64.42	0.1782	80.86	77.17
DFR	0.160	91.96	68.67	0.1609	85.39	81.57
DLI	0.158	95.55	72.34	0.1586	91.95	84.16
DOP	0.143	97.44	73.44	0.1436	96.06	85.97
DPR	0.138	98.45	74.86	0.1381	97.76	86.65
DRI	0.138	99.02	82.45	0.1385	98.72	86.90
DTA	0.137	99.44	83.12	0.1379	99.32	86.97
DTT	0.138	99.75	86.70	0.1380	99.68	86.97
DNE	0.138	100.00	86.97	0.1380	100.00	86.97

Note: N = 140. Legend: Digital Entrepreneurship Ecosystem Components (DEECs), Digital Access (DAC), Digital Adoption (DAD), Digital Finance (DFI), Digital Freedom (DFR), Digital Literacy (DLI), Digital Openness (DOP), Digital Protection (DPR), Digital Rights (DRI), Digital Tech Absorption (DTA), Digital Tech Transfer (DTT), Networking (DNE), cross-validation (CV), variance explanation (X), and magnitude (MG).

). Both models show consistently high cross-validated variance explanation (CV) values across all indicators, indicating robust model fit. Digital Tech Transfer (DTT) and Networking (DNE) emerge as the strongest predictors, with both indicators achieving the maximum CV (cross-validated variance explanation) of 100% and high magnitude of the predictors’ impact (MG) values (86.97 for both in PLRS). These results suggest that DTT and DNE have a substantial impact on sustainable environmental development, making them critical areas for focus. Interestingly, while Digital Access (DAC) has a similar CV in both models (approximately 22.9%), its MG significantly increases from 48.92 in PCR to 57.24 in PLRS, indicating that PLRS may better capture the relationship between DAC and sustainable environmental development. Digital Openness (DOP) and Digital Rights (DRI) also demonstrate strong influence, with MG values consistently above 85 in PLRS, underscoring their importance in sustainable environmental development. Across the board, PLRS tends to show higher MG values compared to PCR, suggesting that PLRS may provide a more detailed understanding of the predictors’ impacts. Overall, indicators like Digital Tech Transfer (DTT), Networking (DNE), Digital Openness (DOP), and Digital Rights (DRI) are critical to driving sustainable environmental development, as evidenced by their high CV and MG values across both models.

The regression results for sustainable economic development using PCR and PLRS models reveal consistent patterns across digital indicators, with PLRS providing a more detailed and impactful explanation of the predictors (see

Table 7. Regression results on sustainable economic development.

Models	PCR			PLRS
DEECs	CV	X	MG	CV	X	MG
DAC	0.229	61.19	48.92	0.2291	60.29	57.24
DAD	0.192	76.02	64.15	0.1929	75.19	71.10
DFI	0.178	86.07	64.42	0.1782	80.86	77.17
DFR	0.160	91.96	68.67	0.1609	85.39	81.57
DLI	0.158	95.55	72.34	0.1586	91.95	84.16
DOP	0.143	97.44	73.44	0.1436	96.06	85.97
DPR	0.138	98.45	74.86	0.1381	97.76	86.65
DRI	0.138	99.02	82.45	0.1385	98.72	86.90
DTA	0.137	99.44	83.12	0.1379	99.32	86.97
DTT	0.138	99.75	86.70	0.1380	99.68	86.97
DNE	0.138	100.00	86.97	0.1380	100.00	86.97

Note: N = 140. Legend: Digital Entrepreneurship Ecosystem Components (DEECs), Digital Access (DAC), Digital Adoption (DAD), Digital Finance (DFI), Digital Freedom (DFR), Digital Literacy (DLI), Digital Openness (DOP), Digital Protection (DPR), Digital Rights (DRI), Digital Tech Absorption (DTA), Digital Tech Transfer (DTT), Networking (DNE), cross-validation (CV), variance explanation (X), and magnitude (MG).

). Digital Tech Transfer (DTT) and Networking (DNE) emerge as the most influential predictors, both achieving perfect cross-validated variance explanation (CV) scores of 100% in PCR and PLRS, and high magnitude (MG) values of 86.97 in PLRS. This suggests that DTT and DNE play crucial roles in driving sustainable economic development. Similarly, Digital Openness (DOP), Digital Rights (DRI), and Digital Tech Absorption (DTA) also show strong predictive power, with MG values exceeding 85 in PLRS, indicating their significant impact on sustainable economic development. While Digital Access (DAC) and Digital Adoption (DAD) have lower CV scores of around 22.9% and 19.2%, respectively, their MG values increase notably in PLRS compared to PCR, particularly for DAC, where the MG rises from 48.92 to 57.24, suggesting that PLRS may capture more nuanced effects.

Overall, PLRS consistently shows higher MG values than PCR, highlighting its strength in modeling the impact of DEE components on sustainable economic development. The findings underscore the importance of Digital Tech Transfer (DTT), Networking (DNE), and related DEE indicators in fostering economic growth, making them key areas for strategic investment and policy development.

It is worth pausing to explore the analysis of Digital Protection (DPR) across the three outcomes—sustainable social, environmental, and economic development—that reveals its substantial and consistent impact in both the PCR and PLRS models, although the magnitude of its influence varies slightly depending on the context. In sustainable social development, DPR shows a strong predictive power, particularly in the PLRS model, with a magnitude (MG) of 89.21, indicating a significant role in fostering social sustainability by ensuring the privacy and security of digital interactions. Similarly, in the context of sustainable environmental development, DPR maintains a high MG of 86.65 in the PLRS model, demonstrating its importance in safeguarding digital infrastructures that support environmental monitoring and conservation efforts. When considering sustainable economic development, DPR again shows a strong influence, with an MG of 86.65 in PLRS, highlighting its critical role in securing the digital foundations necessary for economic resilience and growth. Across all three domains, the consistently high MG values in the PLRS model compared to PCR suggest that PLRS may better capture the nuanced and substantial impact of DPR, making it a pivotal factor in advancing comprehensive sustainable development. The recurring importance of DPR across these diverse sustainability dimensions underscores its essential contribution to creating a secure and resilient digital environment that supports long-term societal, environmental, and economic goals.

Overall, the performance comparison between PCR and PLRS for sustainable development—sustainable social development (SOC), sustainable environmental development (ENV), and sustainable economic development (ECO)—reveals identical results for both models (See

Table 8. Regression models’ performance evaluation.

Outcomes	PCR				PLRS
	MAE	MSE	RMSE	R-Square	MAE	MSE	RMSE	R-Square
SOC	0.106	0.020	0.143	0.894	0.106	0.020	0.143	0.894
ENV	0.093	0.015	0.124	0.870	0.093	0.015	0.124	0.869
ECO	0.073	0.009	0.099	0.890	0.073	0.009	0.099	0.890

Legend: sustainable social development (SOC), sustainable environmental development (ENV), sustainable economic development (ECO); Mean Absolute Error (MAE), Mean Squared Error (MSE); Root Mean Square Error (RMSE); Principal Component Regression (PCR); Partial Least Squares Regression (PLSR).

). Across all three results, both models exhibit the same Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Square Error (RMSE), and R-squared values, indicating no significant difference in predictive accuracy or model fit between PCR and PLRS. Thus, these indicators are strong in the context of evaluating sustainable development. Specifically, the R-squared values for SOP (0.894), ENV (around 0.870), and ECO (0.890) mean both models are similarly effective in capturing the variance of these dependent variables. The identical error metrics suggest that both methods yield comparable precision in estimating sustainable development measures, with no clear advantage of one model over the other in this context.

Finally, as the first robustness check, the bootstrapping results have high original values and low biases, indicating a consistent positive impact on sustainable development metrics (see Appendix B). In general, PLSR tends to exhibit a slightly lower bias compared to PCR. Conversely, the standard errors in PLSR are frequently higher than those in PCR, indicating greater variability in the estimates. This aligns with the understanding that while PLSR often provides more accurate predictions with smaller bias, it can also exhibit higher standard errors due to its reliance on a broader set of latent variables [56]. These insights underscore the trade-offs between bias and variability in PLSR and PCR, suggesting that the choice of method is informed and directed by the specific context and data characteristics of the study. As the second robustness check, comparing the Elastic Net Regression results with the PCR and PLRS models across the three sustainable development outcomes—social (SOC), environmental (ENV), and economic (ECO)—reveals consistent patterns in the impact of digital predictors and the models’ performance (see Appendix C). Overall, while Elastic Net, PCR, and PLRS agree on the importance of Digital Protection (DPR) and Digital Rights (DRI), Elastic Net differs in its treatment of variables like Digital Finance (DFI), Digital Adoption (DAD), and Digital Tech Transfer (DTT), due to its regularization techniques that shrink less relevant predictors more aggressively. Using cross-validation between prediction accuracy and model complexity, we compare the models’ performance. Elastic Net shows performance metrics, with R-squared values of 0.895 for social sustainable development, 0.871 for environmental sustainable development, and 0.887 for economic sustainable development, similar to the high R-squared values observed in PCR and PLRS, which are close to 0.89 across outcomes. RMSE values are also in a similar range, indicating comparable predictive accuracy between the models.

4.2. fsQCA Results

Section 3.1 demonstrated that DEE components impact sustainable social, environmental, and economic development. Here, we propose that a strategic combination of DEE components is equally crucial for developing sustainable development for Latin American countries. We examine whether these countries, at varying stages of SDG development, exhibit distinctive DEE enhancement profiles and how these evolve over time. Our focus is on determining whether countries with similar or different SDG levels have comparable or differing DEE profiles. We aim to identify whether there are more or less effective DEE enhancement profiles within and across different sustainable development levels, investigating whether specific DEE profiles support a transition from lower to higher outcome levels. The findings illustrating how DEE components influence each sustainable development outcome are in Figure 1 for the 2013–2017 period and in Figure 2 for the 2018–2022 period. In Figure 1 and Figure 2, the consistency threshold distinguishes between configurations that are or are not subsets of the outcome [51,59]. Coverage evaluates the extent to which each configuration and the overall solution explain the outcome. Raw coverage measures the proportion of memberships each condition contributes to the outcome, while unique coverage indicates the proportion of cases following the specific configuration leading to the outcome [63].

In comparing the fsQCA results for sustainable social development across the two periods, notable shifts in DEE components influencing sustainable social development emerge. During 2013–2017, six configurations (SC1–SC6) were identified with each configuration demonstrating high raw consistency (1) but varying in raw coverage, with SC2 (0.746) and SC1 (0.451) being the most impactful. Key conditions like Digital Access (DAC), Digital Literacy (DLI), and Digital Protection (DPR) were consistently present across all solutions, indicating their critical role in social development performance. However, Digital Finance (DFI) and Networking (DNE) had a less consistent presence, suggesting varying importance across different configurations. In contrast, the 2018–2022 period shows a slight consolidation into five configurations (SC1–SC5) with improved overall solution coverage (0.774 compared to 0.767 previously). Digital Finance (DFI) becomes more central, appearing in multiple configurations, particularly SC1 and SC5, which could reflect the growing importance of digital financial inclusion in driving sustainable social development. Additionally, the presence of Networking (DNE) across all configurations in 2018–2022 highlights its increasing role in social development, due to the rise of social media platforms. Notably, Digital Openness (DOP) appears less consistently, which may suggest a shift in its relative importance or its integration into broader Digital Literacy (DLI) initiatives. Overall, the transition between these periods indicates a broader, more interconnected digital ecosystem influencing social development, with an increased emphasis on financial inclusion and networking as key drivers.

Sustainable environmental development, across the two periods (2013–2017 and 2018–2022), reveals key shifts in digital conditions driving sustainable environmental development. During 2013–2017, six configurations (EV1–EV6) identified with Digital Access (DAC), Digital Literacy (DLI), Digital Protection (DPR), and Digital Tech Transfer (DTT) consistently appearing across all solutions, underscoring their importance in environmental performance. The overall solution coverage (0.765) and raw consistency (1) indicate a strong and coherent relationship between these DEE components and sustainable environmental development. However, the 2018–2022 period presents a slightly different landscape, with five configurations (EV1–EV5) and an improved overall solution coverage (0.786), reflecting a more concentrated and more effective set of digital conditions. Digital Finance (DFI), which appeared inconsistently in the earlier period, becomes more prominent in the 2018–2022 configurations, particularly in EV1 and EV5, suggesting an increasing role of financial technology in supporting environmental initiatives. Networking (DNE) also gains importance, present across all configurations in this latter period, reflecting the growing use of digital platforms to promote environmental awareness and action. The slight decline in the emphasis on Digital Openness (DOP) could indicate a shift towards more targeted digital interventions rather than broad access, aligning with the overall trend towards greater specificity and integration in digital strategies for environmental performance. This evolution highlights a more strategic use of digital tools in environmental initiatives, with a clear focus on financial inclusion and digital platforms as pivotal elements.

Finally, the results for sustainable economic development from 2013 to 2017 and from 2018 to 2022 reveal key shifts in digital factors. During the 2013–2017 period, six configurations (EC1–EC6) identified with Digital Access (DAC), Digital Literacy (DLI), and Digital Protection (DPR) consistently present across all solutions, reflecting their critical role in economic performance. The overall solution coverage was 0.770, with raw consistency at 1, indicating a strong and uniform relationship between DEE components and sustainable economic development. However, in the 2018–2022 period, the results show a more consolidated set of configurations (EC1–EC5) with improved overall solution coverage (0.797). Digital Finance (DFI), which is in the earlier period, becomes more central in configurations EC1 and EC5, highlighting the increasing importance of financial technology in driving economic growth. Networking (DNE) also gains prominence, appearing in all configurations in the latter period, signaling its enhanced role in economic development through digital platforms that facilitate business and commerce. The decreased emphasis on Digital Openness (DOP) may suggest a shift towards more specialized digital strategies rather than broad digital inclusivity, aligning with a trend towards precision in leveraging digital tools for economic gains. Overall, the transition from 2013–2017 to 2018–2022 suggests a more targeted approach to digitalization in economic strategies, with a significant focus on Digital Finance (DFI) and Networking (DNE) as pivotal elements for sustainable economic development.

5. Discussion

As digital landscapes evolve, the components driving high performance also change, highlighting the need for adaptive and forward-looking digital strategies [64]. Therefore, this study examines the relationship between DEEs and sustainable development growth across social, environmental, and economic dimensions in Latin American countries. Specifically, we (a) analyzed DEE metrics from 14 countries to highlight the significant impact of digitalization on sustainable development; (b) performed econometric tests to evaluate the links between specific DEE metrics and sustainability efforts using SDG data; and (c) investigated changes in DEE profiles across countries from 2013 to 2022.

Adapting the DEE framework from the literature [15], which identifies four digital components across eleven metrics—Digital Access (DAC), Digital Adoption (DAD), Digital Finance (DFI), Digital Freedom (DFR), Digital Literacy (DLI), Digital Openness (DOP), Digital Protection (DPR), Digital Rights (DRI), Digital Tech Absorption (DTA), Digital Tech Transfer (DTT), and Networking (DNE)— the findings reveal that the DEE profiles of Latin American countries significantly influence their prospects for sustainable development and their potential to improve SDG scores.

DEEs may foster innovations and create new digital technologies, driving significant contributions to economic sustainability [65]. Digital Access (DAC) and Literacy (DLI) are foundational, as they enable entrepreneurs to harness the power of digital tools and platforms, increasing productivity and market reach [3,4,66]. Studies indicate that regions with higher levels of digital access and literacy experience more robust entrepreneurial activity, leading to economic diversification and resilience [67]. Through advancements in fintech, Digital Finance (DFI) further democratizes entrepreneurship by providing broader access to capital, particularly for underrepresented groups [68]. This financial inclusion stimulates economic growth and reduces inequalities, aligning with sustainable development goals (SDGs).

An intriguing observation in the fsQCA results is the consistency of country-level configurations across different sustainable development measures (social, environmental, and economic) over the two periods (2013–2017 and 2018–2022). This stability suggests that once countries establish a digital trajectory, they tend to maintain it across multiple dimensions of sustainability. The digital strategies that work for economic performance, for example, are often similarly effective for social and environmental outcomes, indicating a strong, cohesive digital framework within these countries. This consistency also implies that countries do not significantly shift their digital policies or infrastructure based on specific performance goals, but rather employ a broad-based digital approach that supports multiple outcomes simultaneously.

Moreover, it is particularly noteworthy that less developed countries within the sample achieve high sustainable development despite relying on more incomplete configurations. Although these countries may not exhibit the full range of digital capabilities, they effectively leverage the digital elements they do possess. This suggests that even partial Digital Adoption (DAD) or Access (DAC), when strategically utilized, can lead to substantial gains in sustainable development. The ability of less developed countries to perform well with fewer resources challenges the notion that a full suite of digital tools is necessary for high sustainable development performance. Instead, it highlights the importance of strategic focus and the potential for targeted digital interventions to drive significant impact, even in resource-constrained environments. This adaptability among less developed countries underscores the versatility of digitalization and the potential for innovation in achieving sustainable outcomes.

On the social sustainability front, DEEs promote inclusivity and equity [69]. Digital Adoption (DAD) empowers marginalized communities by providing them with the tools to participate in the digital economy, thus reducing social disparities [70]. Research shows that digital platforms can facilitate access to education, healthcare, and social services, improving quality of life and social cohesion [71,72]. Additionally, Digital Rights (DRI) and Digital Freedom (DFR) ensure that the benefits of digital entrepreneurship are equitably distributed, fostering a more inclusive society [73,74]. The protection of these rights is essential for maintaining trust in digital systems and encouraging widespread participation in the digital economy [75]. Networking (DNE), Digital Tech Transfer (DTT), and Digital Protection (DPR) are the most impactful digital components underscoring their crucial role in fostering social sustainability by enabling widespread connectivity and collaboration [76,77]. DPR also emerges as a significant predictor, where it plays a vital role in ensuring the privacy and security of digital interactions, which are foundational for trust and social cohesion [28,78].

Environmental sustainability also benefits from a robust DEE [79]. Digital Tech Transfer (DTT), Digital Networking (DNE), and Digital Protection (DPR) are identified as key components indicating their substantial influence on environmental sustainability by facilitating the dissemination of eco-friendly technologies and promoting environmental awareness through digital networks [80,81]. For example, precision agriculture [82] and smart urban planning [83] utilize digital tools to optimize resource use and reduce environmental impact [84]. Particularly, in the 2018–2022 period, the growing prominence of Digital Finance (DFI) and Networking (DNE) in driving sustainable environmental development reflects the increasing role of digital platforms and financial technologies in supporting environmental initiatives [85,86]. This shift suggests a more strategic and integrated use of digital tools in environmental strategies [83,87].

In the context of sustainable economic development, we identify Digital Tech Transfer (DTT), Networking (DNE), and Digital Finance (DFI) as the most impactful variables, illustrating their critical roles in promoting economic resilience and growth through the diffusion of technology and the facilitation of digital commerce [88,89,90]. In the 2018–2022 period, consolidation of these variables, particularly with the rising importance of DFI and DNE, signifies the growing emphasis on digital financial inclusion and networking as drivers of economic development. This evolution reflects a shift towards more targeted digital strategies that prioritize financial and networked infrastructure as foundational elements for sustainable economic growth [68,91].

Despite these benefits, the development of DEE in Latin American countries also faces several challenges [19,92]. Digital inequality remains a significant barrier, with disparities in access to digital infrastructure and literacy levels hindering the full potential of DEEs [93]. Addressing these inequalities requires targeted investments in digital infrastructure and education, particularly in rural and underserved areas [94,95]. Policymakers must also create an enabling environment that supports digital entrepreneurship through favorable regulations and incentives [18,96]. Moreover, fostering a culture of innovation and collaboration is crucial for the future of DEEs [29,97]. Public–private partnerships can play a vital role in driving digital transformation and creating ecosystems that support entrepreneurial ventures [98].

Indicators such as Networking (DNE), Digital Tech Transfer (DTT), and Digital Protection (DPR) emerge as crucial factors across social, environmental, and economic sustainability, with PLRS better illustrating their impact. On the other hand, the results highlight shifts in DEE components’ importance over time, with a consolidation of Digital Finance (DFI) and Networking (DNE) in the later period (2018–2022). This evolution suggests a growing recognition of financial inclusion and digital platforms as central to sustainable development [74,97]. DEE components significantly impact the sustainability development growth of Latin American countries by driving economic innovation, promoting social inclusion, and supporting environmental sustainability [18,19,29,96]. However, challenges such as inadequate regulatory frameworks, limited cybersecurity infrastructure, and low levels of public awareness about digital risks need to be addressed across all sustainable development outcomes [99,100]. Policymakers must prioritize the development of comprehensive digital protection laws and invest in cybersecurity infrastructure [101,102]. Additionally, raising awareness and providing education on digital security practices can empower individuals and businesses to protect themselves in the digital space. The future of DEEs in the Latin America region hinges on addressing digital inequalities, fostering an enabling regulatory environment, and promoting a culture of digital literacy and rights. By overcoming these challenges, Latin American countries can harness the full potential of DEEs to achieve long-term sustainable growth, thereby contributing to the global SDGs.

Overall, as Latin American countries increasingly prioritize digital strategies to drive sustainable development, the role of DEEs becomes paramount. A new perspective emerging in the literature highlights the importance of regional collaboration within Latin America to foster digital innovation and sustainability [19,103]. Countries that engage in cross-border digital knowledge exchanges and partnerships—such as shared digital platforms for environmental monitoring [85] or regional fintech collaborations [68]—can leverage collective resources and expertise to accelerate their SDGs. This cooperative approach enables countries with more developed digital infrastructures to support those lagging behind, fostering an integrated and inclusive digital economy across the region [13]. Additionally, regional digital policies, supported by organizations such as the Inter-American Development Bank [104,105], can create cohesive strategies for scaling digital solutions, further enhancing economic, environmental, and social outcomes for sustainable development.

Finally, it is worth highlighting these findings with DEE profiles in developing countries. The results indicate that DEEs in Latin American countries often demonstrate unique patterns compared to their counterparts in more developed regions such as the EU and the USA. These differences are particularly evident in areas such as digital access, adoption, and finance, where Latin American countries face more significant infrastructural and socioeconomic barriers. Moreover, while developed countries benefit from advanced digital protections, literacy, and rights frameworks, Latin American nations are still working to establish and enforce these systems. This disparity highlights the challenges faced by the region in fully leveraging digital ecosystems for sustainable development.

Table 9. Major differences across key DEE indicators.

DEECs	Developed Regions	The 14 Latin American Countries
DAC	Better levels of digital access, with widespread broadband and mobile coverage [106].	Digital access varies widely; urban areas generally well-served, rural areas often under-connected [107].
DAD	Faster adoption of digital technologies across sectors [108,109].	Slower adoption rates, often due to financial, infrastructural, or cultural barriers [110].
DFI	Expansion of digital financial services with widespread usage of fintech and digital payment systems [111,112].	Growing fintech sector, but access and adoption remain uneven, particularly in underserved areas [113].
DFR	Stronger protections for digital freedom with established legal frameworks [114].	Varies across countries; some nations face restrictions or limited digital freedoms [115].
DLI	High digital literacy rates, with strong educational programs supporting tech skills development [116].	Digital literacy improving, but there is a significant gap in training, infrastructure, and resources [117].
DOP	High level of digital openness, with open access to global digital markets [118,119].	Digital openness improving but remains constrained by regulatory and infrastructural limitations [120].
DPR	Robust cybersecurity measures and data protection regulations in place [121].	Digital protection is developing, but many countries lag behind in implementing robust cybersecurity posture [122,123].
DRI	Well-established digital rights frameworks ensuring privacy and freedom of expression online [124].	Digital rights protections vary, with some countries lacking strong privacy or data protection laws [122,123].
DTA	Better integration of new digital technologies across industries [125,126].	Moderate absorption, hindered by limited resources and slower technological diffusion in some regions [110].
DTT	High levels of technology transfer, supported by strong institutions and multinational collaborations [127].	More reliance on external tech solutions with slower adaptation to local needs [128].
NET	Stronger networks and ecosystems supporting digital entrepreneurship [129].	Networking opportunities are improving but remain limited compared to more developed ecosystems [12].

, as follows, outlines the major differences across key DEE indicators between developed countries and the sample of 14 Latin American countries analyzed in this study.

Regardless of the challenges faced by Latin American countries in the evolution of DEE indicators, one key distinction is the prevalence of innovative grassroots solutions that arise from constrained digital resources. In developing countries, limited access to advanced digital technologies often sparks creative, low-cost alternatives that effectively address local challenges [130]. For example, mobile-based applications for financial services and health care are frequently developed in response to gaps in traditional infrastructure [131,132]. Additionally, the role of informal digital networks and community-based platforms becomes more pronounced, as they often fill the void left by inadequate formal systems. This reliance on informal networks can enhance social capital and foster local entrepreneurial ecosystems, driving inclusive growth and resilience [133]. Furthermore, developing countries may exhibit a higher rate of digital leapfrogging, where they bypass traditional stages of technology adoption to implement cutting-edge solutions directly [134]. This phenomenon allows them to rapidly integrate advanced digital tools that are tailored to their specific needs, potentially accelerating progress towards sustainable development goals [135]. These differences highlight the importance of context-sensitive approaches in designing DEE strategies, recognizing that the innovative use of limited resources and adaptive practices in developing countries can offer valuable lessons and contribute significantly to sustainability efforts.

6. Conclusions

This paper explored the level at which the DEE framework influences sustainable development across social, environmental, and economic dimensions. We analyzed DEE metrics for 14 countries over a decade (2013–2022). Our findings reveal that different DEE domains, based on their relative weight and development patterns, correlate with various sustainability growth trends. The results indicate that sustainability growth is by ongoing digital transformation and evolving digital activities rather than adherence to a fixed technological target. Political and institutional forces shape this process but also require significant government intervention, as early-stage technology markets often poorly allocate investments (e.g., renewables). Policymakers must anticipate changes in entrepreneurial ecosystems and their impact on digital transformation and sustainability growth. Understanding which digital efforts will best address future demands for SDG achievements is crucial.

Our DEE framework and results suggest that digital domains are interchangeable in generating sustainable development. Each DEE metric serves unique functions and has distinct impacts. Thus, the strategic selection of digitalization is critical. Furthermore, policy should aim to steer the SDG system towards digital initiatives that can significantly drive sustainability growth. However, our analysis indicates that concentrating solely on emerging sectors to develop technologies while neglecting other crucial digital efforts, such as digital rights, digital freedom, and digital literacy, will have a limited effect on long-term SDG growth. Effective innovation policy must be capable of guiding structural rationalization for DEE progress and enhancement.

DEEs are linked to the most substantial impacts on sustainability growth [136]. Yet, emergent technologies are rarely present in the configurations of low- and middle-income countries, including those in Latin America, posing challenges for these economies to engage with frontier technologies. This situation highlights the need for substantial efforts and investments. A thorough investigation into which DEE configurations have the highest potential for driving sustainability growth, and their underlying mechanisms, is essential for future research. Specifically, policy should prioritize developing a DEE strategy portfolio that considers the various functions of each digital effort and how they complement one another.

In this context, the originality of our research lies in updating the DEE indicators needed for shifts towards sustainable development and understanding how these indicators must evolve at various levels of social, environmental, and economic development. Latin American countries face significant challenges with innovation policies that focus solely on emerging technologies. Our analysis emphasizes the importance of monitoring DEEs and understanding their varying impacts and roles in promoting sustainability. A deep understanding of how different combinations of DEE components can enhance sustainability is crucial for setting policy priorities. This scenario presents specific challenges and may necessitate institutional changes that extend beyond related diversification to create additional metrics for SDGs. Promoting capability advancements into new DEE domains might be necessary, requiring institutional and socioeconomic reforms.

6.1. Managerial Implications

This study offers valuable insights for the telecom sector by underscoring the pivotal role that DEEs play in advancing sustainable development across Latin America. The findings highlight how improvements in digital infrastructure, such as enhanced Digital Access (DAC) and Digital Networking (DNE), are directly linked to positive outcomes in economic, social, and environmental development. For the telecom industry, these insights underscore the strategic importance of investing in and expanding digital networks to support entrepreneurship and innovation. By strengthening digital connectivity and access, telecom providers can facilitate greater inclusivity and support the growth of local startups, which in turn drives broader economic development. Additionally, the study’s focus on regional collaboration and the adaptation of digital technologies in resource-constrained environments presents opportunities for telecom firms to tailor their services and infrastructure investments to meet specific local needs. This approach not only aligns with corporate social responsibility goals but also positions telecom firms as key enablers of sustainable development, thereby enhancing their market presence and fostering long-term growth in emerging markets.

Additionally, it provides significant value to the tech sector by illustrating that the integration and advancement of digital technologies are crucial for fostering innovation and achieving sustainable outcomes. For tech firms, this research highlights the importance of focusing on digital solutions that address local challenges and support entrepreneurial growth. By understanding the unique DEE profiles and their impacts, tech firms can better tailor their products and services to the needs of emerging markets, enhancing their relevance and effectiveness. Furthermore, this study’s emphasis on regional collaborations and adaptive digital strategies offers tech companies valuable insights into how they can engage in partnerships and invest in technologies that promote inclusivity and resilience. This knowledge not only supports the development of targeted tech solutions but also helps tech companies align their innovation efforts with broader sustainability goals, ultimately contributing to their long-term success and competitive advantage in the global market.

6.2. Limitations

This study does not go without limitations. First, the constraints of available data sources mean we could not capture the full spectrum of digital initiatives across Latin American economies. Due to data limitations, we could not include indicators such as digital payments in our analysis. Despite our efforts to minimize data bias, this limitation affects the generalizability of our findings. Second, our aggregate analysis focuses on countries based on patterns of SDG growth dynamics, which prevents us from examining the varied impacts of specific digital initiatives, such as patents, on market competitiveness and sustainability. Additionally, different methods of combining DEE components could yield varying insights into the relationship between DEE diversification and sustainable development efforts.