1. Introduction
Global South economies are highly vulnerable to climate change but have limited capacity for adaptation (African Development Bank, 2024). While undergoing rapid economic transformations largely reliant on high-carbon energy sources, these countries have ratified the Paris Agreement and committed to Nationally Determined Contributions. To align with the green economy agenda, low-carbon energy transition (LCET) initiatives are essential. Given LCET's knowledge-intensive nature, research plays a pivotal role, although existing information is often limited and unclear. This study contributes to the literature by reviewing existing evidence, identifying key gaps and offering actionable insights for research and policy on LCET in the Global South.
Theoretically, the LCET process involves complex interactions among various social actors (Geels et al., Reference Geels, Sovacool, Schwanen and Sorell2017). The multi-level perspective (MLP) theory, rooted in socio-technical transition, explains transitions through interactions at three levels: niches, regimes and landscapes (Rip and Kemp, Reference Rip, Kemp, Rayner and Malone1998; Geels, Reference Geels2002; Smith et al., Reference Smith, Voß and Grin2010). The niche level focuses on emerging innovations that challenge prevailing socio-technical systems, accelerating sustainable paths (Geels and Schot, Reference Geels and Schot2007). The socio-technical landscape encompasses broader, slow-changing trends – such as geopolitics, demography, ideology, spatial structures and external factors like economic crises or political upheavals (Geels et al., Reference Geels, Sovacool, Schwanen and Sorell2017). MLP argues that transition can only happen when processes within and between these levels align. Analysing these complex processes and their interactions is key to understanding socio-technical change.
Complementarily, transition management theory emphasizes strategic collaboration among government, policymakers, businesses and civil society actors (government, policymakers, civil society and businesses) as key in shaping green energy transitions (Loorbach, Reference Loorbach2010). Together, these theories highlight the multi-dimensional nature of LCET, emphasizing the role of socio-economic and technological factors (Wang and Huang, Reference Wang and Huang2022; Saraji and Streimikiene, Reference Saraji and Streimikiene2023). Empirical evidence shows that LCET is influenced by multiple drivers – economic, political, social, environmental and technological factors. Comprehensive reviews by Muhire et al. (Reference Muhire, Dickson, Adaramola and Natongo2024) and Saraji and Streimikiene (Reference Saraji and Streimikiene2023) detail these factors, while Sovacool and Griffiths (Reference Sovacool and Griffiths2020) explore the influence of culture.
Technological factors, such as infrastructure, technology and technical standards, are especially pivotal. In this context, digital technologies have emerged as transformative tools for decarbonization and advancement of green economies (Yang et al., Reference Yang, Nie, Li and Wang2023, Reference Yang, Wang, Dong, Dong, Wang and Fu2024). Global policy forums, including the 2015 World Summit on the Information Society +10 and the 2004 Partnership for Measuring ICT, have emphasized the role of digital tools in achieving sustainable development goals, especially through increased information and communication technology (ICT) penetration in the Global South (IISD, 2010).
Digital infrastructure offers numerous benefits (Röller and Waverman, Reference Röller and Waverman2001; IISD, 2010; Munoz and Naqvi, Reference Munoz and Naqvi2017; Hjort and Poulsen, Reference Hjort and Poulsen2019; Tao et al., Reference Tao, Sui, Liu, Qi, Zhang, Song, Guo, Lu and Nee2019; Hawash and Lang, Reference Hawash and Lang2020; IEA, 2020). It expands firms' knowledge base (Yin and Yu, Reference Yin and Yu2022), supports green technology innovation (Yang et al., Reference Yang, Nie, Li and Wang2023), facilitates information sharing and knowledge absorption and helps match green finance supply with demand (Yang et al., Reference Yang, Nie, Li and Wang2023). Moreover, digital platforms such as energy management systems help firms obtain verifiable climate data for low-carbon investments. Technologies such as artificial intelligence (AI), internet of things (IoT), blockchain and big data help optimize production, distribution and consumption processes, reducing material and fuel demand and supporting LCET (Belhadi et al., Reference Belhadi, Venkatesh, Kamble and Abedin2024).
In the energy sector, digital technologies are significantly reshaping energy demand and supply dynamics. They enable energy demand management via automation (Lee, Reference Lee2020; Li et al., Reference Li, Zhang, Li and Hao2024) and improve the monitoring and optimization of renewable energy systems (Abdalla et al., Reference Abdalla, Nazir, Tao, Cao, Ji, Jiang and Yao2021; Zhang et al., Reference Zhang, Ling and Lin2022). The Climate Group and the Global e-Sustainability Initiative (2008) estimate that effective energy demand management and grid monitoring could reduce global energy consumption by up to 30 per cent. In China, Wang (Reference Wang2025) finds that AI, big data, blockchain and IoT significantly advance low-carbon transitions among energy enterprises. Given the strong link between carbon emissions and fossil fuel use, such technologies can substantially lower energy needs and carbon footprints by streamlining processes and operations.
However, the impact of digital technologies on LCET is complex and context-dependent. On one hand, they enhance efficiency; on the other, they contribute to rising energy demand – particularly in data centres and device production (Jones, Reference Jones2018). Their net effect on LCET is therefore not straightforward. Moreover, disparities in access to digital tools and energy resources across regions and demographic groups (Schopp et al., Reference Schopp, Schelenz, Heesen and Pamelec2019) affect both the inclusiveness and effectiveness of LCET. Addressing these challenges requires a deeper understanding of how digital technologies interact with energy systems, social structures and policy environments in the Global South. In response, this review attempts to provide some practical ways of addressing these challenges by identifying the key knowledge gaps and offering actionable insights to guide both future research and policymaking in support of an inclusive and just LCET in the Global South.
2. Digitalization in the Global South
Over the past decade, the Global South has experienced gradual integration of digital technologies, driven by key regional initiatives. For instance, the African Union's Africa Smart initiative, launched in 2013, aimed to introduce digital products, harmonize communication infrastructure and expand broadband access across the continent (Sausen, Reference Sausen2020). Similarly, the African Continental Free Trade Area prioritizes digital trade, especially e-commerce, among member states. In the financial sector, the rise of e-money, mobile payment platforms and digital banking has been notable. In the energy sector, the adoption of smart pre-paid meters and electronic billing systems is expanding.
Despite this progress, substantial digital potential remains untapped in Africa and other Global South economies. The Dalberg Report (2013) emphasized underutilized opportunities among businesses and small- to medium-sized enterprises. While internet access has improved, the digital divide persists. International Telecommunication Union (ITU) data from 2017 to 2020 (see figure A1 in the online appendix) show continuing disparities. In 2018, total telecommunications investment in Africa reached US$6.7 billion, with South Africa and Nigeria receiving the largest share (ITU, 2021).
In contrast, regions such as Asia, the Pacific and South and Central America host some of the world's leading digital economies, yet they also exhibit stark internal disparities. While developed economies in these regions boast near-universal internet access, least-developed countries lag significantly. For example, only 15 per cent of least-developed economies in Asia and the Pacific had internet access, compared to 90 per cent in developed economies; the figures are 30 and 90 per cent, respectively, in South and Central America (ITU, 2021).
Across the Global South, the digital divide remains a pressing concern (Schopp et al., Reference Schopp, Schelenz, Heesen and Pamelec2019). ITU data from 2021 indicate that gender and youth disparities are more pronounced in Africa, Asia and the Pacific than in Latin America. Young people generally have better internet access than older generations, and urban areas are significantly better connected than rural areas. While gender gaps are narrowing – with near parity in the Americas – significant disparities persist in Africa, the Arab States, least-developed countries and landlocked developing countries (see figure A2 in the online appendix). These inequalities raise critical questions about the implications of digital exclusion for the digital revolution and broader sustainability outcomes. Bridging the digital divide is essential for ensuring equitable access to digital technologies and their benefits.
A variety of economic, social and political constraints continue to limit digitalization in the Global South. Economically, high costs of digital tools, coupled with low incomes, restrict access – particularly for women and marginalized groups (Maxwell and Maxwell, Reference Maxwell and Maxwell2014; Majama, Reference Majama2018; Keja and Knodel, Reference Keja and Knodel2019; Kemp, Reference Kemp2019; Rizal et al., Reference Rizal, Rusdiana, Setiawan, Siahaan and Ridwan2020; ITU, 2021). In some cases, digital taxes further exacerbate the problem (Lirri, Reference Lirri2021).
Socially, the dominance of major tech firms from the United States and China raises concerns about dependency and digital colonialism (Kwet, Reference Kwet2019; Wakunuma, Reference Wakunuma2019). These monopolies often control access, user data and infrastructure, posing risks to civil liberties and privacy (Moore, Reference Moore2016; Kwet, Reference Kwet2019). The banning of companies like Huawei and ZTE by the Five Eyes intelligence alliance highlights these geopolitical tensions.
Digital literacy is another key challenge, especially for non-English speakers and marginalized populations (World Wide Web Foundation, 2015, 2016; Sègla, Reference Sègla2019). Limited literacy reduces digital tool adoption and deepens existing inequalities. Calls for integrating local cultures and languages into digital platforms to enhance accessibility (Sègla, Reference Sègla2019) remain largely unexplored in the literature. Politically, weak infrastructure, ineffective regulatory frameworks and low public expectations (Dalberg Report, 2013; Mahrenbach, Reference Mahrenbach2018) further limit digital expansion. These factors contribute to urban-centric digitalization, leaving rural communities behind and widening the divide.
Overcoming these challenges requires comprehensive strategies – improving core infrastructure, expanding digital literacy programmes, lowering costs and enacting policies that ensure data privacy and cybersecurity (Schia, Reference Schia2018). Equally important is fostering inclusive innovation that reflects local needs and cultural contexts.
Understanding these constraints within specific sectors, such as energy, is also vital, as challenges often differ across supply chains. Further research is needed to identify sector-specific barriers to digital adoption and devise tailored solutions. This review attempts to contribute to that effort by highlighting key constraints and offering practical pathways for addressing the digital divide and enhancing the role of digital technologies in sustainable development across the Global South. The literature on digitalization in the energy sector, though growing, remains underdeveloped. Additionally, the complexity of the underlying relationships and the context-specific nature of existing evidence often result in a body of research that is large, fragmented and heterogeneous.
3. Energy system in the Global South
The energy sector in the Global South faces persistent and complex challenges. Despite the growing availability of cleaner alternatives, unclean energy sources continue to dominate the energy mix (Enerdata, 2019). In regions such as Africa, Central and South America and Southeast Asia, coal, biofuels, waste and crude oil remain central to electricity generation (see figures A3a–A3c in the online appendix). Additionally, millions of people still rely on biomass, coal and kerosene for cooking (IEA, 2018), underscoring the continued prevalence of unsustainable energy practices.
Governments across the Global South are increasingly urged to diversify their energy portfolios – not only to improve energy security but also to harness opportunities presented by the Fourth Industrial Revolution, driven by AI and digital technologies. As energy demand surges and concerns of climate change intensify, there is an urgent need to accelerate the shift towards low-carbon energy sources to support sustainable economic development.
Access to electricity remains a critical concern, particularly in sub-Saharan Africa, where only 48.4 per cent of the population had electricity access in 2020 (ESMAP (Energy Sector Management Assistance Program), 2022). Chronic energy deficits – averaging 1.5 per cent of GDP – are largely attributed to both technical and non-technical losses (Pistelli, Reference Pistelli, Hafner and Tagliapietra2020). Addressing these energy challenges raises several key questions:
How can the share of renewable and clean energy in the energy mix be increased? Investments in renewable energy, especially wind and solar, are growing across Africa and other countries in the Global South (Enerdata, 2019), driven by falling costs, international pressure and energy security concerns. However, structural, institutional, economic and socio-cultural barriers still limit the full adoption of clean energy sources.
How can cleaner energy sources be effectively integrated into the broader energy system? Many developing countries, particularly in Africa, face challenges such as high upfront costs and limited technical capacity, which hinder the integration of gas and renewable sources (Hostettler, Reference Hostettler, Hostettler, Gadgil and Hazboun2015). However, advances in digital technologies are helping enable hybrid systems that combine variable renewables like solar and wind with more stable sources such as gas and nuclear power.
How can energy efficiency be improved across the sector? Enhancing energy efficiency requires tackling a range of barriers – economic, technical, market-related and regulatory. While the UN Sustainable Development Goal Target 7.3 seeks to double the global energy efficiency improvement rates by 2030, most governments in the Global South – aside from China – have yet to make significant financial commitments to meet this goal, especially in Africa (IEA, 2020).
4. Role of digitalization in the energy sector
Digital technologies play a crucial role in transforming the energy sector, particularly by enabling real-time data management, supporting the shift to cleaner energy, integrating variable renewable energy sources with natural gas and nuclear energy and enhancing energy efficiency. Tools such as AI, IoT and digital sensors enable managers to coordinate supply and demand in decentralized electricity systems, optimize energy demand-side management, facilitate carbon capture and storage and improve operational efficiency. They also support the transition towards electrified and automated transportation systems (IEA, 2020). As a result, variable renewables are becoming increasingly dispatchable, making it easier to integrate them into national energy grids.
While the potential of digital technologies in the Global South's energy sector is widely acknowledged, there remains a lack of empirical understanding in the literature about how these technologies support clean energy transitions, boost energy efficiency and enable energy source integration in the region.
Several characteristics of the energy sector may hinder the adoption of digital solutions. Energy investments are typically capital-intensive and long-term, requiring economic stability – something that can be difficult to ensure amid structural transitions in the region. Moreover, the sector's highly regulated nature may limit space for innovation, while the digital transformation of energy systems can generate spillover effects that impact other sectors in unforeseen ways.
To successfully implement digital technologies in the energy sector, conducive economic, political, social and legislative frameworks are essential. Further research is therefore needed to examine how these contextual factors influence the digital transformation of energy systems in the Global South.
5. Theory on digital technology and energy consumption debate
This section explores how digital technologies impact energy consumption and the transition to a low-carbon energy system. Scholars such as Brookes (Reference Brookes1978), Khazzoom (Reference Khazzoom1980), Daly (Reference Daly1990), Ayres (Reference Ayres1999), Sorrell (Reference Sorrell2009) and Brockway et al. (Reference Brockway, Barrett, Foxon and Steinberger2014) identify four key mechanisms through which digital technologies affect energy use: direct, scale, efficiency and structural effects (see also Brock and Taylor (Reference Brock, Taylor, Aghion and Durlauf2005) and Lange et al. (Reference Lange, Pohl and Santarius2020) for detailed discussions).
Direct effect: The production, operation and disposal of digital technologies increase energy consumption throughout their lifecycle (Morley et al., Reference Morley, Widdicks and Hazas2018; Lange et al., Reference Lange, Pohl and Santarius2020).
Scale effect: Digital technologies stimulate economic growth but may also displace labour through automation (Frey and Osborne, Reference Frey and Osborne2017) and the substitution of cognitive human work (Brynjolfsson and McAfee, Reference Brynjolfsson and McAfee2014; Wolter et al., Reference Wolter, Mönning, Hummel, Weber, Zika, Helmrich, Maier and Neuber-Pohl2016). These changes can create new products and consumption patterns, thereby raising overall energy demand. Moreover, they can widen wage disparities and exacerbate income inequality (Brynjolfsson and McAfee, Reference Brynjolfsson and McAfee2014; Lange and Santarius, Reference Lange and Santarius2020). Together, the direct and scale effects may increase energy use and carbon emissions, potentially slowing the transition to low-carbon energy systems.
Efficiency effect: Digital technologies improve energy efficiency by optimizing production, enhancing operational performance and streamlining smarter distribution and logistics (Berkhout and Hertin, Reference Berkhout and Hertin2004; Koomey et al., Reference Koomey, Berard, Sanchez and Wong2011; Coroama and Hilty, Reference Coroama and Hilty2014). However, these gains are often offset by rebound effects, where increased efficiency lowers costs and stimulates greater consumption. The net impact, therefore, depends on the magnitude of these rebound effects.
Structural effect: Digitalization can reshape economies by shifting the activity towards the service sector (tertialization), which typically has lower energy intensity. However, because digital technologies themselves are energy-intensive, the overall impact on energy consumption remains uncertain.
In sum, the relationship between digital technologies and energy use is complex and context-dependent. Whether these technologies ultimately reduce or increase energy consumption and carbon emissions hinges on which effects dominate. As such, their role in advancing LCET remains uncertain and requires further empirical investigation.
6. Method and data
The primary aim of this study is to establish the boundary of evidence on the emerging topic of digitalization in the energy sector and to identify gaps in the current literature. To achieve this, a systematic literature review was conducted to assess the size, depth and thematic focus of the available research.
The review began with a broad definition of the topic, using keywords such as digitalization, ICT, social inclusion, energy demand and ecological footprint, without restrictions on geography or publication year. This initial approach was intended to capture the full breadth of literature on the subject. From this broad pool, we then identified studies specifically focused on countries in the Global South, guided by the core research questions of this review.
The next step involved selecting the types of studies to include and identifying appropriate sources. While both qualitative and quantitative studies were considered, priority was given to those offering quantitative evidence. This was to help assess methodological advancement in understanding the nexus between digitalization and energy sector, including its spillover effects on household welfare and socio-economic inclusion.
To build the primary database, we conducted searches using major search engines – Scopus, Web of Science and Google Scholar. The search results were merged, and duplicates were removed. Bibliographies of the selected studies were reviewed to identify additional relevant grey literature. Supplementary documents, such as reports, working papers and conference papers, were also included from online sources. The core database for this review was constructed using the following inclusion and exclusion criteria, as shown in table A1 in the online appendix.
7. Empirical evidence of energy effects of digital technologies from the Global South
7.1. Digital technologies and energy supply
Digital tools are increasingly being integrated into the energy sector, offering substantial benefits across the supply chain. In oil and gas, they enhance data management and processing, reduce operational costs, improve safety, increase production and boost profitability (IEA, 2017). In the power sector, digital tools improve efficiency in electricity generation, transmission and distribution – lowering investment needs, fuel consumption, operational costs and carbon emissions (IEA, 2017). Beyond efficiency gains, digital technologies support the broader transition towards a low-carbon energy system by reducing the sector's carbon footprint.
Despite these benefits, research on the impact of digitalization on energy supply – particularly in the Global South – remains limited. Anecdotal evidence suggests that digital technologies can facilitate decentralized renewable energy production and strengthen supply systems (Banales, Reference Banales2020). Case studies from Tamil Nadu, India (Shahinzadeh et al., Reference Shahinzadeh, Moradi, Gharehpetian, Nafisi and Abedi2019; Priya and Rekha, Reference Priya and Rekha2020) and Nigeria (Chukwuorji et al., Reference Chukwuorji, Saka, Inuwa, Hussein, Thomas and Adeshina2019) demonstrate their role in optimizing renewable energy integration, improving power generation and distribution and enhancing coordination across supply and storage systems (Shahinzadeh et al., Reference Shahinzadeh, Moradi, Gharehpetian, Nafisi and Abedi2019; Al-Othman et al., Reference Al-Othman, Tawalbeh, Martis, Dhou, Orhan, Qasim and Ghani Olabi2022; He et al., Reference He, Zheng, Ma, Wang, Kong and Zhu2022).
However, most of these studies rely on optimization algorithms and simulation models, which offer limited internal validity. There is a notable lack of rigorous research using causal inference techniques (experimental or observational) to assess the impact of digital technologies across various segments of the energy supply chain in these contexts.
7.2. Digital technologies and energy demand
Digital technologies also influence energy consumption through both direct and indirect channels, with the overall impact depending on which effects dominate. While the direct effects are well documented, comprehensive assessments that account for both are rare. Approximately 90 per cent of the studies reviewed focus exclusively on direct effects, with few investigating the net impact of digitalization on energy demand (see table A2 in the online appendix for details). This highlights the methodological difficulties in capturing the full spectrum of digitalization's influence on energy consumption.
7.2.1. Direct energy-related effects of digitalization
Studies examining the direct impact of digital technologies on energy demand often rely on observational data, limiting their internal validity. The absence of robust counterfactuals means these studies often identify correlations rather than causal effects.
Beyond methodological limitations, the findings in this literature are highly mixed, with studies reporting positive, negative, non-linear or even negligible effects. These inconsistencies can be attributed to variations in local context, types of digital technology interventions, methodological rigor and the scope of sample. For example, Higon et al. (Reference Higon, Gholami and Shirazi2017) and Danish et al. (Reference Danish, Wang and Latif2019) found that digital technologies reduce energy demand in developed economies but increase it in developing ones, highlighting the importance of context-specific analysis and cautioning against broad generalizations.
Understanding this ambiguity is crucial. It points to the influence of local context conditions, technological diversity and methodological differences in shaping research outcomes. While a meta-analytic review could help synthesize and clarify these inconsistencies, no such study currently exists, underscoring a notable gap in the literature.
One promising approach to addressing this ambiguity is sector-specific analysis. Since energy and technology needs vary significantly across industries (see Cho et al., Reference Cho, Lee and Kim2007; Khayyat, Reference Khayyat2015), research that focuses on specific sectors can avoid overgeneralization and provide insights within relevant contextual boundaries. However, energy sectoral studies – most of which are concentrated in Asia – suggest that the same digital technology can have differing effects depending on the sector.
For instance, Zhou et al. (Reference Zhou, Zhou and Wang2018a) found that ICT increased energy demand more in China's devices sector than in its services. Similarly, Dehghan Shabani and Shahnazi (Reference Dehghan Shabani and Shahnazi2019) reported that ICT investments raised energy consumption in Iran's industrial sectors but reduced it in transportation and services. Malmodin et al. (Reference Malmodin, Moberg, Lundén, Finnveden and Lövehagen2010) and Malmodin and Lunden (Reference Malmodin and Lunden2018) further showed that ICT contributes significantly to global energy use, with sectoral differences in intensity depending on network operation and user devices.
7.2.2. Sectoral insights: transport, buildings and industry
The transport sector accounts for 28 per cent of global energy consumption and 23 per cent of carbon emissions (IEA, 2017), with demand projected to reach 165 exajoules by 2050 – driven largely by road freight and passenger vehicles. Yet, digital technologies such as intelligent transportation systems, digital sensors and satellite communication are improving efficiency and reducing energy use (US Department of Transportation, 2017). However, research on their energy impacts, particularly in developing countries, remains scarce. Tools like road traffic detectors, GPS, radio-frequency identification and automation systems are vital for building sustainable transport systems in the Global South. In China, Lee (Reference Lee2020) found that AI enhances electric vehicle automation, while Hu et al. (Reference Hu, Lin, Li, Hou, Chu, Zhao, Zhou, Jiang and Zhang2024) demonstrated that reinforcement learning algorithms can significantly improve fuel efficiency in plug-in hybrid and fuel-cell vehicles.
Buildings account for 55 per cent of global final energy consumption, with electricity use in the sector rising steadily over the past 25 years (IEA, 2017). Projections suggest energy demand in this sector will double by 2040. However, digital technologies – such as smart meters, light sensors and automated heating and cooling systems – could reduce energy use by up to 10 per cent, potentially saving an estimated 65 PWh between 2017 and 2040. Despite their promise, research on the energy impacts of these technologies, especially in the Global South, is limited. For example, Gray et al. (Reference Gray, Ayre, Hinton and Campbell2020) assessed the energy footprint of home automation systems, finding considerable impacts, particularly in mid-sized residential homes. Broader and more context-sensitive research is needed to evaluate the sustainability implications of these tools across diverse building environments.
The industrial sector consumes 38 per cent of global final energy and contributes 24 per cent of carbon emissions (IEA, 2017). As industrial expansion accelerates in emerging economies, digital technologies are playing a growing role in boosting energy efficiency. Smart meters and sensors help firms identify inefficiencies, optimize maintenance scheduling and reduce downtime.
Beyond the plant level, digitalization is transforming production processes through innovations like industrial robots, 3D printing and digital twins. These technologies enhance precision, reduce material waste and enable real-time process monitoring, with increasing adoption in emerging markets (IFR, 2016). For example, chemical manufacturers have shortened batch production time by 30 per cent using digital simulations (World Economic Forum, 2016). Cloud-based platforms and autonomous control systems further optimize logistics and supply chains, contributing to energy savings.
Despite promising developments, comprehensive, large-scale studies on the energy implications of industrial digitalization – particularly in the Global South – remain limited. More empirical research across diverse regional contexts is essential to fully understand how digital technologies are reshaping industrial energy dynamics and advancing sustainability.
7.2.3. Digitalization and sustainable use of energy (energy efficiency)
Digital technologies hold significant promise for advancing energy efficiency and promoting sustainable energy use, aligning with environmental and energy targets (World Energy Council, 2018). ICT-based monitoring and environmental management systems can enhance transparency and accountability in energy management (Verma et al., Reference Verma, Savickas, Strüker, Buettner, Kjeldsen and Wang2020). However, some digital tools – particularly AI, IoT and blockchain – can paradoxically increase energy consumption, especially in data centres and network infrastructure (Verma et al., Reference Verma, Savickas, Strüker, Buettner, Kjeldsen and Wang2020). Although end-user devices such as computers and cell phones generally have low power requirements due to efficient transistors and semiconductors, the overall impact of digital tools on energy consumption efficiency remains difficult to predict.
This review reveals a notable scarcity of studies focusing on the effects of digital technologies on energy efficiency in the Global South, with existing findings being inconclusive. Some research, such as Anjana and Shaji (Reference Anjana and Shaji2018), suggests that ICT adoption reduces energy costs and enhances energy efficiency. Similarly, studies by Bento (Reference Bento2016), Wu and Raghupathi (Reference Wu and Raghupathi2015), Murshad (Reference Murshad2020) and Yan et al. (Reference Yan, Shi and Yang2018) report positive associations between digital technologies and energy efficiency. Hu et al. (Reference Hu, Chen and Yang2022) further support these conclusions, noting a generally positive net effect of ICT on energy savings, economic growth and emissions reduction.
However, others caution against overly optimistic interpretations. Court and Sorrell (Reference Court and Sorrell2020) and Koot and Wijnbioven (Reference Koot and Wijnbioven2021) argue that energy gains from digital tools may be outweighed by increased energy consumption. Horner et al. (Reference Horner, Shehabi and Azevedo2016) emphasize the sensitivity of ICT's impact to contextual factors such as user behaviour and implementation/deployment specifics. Overall, while ICT can contribute significantly to energy efficiency, the magnitude and direction of its impact are highly contingent on deployment and behavioural variables.
Several studies emphasize sector- or technology-specific gains in energy efficiency. For instance, Putra et al. (Reference Putra, Rizky Pratama, Lazovik and Aiello2017) found that Bluetooth communication is approximately 30 per cent more energy-efficient than Wi-Fi for transmitting residents’ data. Amasawa et al. (Reference Amasawa, Ihara and Hanaki2017) reported that e-book reading has a lower global warming potential compared to paper books. Similarly, Weber et al. (Reference Weber, Koomey and Matthews2010) showed that digital music purchases are less energy- and carbon-intensive than traditional delivery methods. Zhou et al. (Reference Zhou, Fang, Li and Liu2018b) observed greater energy efficiency gains in China's devices sector compared to services, while Hao et al. (Reference Hao, Guo and Wu2022) found that ICT-driven efficiency improvements were more pronounced in Western China. These studies underscore the nuanced and context-specific nature of digital technologies' contributions to energy efficiency.
Yet, energy efficiency gains from digital tools may also generate rebound effects, offsetting initial savings. The literature addressing this phenomenon is limited. The most cited study, Galvin (Reference Galvin2015), using global data, found that ICT-related energy efficiency gains resulted in increased ICT and electronic device usage, leading to rebound effects ranging from 115 per cent to 161 per cent. This implies that energy consumption could rise by 15–61 per cent above the initial savings, underscoring the need to account for rebound effects in energy assessments of digital technologies.
Several factors contribute to the limited research on this topic. First is the methodological challenge of accurately measuring the true extent of the rebound effect (Kunkel and Tyfied, Reference Kunkel and Tyfied2021). Multiple approaches have been proposed, including environmental and economic methods, yet none are considered definitive. Environmental methods, such as life cycle assessment (LCA) techniques, are often criticized for being inadequate to evaluate systemic and multi-service applications. Economic techniques – experimental and non-experimental – struggle to capture time-dependent or long-term rebound effects, which can bias estimates (Rivera et al., Reference Rivera, Håkansson, Svenfelt and Finnveden2015).
Economic models face further limitations when rebound effects are uncertain. Scenario-based approaches offer an alternative, but only estimate potential rather than actual rebound outcomes. Hertwich (Reference Hertwich2005) critiques this limitation and calls for refinement. Börjeson et al. (Reference Börjeson, Höjer, Dreborg, Ekvall and Finnveden2006) propose combining scenario-based approaches with social practice approaches to better capture second-order rebound effects, especially in the context of digital technologies.
Even sophisticated global models such as computable general equilibrium frameworks have failed to account comprehensively for rebound effects (Thomas and Azevedo, Reference Thomas and Azevedo2013a, Reference Thomas and Azevedo2013b). Addressing rebound effects effectively may require a systems-dynamics approach that integrates and builds upon the strengths of existing methodological frameworks.
7.2.4. Methodological approaches – digitalization and energy demand nexus
The literature on the direct and indirect effects of digitalization has primarily centred on two key questions. The first concerns estimating the energy embodied in digital technologies, with LCA being the most commonly used method. LCA relies on life cycle inventory data to assess environmental and energy impacts across the production, use and disposal phases of digital tools. However, the bottom-up nature of LCA requires extensive and detailed data. The method is also often based on generic rather than specific assumptions, which limits its effectiveness and applicability, particularly when applied to sector-specific analyses or in contexts where data availability is limited. While LCA is useful for estimating energy savings from substituting one digital technology for another, it cannot account for rebound effects, potentially resulting in an underestimation of net energy impacts.
The second major area of inquiry focuses on how energy systems respond to digital technology adoption. In this domain, studies have primarily employed two methodological approaches: observational and experimental. Over 90 per cent of the reviewed literature relies on observational techniques, such as regressions and forecasting models. These approaches offer high external validity, allowing for generalization across contexts. However, they typically lack internal validity, making it difficult to establish causal relationships and raising the risk of biased or spurious correlations.
Only a limited number of studies employ experimental methods, which are better suited to establishing causality due to their stronger internal validity. Nevertheless, the findings from these studies are often mixed or inconclusive. For example, experimental research conducted in Morocco (Putra et al., Reference Putra, Rizky Pratama, Lazovik and Aiello2017; Rochd et al., Reference Rochd, Benazzouz, Abdelmoula, Raihani, Ghennioui, Naimi and Ikken2021), Bosnia and Herzegovina (Hadzovic et al., Reference Hadzovic, Seremet, Mrdovic and Causevic2020) and Nigeria (Bento, Reference Bento2016; Popoola et al., Reference Popoola, Atayero, Okanlawon, Omopariola and Takpor2018) has reported varying effects of digital technologies on energy consumption and efficiency. These inconsistencies in the results underscore the importance of methodological rigor and contextual sensitivity. While experimental methods are generally recommended for causal inference, the choice between observational and experimental methods should align with the researcher's goal – whether to establish causality or association.
7.2.5. Geographical distribution of studies and type of digital infrastructure examined
The distribution of studies on the energy impacts of digital technologies in the Global South is uneven, with a pronounced concentration in Asia and relatively limited research on Africa, Central or South America (see figure A4 in the online appendix). This regional imbalance may reflect difficulties in accessing reliable data on digital technologies in the energy sectors of underrepresented regions.
The literature covers a broad spectrum of digital technologies, including digital infrastructure, access, usage and digital skills or knowledge. Some studies also employ monetary-based indicators such as ICT trade and investment. The choice of technology examined often depends on data availability and the study's analytical focus. While some studies investigate specific digital technologies, others adopt a more holistic view of the digital ecosystem. Notably, there is a marked emphasis on digital infrastructure – particularly ICT, hardware and software – with internet access, mobile and landline phone ownership, and computer usage being the most commonly used proxies (see figure A5 in the online appendix). Approximately 60 per cent of reviewed studies use such proxies to assess the influence of digital technologies on energy systems.
Despite the growing interest in this area, the evidence on the impact of digital technologies on energy systems in the Global South remains inconclusive. Results vary widely depending on contextual factors, methodological approaches and model specifications. These inconsistencies suggest that no single digital infrastructure has been universally effective in driving the LCET. Moreover, the adoption of low-carbon digital technologies may entail trade-offs, underscoring the complexity of aligning digitalization with sustainability objectives.
The review also highlights notable gaps in the literature regarding the energy impacts of specific digital technologies in the Global South. Few studies address advanced technologies such as AI, teleworking, SDN, digital smart meters, sensors and automated processes. For instance, Gray et al. (Reference Gray, Ayre, Hinton and Campbell2020) found that Home Automation Management Systems can reduce household energy use, while Hadzovic et al. (Reference Hadzovic, Seremet, Mrdovic and Causevic2020) showed that switching to SDN improves energy efficiency. Other research, including Hook et al. (Reference Hook, Sovacool and Sorrell2020), Amasawa et al. (Reference Amasawa, Ihara and Hanaki2017) and Court and Sorrell (Reference Court and Sorrell2020), underscores the energy-saving potential of e-materialization and teleworking. However, studies examining the energy implications of digital platforms such as Uber, Airbnb, Netflix and FinTech in the Global South are notably absent.
7.2.6. Social inclusion, welfare and energy-related effects of digitalization
Social inclusion seeks to enhance the capabilities, opportunities and dignity of society's most disadvantaged members (World Bank, 2019), thereby improving quality of life and fostering equity and social cohesion (Levitas et al., Reference Levitas, Pantazis, Fahmy, Gordon, Ehrr and Patsios2007). Energy access plays a critical role in supporting economic development and inclusive growth. However, in the Global South, access to energy services remains highly unequal. Women, girls and rural populations are disproportionately underserved, exacerbating social exclusion and undermining development goals: quality of life and societal equity and cohesion (UN Women and UNIDO, 2023).
Addressing energy-related social exclusion requires targeted efforts to close service delivery gaps across socio-economic groups. Donor- and research-led interventions in rural areas – such as off-grid electricity solutions, on-grid electricity projects and clean cooking technologies (Williams et al., Reference Williams, Jaramillo, Taneja and Ustin2015; Mandelli et al., Reference Mandelli, Barbieri, Mereu and Colombo2016; Peters et al., Reference Peters, Sievert and Toman2019) – have sought to improve access. Yet, evidence on the effectiveness of these interventions in promoting social inclusion and development outcomes remains inconclusive and contested (see Jeuland et al., Reference Jeuland, Fetter, Li, Pattanayak, Usmani, Bluffstone, Chávez, Girardeau, Hassen, Jagger, Jaime, Karumba, Köhlin, Lenz, Litzow, Masatsugu, Naranjo, Peters, Qin, Ruhinduka, Serrano-Medrano, Sievert, Sills and Toman2021).
Technological innovation is increasingly viewed as a pathway to promoting social inclusion by offering leapfrogging opportunities and empowering marginalized groups – including women, girls, youth, the elderly and persons with disabilities (Hayes et al., Reference Hayes, Gray and Edwards2008; Ashraf et al., Reference Ashraf, Hassan, Lewis, Hassan and Ray2017). These innovations aim to enhance access to public, economic and social services for groups that have traditionally been excluded. In particular, digital technologies offer promising avenues to extend energy services to underserved populations in remote or marginalized areas (World Bank, 2019).
This underscores the potential of digital technologies – closely tied to both energy systems and carbon footprints – to reduce disparities in energy access and enhance social inclusion. Tools such as AI, IoT and energy demand management systems can simultaneously strengthen energy security and reduce emissions, contributing to broader social, environmental and economic integration. However, there remains a significant empirical gap in understanding the role of digital technologies in promoting equitable energy access and social inclusion, particularly in the Global South. Bridging this gap requires a deeper examination of the digital divide, including disparities in access, skills and infrastructure.
The literature also lacks robust analysis on how digitalization affects household welfare through energy-related pathways. While previous research has attempted to link energy access with development outcomes, the evidence remains fragmented and inconclusive (see Jeuland et al., Reference Jeuland, Fetter, Li, Pattanayak, Usmani, Bluffstone, Chávez, Girardeau, Hassen, Jagger, Jaime, Karumba, Köhlin, Lenz, Litzow, Masatsugu, Naranjo, Peters, Qin, Ruhinduka, Serrano-Medrano, Sievert, Sills and Toman2021). Similarly, studies on the energy-related impacts of digitalization have yet to reach consensus. Despite these uncertainties, the conceptual links between energy access, digitalization and welfare outcomes suggest that digital technologies could mediate the relationship between energy service delivery and welfare indicators. However, to date, no empirical studies have explicitly tested this hypothesis in the context of the Global South.
8. Conclusion and direction of future research
This study provides a comprehensive synthesis of the literature on the role of digital technologies in advancing LCET in the Global South. It evaluates current evidence, identifies critical gaps and highlights priority areas for actionable future research necessary to support the transition to sustainable low-carbon energy systems. The key conclusions from the study include the following:
Digital technologies contribute to low-carbon transitions by enhancing energy demand management, improving operational efficiency and supporting carbon neutrality. However, the empirical evidence on their impact – especially in the Global South – remains limited and fragmented.
Most existing studies focus on demand-side effects, with significantly less attention paid to supply-side dynamics. The latter often lack robust, quantitative methodologies and tend to be descriptive in nature.
A large share of demand-side studies assesses only direct effects of digitalization, often using observational rather than experimental or quasi-experimental designs. This limits causal inference and undermines internal validity. Indirect, secondary or system-level effects remain underexplored.
There is a pressing need for more holistic methods – such as system dynamics modelling – to capture complex interrelationships within energy systems. However, the reliance on strong and sometimes unrealistic assumptions limits the practical applications of such models.
The effect of digital technologies varies significantly by location, technology, region and sector. Yet, research on critical energy-intensive sectors – transportation, buildings and industry – is still underdeveloped. Moreover, the impact of specific digital tools like AI, blockchain, digital sensors and GPS in the Global South remains largely underexplored. Since different technologies may produce distinct energy-related effects, comparative studies across digital tools are essential.
Studies that examine how the energy-related effects of digitalization affect social inclusion, especially in terms of gender equity and overall welfare, are lacking. Additionally, the interaction between digital policies (digital taxation), digital access and inclusion, and cultural norms further complicates the landscape and is insufficiently addressed in the literature.
To advance the understanding of the role of digitalization in LCETs in the Global South, the following actionable research questions are proposed:
1. What drives the use (adoption) of digital technologies in support of LCET, sustainable natural resources management, consumption efficiency and substitution for energy-intensive materials?
2. Which digital technologies are most effective in enabling LCET across key sectors, and what factors influence their adoption?
3. How are digital technologies transforming within and outside industrial processes to support LCET?
4. How does digital exclusion reinforce disparities in energy service delivery, and what strategies can promote digital inclusion?
5. What is the relationship between digital taxation, social digital exclusion and energy access gaps? How do these dynamics affect welfare and progress towards LCET?
6. How can local culture and content be embedded in digital technologies to reduce social digital exclusion and improve energy service equity?
7. What is the scale of digital rebound effects? What factors drive them, and how do they influence progress towards LCET?
8. In what ways does social digital inclusion contribute to more equitable energy access and broader social sustainability outcomes?
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1355770X25100132
Acknowledgements
The authors are grateful to two anonymous reviewers and the editor for the initial thoughtful comments, which helped improve the paper. This study is a part of the Environment for Development (EfD) report commissioned by the Sustainable Inclusive Economy Division of the International Development Research Centre (IDRC). The authors acknowledge financial support from IDRC-ICRD, Canada, for their work on this initiative.
Competing interests
The authors declare none.
Open access
