Dynamic Clustering of Renewable Energy Capacity: A Comparative Study of the EU-27 and 15 RCEP Countries

Abstract

This study investigated the performance and dynamics of progress for renewable energy resources (RES) using data from the International Renewable Energy Agency (IRENA). The main objective was to recognize and interpret patterns of dynamic clustering among countries based on the temporal evolution of total installed renewable energy (RE) capacity. The study covers 27 European Union (EU) states and 15 RCEP countries (including China, Japan, South Korea, Australia, and ASEAN members) over the period from 2015 to 2024. Unlike static cross-sectional classification, the proposed clustering approach captures longitudinal trajectories of total installed RES capacity. Moreover, 2020 was considered a structural reference point (the energy shock, COVID-19, and the intensification of climate policies), marking a clear acceleration in renewable energy expansion. This scale allows differentiation between energy systems that experienced a post-2020 acceleration in transformation and those that either maintained earlier growth trends or experienced deceleration. A hierarchical clustering using Ward’s scheme was applied, which identified three distinct dynamic clusters: (1) moderate growth with transformation constraints; (2) sustained expansion and accelerated structural development; and (3) stagnation reflecting minimal structural change. To further segregate states within clusters, the Compound Annual Growth Rate (CAGR) has been employed, presenting a dual classification: (1) installed capacity levels, and (2) reflecting the momentum of renewable energy growth. An Acceleration indicator has been constructed to capture shifts in development dynamics leading up to 2020 and beyond. By measuring changes in RES capacity growth from 2015 to 2019, and from 2020 to 2024, this index allows us to determine whether the RES transition accelerated after the energy shock and the ramp-up of climate policies. According to the findings, the patterns of RE development are largely defined by investment capacity, economic scale, and institutional maturity, with continental or regional belonging playing a secondary role.

1. Introduction

Renewable energy plays a crucial role in reducing greenhouse gas emissions. The development of equipment that generates energy from renewable sources and stores excess energy, as well as research aimed at improving its efficiency, is progressing rapidly, leading to the expansion of renewable energy infrastructure. The latest IRENA report [1], whose data are used in this analysis, presents renewable power generation capacity statistics for the past decade (2015–2024) in trilingual tables. Using the data, the authors of this paper conducted a clustering of RES capacity performance across 27 EU countries and 15 RCEP countries.

Such an analysis is important for several reasons. First, recent research suggests that the use of renewable energy from various sources as green infrastructure must be spatially and market-balanced [2]. In the European Union, there is an ongoing debate on capacity adequacy in electricity markets in the context of significant differences in RES performance, supported by long-term production subsidies (feed-in tariffs, etc.) [3]. Most renewable energy comes from variable sources such as wind and solar. The amount of energy obtained from wind and solar radiation also differs significantly depending on the geographical location of the generating unit. A direct requirement of the energy system is the ability to balance not only variable demand but also increasing fluctuations on the generation side. Spatial fluctuations can be balanced through transmission, whereas temporal balance can be achieved through energy storage. Consequently, systems will continue to evolve to ensure energy flexibility. The key components providing flexibility in energy systems include networks, storage, demand-side management (DSM), and even curtailment [4].

Second, there are differences in regional energy transition models. The EU has set ambitious climate targets (net zero by 2050), while the RCEP is expanding both renewable energy production capacity and continuing to rely on fossil fuels. The EU’s climate neutrality is aligned with the “Fit for 55” strategy [5] and REPowerEU [6]. The EU aims to achieve at least 42.5% (with a target of 45%) of renewable energy in final energy consumption by 2030. RCEP countries have a free trade agreement, so their climate goals differ somewhat. China, the largest participant in RCEP, aims to peak its emissions before 2030 and achieve carbon neutrality by 2060. In achieving these goals, China prioritizes energy security and economic development. Chinese strategy is implemented through a comprehensive policy framework known as “1 + N” [7,8]. There is a need for strategic dialogue between the EU and ASEAN (part of RCEP) regarding the harmonization of standards [9], which would enable more sustainable trade and contribute to a more unified global energy market. This direction may be considered the third rationale for our analysis, as it provides answers regarding the current clusters of RES capacity performance in EU and RCEP countries. It should also be emphasized that the EU is a pioneer in terms of regulatory frameworks and the pace of decarbonization, while RCEP is the main global producer and executor of RES investments [10]. The industries of many RCEP countries are important for the development of the EU, and the governments of RCEP countries are key partners in the global energy transformation.

Significant progress has already been made in renewable energy. By 2025, wind and solar energy will have become the main energy sources in the European Union. Renewable energy will overtake fossil fuels, meaning that green energy will begin to dominate the energy mix (for the first time in history, wind and solar energy generated more electricity in the EU (30%) than fossil fuels (29%) [11]. Global cumulative photovoltaic capacity increased significantly to over 2.2 TW at the end of 2024, compared to 1.6 TW in 2023, with more than 600 GW of new PV systems commissioned. China’s annual capacity increased to 357.3 GW, which accounted for almost 60% of the new global capacity, or more than 1 TW of cumulative capacity. Notably, China accounted for nearly half of global photovoltaic capacity at the end of 2024. The rest of the world represented slightly more than 40% of new installations, which also increased significantly in 2024, adding 244.6 GW and reaching 1198.0 GW of installed capacity. Europe recorded strong growth, installing 71.4 GW (of which 62.6 GW in the EU), led by Germany (16.7 GW) and Spain (7.5 GW). In the Americas, both major markets expanded—the United States continued strong growth, adding 47.1 GW (224.1 GW total), while Brazil increased by 14.3 GW, raising its cumulative capacity to 52.1 GW. India recorded a growth year, reaching 31.9 GW, mainly in centralized systems. Pakistan recorded a high number of installations, reaching 17 GW, placing it fourth globally in terms of annual capacity; other Asia-Pacific markets experienced a slowdown (Australia—4.0 GW, Japan—5.5 GW) [12].

In light of the mentioned conditions (identified research rationales), we undertook a cluster analysis in order to achieve the research objective, which is the identification and interpretation of dynamic clustering patterns of countries based on the temporal evolution of total installed renewable energy capacity (Total RES) in the years 2015–2024, considering the structural breakpoint in 2020.

To accomplish this objective, the following research questions were formulated:

• RQ1: Do EU-27 and RCEP countries form distinct, repeatable dynamic patterns of RES capacity development in the years 2015–2024?
• RQ2: Is the dynamic of the energy transition determined by regional affiliation (EU vs. RCEP), or rather by the scale and systemic capacity of the economy?
• RQ3: Did the year 2020 constitute an acceleration point in the energy transition in the analyzed countries?

Several contributions have been made to the current body of knowledge regarding renewable energy clustering. Firstly, while other previous research mainly makes use of a static classification method, the current paper makes use of dynamic clustering based on the longitudinal trajectories of installed renewable energy capacity and, hence, can capture the structurally differentiated development path within the same region. Secondly, through the introduction of the concept of Acceleration Index, this paper captures the change in growth dynamic of renewable energy before and after the structural break point in 2020, thereby adding an additional temporal dimension to the traditional clustering methods. Thirdly, the paper provides a three-tiered analysis through the combination of dynamic clustering with growth rate decomposition (CAGR) and acceleration. Finally, rather than making regional comparison alone, the paper illustrates that economic size, investment capability, and systemic maturity are more significant than geographical proximity in influencing structural development patterns, therefore enriching the understanding of the cluster results. The clustering results are subsequently interpreted using socio-technical transition theory [13,14] and ecological modernization theory [15,16].

2. Background of Analysis

2.1. Statistical Background

Renewable energy capacity is measured in megawatts (MWs) or gigawatts (GWs). Global renewable energy generation capacity reached 4448 GW in 2024, with renewable energy accounting for 46% of total global installed capacity. IRENA statistics show the continued progress of the global energy transition (Figure 1).

The rapid growth of renewable energy is a result of the dynamic growth of solar and wind energy. The total installed capacity of renewable energy is planned to triple by 2030 to achieve climate goals [1]. During the United Nations Climate Change Conference COP28 (30 November–12 December 2023), participants in the meeting in Dubai stated that they aim to triple the capacity of renewable energy sources and double the rate of energy efficiency improvement by 2030 [17]. The International Energy Agency (IEA) predicts that by 2028, renewable energy sources will generate over 42% of global electricity production, with wind, solar, and photovoltaic energy dominating other renewable sources [18]. According to an IEA report [18], the global capacity of renewable energy sources could increase by two and a half times by 2030. This remains insufficient to achieve the COP28 objective of tripling renewable capacity, although governments possess multiple instruments to accelerate RES deployment, including dedicated programs, digital platforms, and financial mechanisms.

In a subsequent report [19], the IEA indicates that global renewable growth is likely to exceed existing government targets for 2030. In the IEA Scenario, around 5500 gigawatts (GW) of new renewable capacity will be commissioned by 2030. Solar photovoltaics and wind power combined are projected to account for 95% of the total renewable capacity growth by 2030 [19]. Solar power alone has already contributed to more than three-quarters of the renewable energy growth, achieving a record annual growth of 452 GW, while wind power added 113 GW [1]. In light of these aggregate statistics, both global and regional RES performance constitute a significant field of inquiry, particularly given the persistent asymmetries in the energy transition identified by IRENA [1]. China, the United States (not examined in the present study, given its established leadership position in its region), and the European Union accounted for 489 GW—83.6% of all new renewable capacity installed in 2024—whereas Africa contributed only 4.2 GW, or 0.7% [1]. Figure 2 presents a comparative summary of RES performance for RCEP countries (aggregate total) and the total for the European Union.

China is a world leader in renewable energy deployment. According to the International Energy Agency’s (IEA) forecasts [19], this country (or region) is expected to account for 60% of the global renewable energy capacity growth by 2030. Currently, one in two megawatts of the world’s total installed renewable energy capacity is expected to come from China, following the achievement of its 1200 GW wind and solar target six years ahead of schedule [19]. China’s dominance in the RES market derives from cost competitiveness and sustained policy support. The Chinese government supports investment in both industrial-scale and distributed renewable technologies. China participates in global supply chains for raw materials, including rare earth elements (REEs). Furthermore, China is the largest producer of photovoltaic (PV) energy (80% of key photovoltaic components are produced in China). Moreover, over 30% of cumulative global manufacturing capacity is located there, and China remains the world’s largest exporter of photovoltaic panels, with production costs below USD 0.24 per watt [20].

It is also important to note that China has made significant investments in Malaysia and Vietnam—both RCEP countries—contributing to their emergence as major exporters of photovoltaic products [21]. The Chinese carbon neutrality strategy combines large-scale grid investments [22] and accelerated RES market development, while coal-based generation continues to play a dominant role in the national energy mix. Coal, oil, and natural gas remain primary energy sources in China, particularly within the power generation sector [23].

In Japan, renewable energy is also growing, with policy goals to achieve a 36–38% share of the national energy mix by 2030 and 50% by 2040. Such shares represent a market value exceeding USD 40 billion. Solar energy dominates the Japanese renewable energy market, accounting for approximately 11.4% of total electricity production. In its energy policy, the Japanese government also emphasizes the development of offshore wind energy. This expansion is driven by governmental decarbonization objectives, although it faces structural constraints related to grid capacity limitations, restricted land availability, and high operational costs [24].

The future trajectory of Japan’s renewable energy market appears structurally robust, supported by technological progress and increasing consumer demand for sustainable energy solutions. Japan is expected to be a leader in the energy transition in the Asia-Pacific Region [24]. Already, the share of renewable energy in Japan has increased from about 15% in 2016 to 22% by 2021, and to 26.7% in 2024 [25]. Variable renewable energy (VRE), including solar and wind, reached 12.6% in 2024 [25]. During the same period, the share of electricity generated from fossil fuels fell to 65.1% [25].

At the same time (2024), some European Union countries exceeded 50% of their annual electricity production from renewable energy sources. The EU-27 average in 2024 was 47.5% [25]. Denmark is the leader in the renewable energy market (69%). Portugal, Germany, and Spain are close behind, with renewable energy shares exceeding 40%. The EU-27 average for VRE stands at 28.6%. Parallel to the diffusion of solar energy, stationary battery storage systems are increasingly being installed across different regions. This trend is particularly advanced in China, the United States, and within the EU—especially in Germany and Italy—all characterized by high solar penetration. In Japan, more than 10 GWh of battery capacity has already been installed, primarily on the demand side, especially in residential applications. Globally, total installed battery capacity is approaching levels comparable to pumped-storage hydropower plants [25].

Another example of a successful transition is Australia, where the federal government has set a new target to reduce emissions by 62–70% by 2035 compared to 2005 levels. In recent years, the country has doubled its share of renewable energy sources from 17% to 40%. Wind and solar technologies have become more cost-competitive and accessible [26].

In South Korea, a long-term “green growth” strategy was initiated in 2008 to promote economic development through low-carbon technologies and clean energy deployment. Energy production encompasses fossil fuels extracted and processed for electricity generation and fuel use, nuclear energy generated through fission, and renewable sources such as hydropower, wind, and solar. Bioenergy—including both modern and traditional sources, such as municipal waste combustion—also constitutes a significant domestic energy component in many economies [27].

A critical dimension of energy system transformation concerns oil refining processes that convert crude oil into petroleum products used in transport sectors, including automotive, maritime, and aviation fuels [27]. In terms of the total volume of refined petroleum products (data from 2023), the leading countries in the Asia-Pacific are China, India, Japan, South Korea, and Thailand [27]. Conventional energy infrastructure is still present in regional energy systems. Therefore, it seems reasonable to pursue hybrid transformation paths rather than linear decarbonization trajectories.

At the strategic level, the European Green Deal provides the overarching objective of achieving climate neutrality by 2050. The “Fit for 55” package is a legislative reform of EU climate policy. At the sectoral level, the Renewable Energy Directive (RED III) is an important document [28]. This document indicates the direction of changes in the energy mix. At the market level, emission allowances (ETS) are an important instrument. The EU pricing mechanism considers environmental externalities and encourages investment in low-emission technologies in the economy. In turn, REPowerEU [6] and the Just Transition Mechanism are essential for achieving energy security and socio-economic restructuring.

Table 1. Key climate and energy policies in European Union [5,6,29,30,31,32,33].

Program/Policy	Year	Instrument Type	Core Objectives	Time Horizon
European Green Deal	2019	Overarching strategic framework	Climate neutrality; systemic economic transformation; large-scale RES deployment; circular economy; just transition	2050
Fit for 55	2021	Legislative package	At least 55% GHG reduction (vs. 1990); ETS reform; Carbon Border Adjustment Mechanism (CBAM); transport decarbonization	2030
REPowerEU	2022	Energy security and acceleration plan	Reduce dependence on Russian fossil fuels; accelerate renewable deployment; improve energy efficiency	2030
Renewable Energy Directive (RED III)	2023	Binding sectoral directive	Minimum 42.5% renewable energy share in final energy consumption (45% indicative target)	2030
EU Emissions Trading System (EU ETS)	2005/revised 2023	Market-based carbon pricing mechanism	Progressive emission caps in power and industry; extension to buildings and transport; strengthening carbon price signal	2030+
Just Transition Mechanism	2020	Financial and cohesion instrument	Support coal-dependent and carbon-intensive regions; socio-economic restructuring	2021–2027

depicts the dimensions of the EU’s highly structured and multi-layered architecture, aiming to achieve the objective of climate neutrality through regulatory coherence and supranational authority.

Table 2. Key climate and energy policies in RCEP region (selected national frameworks) [7,8,34,35].

Program/Policy	Country/Region	Instrument Type	Core Objectives	Time Horizon
Regional Comprehensive Economic Partnership	15 Asia-Pacific countries	Trade and economic integration agreement	Trade liberalization; regional value chains (including clean energy technologies); industrial integration	Ongoing (since 2020)
Carbon Neutrality 2060 (“1 + N” framework)	China	National climate strategy	Emissions peak before 2030; carbon neutrality before 2060; structural energy transition	2030/2060
14th Five-Year Plan	China	National development plan	Expansion of renewable capacity; grid modernization; energy security; industrial upgrading	2021–2025
Strategic Energy Plan	Japan	National energy strategy	36–38% renewable electricity share; offshore wind expansion; hydrogen strategy	2030
Green Growth Strategy	South Korea	Long-term green development strategy	Low-carbon technological leadership; green industry development; carbon neutrality	2050
National Energy Transition Plans	ASEAN members (e.g., Indonesia, Vietnam, Malaysia, Thailand)	National policy frameworks	Gradual RES expansion; energy security; diversification of the energy mix	2030–2060

also shows the structurally decentralized and nationally driven nature of the energy transition process. Unlike the case in the European Union, RCEP does not operate within the context of a supranational authority for the management and regulation of the climate. Instead, renewable energy and decarbonization goals and targets are set at the national level. The RCEP is simply a regional trade and economic integration agreement that facilitates industrial cooperation and supply chain integration. However, large economies such as China, Japan, and South Korea have set and implemented long-term plans for the achievement of the goal of carbon neutrality. These economies combine renewable energy and decarbonization goals with industrial goals and plans. For instance, China operates within the context of the “1 + N” plan and successive five-year plans. ASEAN member states, on the other hand, tend to adopt gradual and nationally driven approaches to the transition process. The approaches often combine renewable energy and decarbonization goals with fossil fuel-based goals and plans.

There are two different paradigms of the governance of the energy transition, which differ in their structures. While the European Union has a regulation-based model of the governance of the energy transition, which relies on collective regulation, legislation, and a carbon pricing strategy, the RCEP model of the governance of the energy transition lacks a centralized climate policy, which means that the strategy of the energy transition varies among the different countries of the region. Although the countries of the region, such as China, Japan, and South Korea, have set carbon neutrality targets, the RCEP model of the governance of the energy transition lacks a unified approach to the issue of the regulation of the energy transition, which means that the strategy of the energy transition varies among the different countries of the region. While the European Union model of the governance of the energy transition relies on a unified approach to the issue of the regulation of the energy transition, the RCEP model of the governance of the energy transition relies on a production-oriented model of the governance of the energy transition, which means that the capacity of the production industry, the size of the economy, and the investment strategy of the government of the country drive the expansion of renewable energy in the region. Table 1 presents the most important EU policies, and Table 2 presents the RCEP.

Table 3. Comparative governance and energy transition model: EU vs. RCEP [29,30,31,32,33,34,35,36,37,38,39].

Dimension	European Union (EU)	RCEP Region
Institutional Structure	Supranational political and regulatory union with binding legislation	Intergovernmental trade agreement without supranational climate authority
Nature of Integration	Policy-driven and regulatory integration	Trade-driven and economic integration
Climate Target Structure	Legally binding collective targets for all Member States	Nationally determined targets; heterogeneous ambition levels
Carbon Neutrality Objective	2050 (EU-wide commitment)	2050 (Japan, South Korea); 2060 (China); varied or undefined in several ASEAN states
Emission Reduction Target (2030)	At least −55% vs. 1990 levels	Country-specific; no unified regional reduction target
Renewable Energy Target	42.5% binding target (45% indicative) by 2030	Country-specific renewable targets; no common regional RES quota
Carbon Pricing Mechanism	EU-wide Emissions Trading System (ETS); expansion to new sectors	National ETSs (e.g., China); no regional carbon market
Energy Security Strategy	Diversification, electrification, demand reduction, cross-border grids	Emphasis on supply security; coexistence of fossil fuels and renewables
Transformation Model	Regulatory-institutional and market-based decarbonization	Industrial production drives expansion of clean technologies
Role in Global Value Chains	Regulatory standard-setter; climate governance leader	Manufacturing hub for renewable technologies (especially solar PV supply chains)
Post-2020 Acceleration Logic	Reinforcement through crisis response (energy security + climate)	Acceleration primarily in large economies; uneven across smaller systems
Structural Determinant of Transition	Institutional maturity, policy coherence, internal market scale	Economic scale, industrial capacity, state-led investment strategies

compares the EU and RCEP energy transition models.

2.2. Literature Background

The pace of energy transformation has accelerated in recent years. The global renewable energy (RES) market is shaped by political and social processes, as well as economic factors. Governments play a central role in implementing and financially supporting renewable energy deployment. Political and economic determinants significantly influence renewable energy policy and its implementation, and existing research has confirmed their importance in attracting investment [40]. Beyond political variables, an important determinant is societal awareness of the ongoing transition. When society possesses such awareness, it more readily adopts RE innovations [41]. The impact of many macroeconomic (economic growth) and legislative (environmental regulations), market (raw material prices, energy prices) and technological (technological progress) factors on the use of renewable energy is often uncertain and varies on a global, regional and national scale.

The impact of economic, social, political, and institutional factors on renewable energy development has generated extensive scientific debate concerning the main drivers and barriers to RE expansion [42,43,44,45,46]. Many researchers regard economic growth as a driving force of renewable energy development [47,48,49], accompanied by technological advancement. Technologies and infrastructure constitute the backbone of renewable energy development, enabling energy transition, improving efficiency, and ensuring security of supply under conditions of variable generation [50,51]. While technologies (e.g., more efficient panels and turbines) are responsible for producing clean energy, infrastructure (e.g., grids and storage systems) ensures its effective delivery and balancing. These two pillars—energy technology and infrastructure—are supported by digitalization and artificial intelligence: the Internet of Things (IoT), artificial intelligence (AI), and data analytics tools enable intelligent real-time monitoring and management of energy production. These solutions form part of Industry 4.0 and Industry 5.0 paradigms [52,53]. Digitalization facilitates the development of smart grids—modern networks that use digital technologies to detect and respond to local changes in energy consumption, increasing system flexibility in the presence of distributed sources such as prosumer photovoltaics—and other smart energy management techniques that integrate sectors across the energy market (electricity generation, heating, cooling, manufacturing, residential construction, and maritime transport). It also supports the identification of feasible and accessible technologies for transitioning toward sustainable and renewable energy solutions [54]. A major challenge for global RES efficiency growth lies in integrating distributed energy sources through advanced market design mechanisms, such as energy trading and flexible demand management, in order to accommodate high levels of energy source penetration [55], alongside strong governmental support for emerging RE markets [56,57] and the promotion of local initiatives, such as energy cooperatives [58].

Various indicators are used to measure renewable energy performance. These include the share of renewable energy in total energy supply, total renewable energy production (REP), the share of renewable energy in total electricity generation (RES), renewable energy consumption (REC), the share of renewable energy consumption in total energy consumption, and overall RES efficiency and the efficiency of specific renewable energy categories. These indicators, across different spatial (geographical) arrangements, are compiled in reports by agencies specializing in RE market analysis, such as the International Energy Agency (IEA), the International Renewable Energy Agency (IRENA), Ember, research institutes such as the Energy Institute, and statistical offices including Eurostat.

Static datasets often constitute the empirical basis for scientific research. Authors of academic papers present renewable energy development models incorporating country- or region-specific determinants and organize results into clusters or develop scenarios. RE development is assessed across various dimensions and specifications (e.g., supply, consumption, or installed capacity of RES measured in absolute terms, per capita, or as a share of total energy or electricity). Bourcet [59], conducting a literature review on quantitative determinants of renewable energy development at the national level, demonstrated that there is limited consensus regarding the influence of economic, environmental, and energy-related factors, which are most frequently examined. Other key determinants considered include regulatory, political, and demographic factors. Researchers apply diverse measures of RE implementation and performance evaluation [60,61,62]. Differences across studies concern the choice of renewable energy utilization indicators, the determinants included in subsequent analyses, and the adopted methodologies. Among studies producing clusters, variations appear in the scope of comparisons (

Table 4. Research on renewable energy source (RES) clustering [59,60,61,62,63,64].

Authors	Year	Spatial Scope	RES Indicator(s)	Methodological Approach	Main Findings
Bourcet [59]	2020	OECD countries	Share of RES in total energy supply	Literature review + comparative grouping	No clear consensus on determinants; strong role of political and regulatory variables
Apergis & Payne [61]	2014	80 countries	Renewable energy consumption	Panel data models + causality tests	Heterogeneous RES–growth relationship across regions
Sadorsky [62]	2009	G7 countries	Renewable energy consumption per capita	Panel regression analysis	Income level and CO2 emissions significantly affect RES development
Marques et al. [63]	2010	EU-15	Share of RES in electricity generation	K-means clustering	EU countries form distinct groups reflecting different transition trajectories
Menegaki [64]	2011	European countries	Renewable energy production	Panel econometric analysis	Weak and context-dependent relationship between economic growth and RES

), yet the underlying idea remains common, as clustering helps identify areas of development [60]. Table 4 shows the research on RES. Earlier studies predominantly relied on panel regression frameworks or relatively simple clustering procedures to examine the relationship between renewable energy consumption, economic growth, and emissions, frequently revealing heterogeneous and region-specific patterns. Subsequent research introduced more formal clustering techniques—such as k-means and hierarchical clustering—to identify groups of countries or regions with similar renewable energy development trajectories, thereby emphasizing the roles of economic scale, industrial capacity, and institutional maturity. More recent contributions extend this analytical approach by incorporating longitudinal trajectory analysis and synthetic growth indicators (e.g., Compound Annual Growth Rates and acceleration metrics), enabling scholars to distinguish between mature, incremental, and structurally constrained transformation regimes. Overall, the literature indicates that renewable energy dynamics are less determined by geographic location per se and are more strongly shaped by systemic capacity, investment potential, and policy continuity. Table 4 summarizes the literature review.

The review is supplemented by a list of examples of renewable energy programs implemented around the world in 2026 (

Table 5. Examples of renewable energy programs in selected countries from EU and RCEP in 2026.

Country	Key Accelerated Programs & Policies & Tools	Description	More on This & Additional Information
Belgium	Tax credits & tariff in grid use (Green, Amber, Red) & energy communities and others (Green Certificates, Green Loans (EIB), promoting EVs, regional programs for industry)	Belgium has increased tax credits for green investments: solar panels: 120% tax credit (compared to the standard 100%); heat pumps: 125% tax credit; electric vehicles: 120% tax credit. The energy supplier in this country offers three pricing tiers (green, amber, red) that reward consumers for using smart meters that adjust energy consumption to match grid demand and energy prices. In Belgium, as in other EU countries, regulations are in place that allow for active “energy communities,” enabling collective cost reduction, peer-to-peer sales, and energy sharing (e.g., in schools or among neighbors). In addition, the Belgian government has a streamlined permitting process for projects that meet environmental criteria. The European Investment Bank (EIB) offers low-interest loans through Belgian banks (Belfius, KBC, etc.) for renewable energy projects with a capacity of over 100 kW. Industry Decarbonization: regional programs, such as the Walloon Marshall Plan 4.0 and Flemish “Vlaanderen Circulair,” provide significant grants for industrial electrification and waste heat recovery. Electric vehicles (EVs): 100% electric company vehicles are 100% tax-deductible in 2026.	[65]
Chine	End of Feed-in Tariffs (FiTs & Green Electricity Certificate & “Quota Policy”	China has adopted a “grid parity” policy under which renewable energy competes directly with the prices of coal-fired power. China’s “Quota Policy” (2019) requires regions to meet renewable energy consumption targets (similar to renewable energy portfolio standards). It is currently supported by a thriving renewable energy certificate market that rewards renewable energy producers. In addition, the country is consistently modernizing its power grid: Investments in grid infrastructure, particularly in ultra-high-voltage (UHV) lines, are being made to transmit electricity from resource-rich regions in the west to energy-consuming centers in the east. The Chinese government actively supports private-sector involvement in energy infrastructure, allowing private companies to hold stakes in large-scale projects, although government-led projects remain the dominant form.	[66,67]
Finland	Green Transition Tax Credits & RENEWFM & others	The government is offering 20% tax credits (up to €150 million per company) for large-scale investments in renewable energy, energy storage, and industrial decarbonization. In addition, Business Finland awards grants for projects worth over EUR 30 million focused on industrial decarbonization and energy efficiency, and the Renewable Energy Financing Facility (RENEWFM) supports the development of photovoltaics.	[68]
France	EDF “Obligation d’Achat” (OA) & Feed-in Tariffs (FiTs) & Prime à l’Autoconsommation & Subsidies & Agri-PV Premium	This country has a feed-in tariff system (EDF) that is adjusted quarterly for solar systems with a capacity of less than 100 kWp. The FIT rates are set to encourage the integration of batteries and direct energy consumption. PV systems are also eligible for a capital subsidy (“bonus”) for self-consumption (systems up to 100 kWp). In addition, photovoltaic systems in agriculture covered by the OA scheme: a premium of 1.5–2.0 ct/kWh above standard rates until 2028. Private capital is actively involved in large-scale battery energy storage (e.g., Eiffel Investment Group) for battery energy storage systems (BESS). Public–private auctions (CRE) are also popular in France. Systems with a capacity exceeding 100 kWp typically participate in competitive bidding (CRE auctions) with a market premium, which encourages the development of larger and more efficient projects.	[69]
Germany	EEG 2026 & KfW Subsidies & Solarpaket 1 &) & energy communities and others	This country is transitioning from high feed-in tariffs (FiTs) to market-based premiums for large-scale projects; Germany continues to use KfW loans and grants to cover up to 30% of the costs of home energy storage and the expansion of photovoltaic systems. The Solarpaket I package further simplifies regulations regarding photovoltaic installations on balconies and reduces red tape. The government supports “energy communities,” enabling collective cost reductions, peer-to-peer sales, and the sharing of renewable energy.	[70,71]
Greece	Feed-in Premiums (FiPs) & Small Rooftop Solar Program	Feed-in Tariffs (FiTs): Larger solar power plants (>500 kW) and onshore wind farms (>3 MW) participate in competitive auctions to secure a “feed-in tariff” on top of the market price. Program for small rooftop photovoltaic installations: Starting in 2022, a 20-year guaranteed price of 0.087 EUR/kWh is offered for small rooftop photovoltaic installations (up to 6 kWp).	[72]
Italy	FER X/FER 2 & agrivoltaic incentives & energy communities & Energy Release 2.0 & “Bills Decree” 2026 (DL 21/2026)	The FER X/FER 2 decrees provide incentives for new renewable capacity (solar, wind, and agrivoltaic) through competitive tenders for bilateral CfD contracts, with tariffs paid by the Gestore dei Servizi Energetici (GSE). In addition, the Italian government has approved a program supporting agrivoltaic power plants. It includes capital grants (up to 40% of costs) and 20-year incentive tariffs. Italy also offers high premium tariffs (8–11 cents/kWh) for energy shared in RECs to encourage decentralized generation, especially in urban areas. Energy Liberation 2.0 (2026): Energy-intensive companies can purchase electricity from GSE at a fixed price (65 EUR/MWh) for 36 months in exchange for building new renewable capacity, which will repay this energy over 20 years. According to DL 21/2026 (February 2026), operators may benefit from a temporary reduction in the applicable guaranteed tariffs (Conto Energia) (a reduction of 15–30%) in exchange for extending incentive agreements aimed at managing grid saturation and promoting.	[73,74]
Japan	“GX 2040” & GX-ETS & others	The “GX 2040” Grant Program: Funding of up to 50% of investment costs for companies (including data centers) that use 100% decarbonized electricity and contribute to the development of regional economies. Regarding PV, Japan has amended its Energy Conservation Act; the law mandates the installation of solar panels on the roofs of new buildings and warehouses of a certain size starting in fiscal year 2026. In the area of decarbonization, Japan introduced an emissions trading system (GX-ETS) in April 2026. Japan is promoting the development of offshore wind projects in its exclusive economic zone (EEZ).	[75,76,77]
Malaysia	Dynamic Line Rating (DRL)	The real-time weather monitoring system used in this country increases the capacity of existing power lines by 10–50%, providing a cost-effective, smart grid solution for the energy sector.	[78]
Poland	“Mój Prąd Program” & energy communities and others (e.g., PPP)	This program provides substantial subsidies (up to 50% of costs) for residential solar installations, combined with a net metering system that operates at spot prices, thereby encouraging self-consumption. The program was announced on 23 July 2019. The “Mój Prąd Program” for 2024–2027 is currently underway. Moreover, the government supports “energy communities,” enabling collective cost reductions, peer-to-peer sales, and the sharing of renewable energy.	[79]
Portugal	Industrial Decarbonization & Green Tech: Ministerial Order No. 160/2024/1 & “Investment in Strategic Sectors” & Rooftop & Residential Solar Support & others (simplified licensing, Accelerated Implementation Zones (PSZAER))	Ministerial Regulation No. 160/2024/1 provides for non-repayable grants for the production of climate-related equipment, including photovoltaic (PV) systems, wind turbines, batteries, and green hydrogen technologies. The country also offers grants for strategic sectors: the “Investments in Strategic Sectors” program (up to 35% of eligible costs) for the large-scale production and storage of energy from renewable sources, with up to €350 million available per project in designated regions. In addition, the government supports the purchase of solar panels for households and is launching a new tender for light electric vehicles. Portugal uses simplified environmental impact assessment licenses—they are no longer mandatory for photovoltaic projects with a capacity of less than 50 MW. The government is creating “Accelerated Implementation Zones” (PSZAER) for green renewable energy projects. In addition, the government is focusing on reducing grid connection costs for renewable energy sources and increasing flexibility in access to grid capacity.	[80]
South Korea	“Sunlight Income Village” & RE100 Industrial Complexes & Tax Incentive & The 12th Basic Power Supply Plan (2026–2040)	The South Korean government is promoting a proactive auction-based system, with increased public and private funding, to expand solar energy (particularly rooftop and agrivoltaic systems) and offshore wind energy. The “Sunlight Income Village” program is being implemented in this country, and its main goal is to establish 2500 village-scale community solar systems (with capacities ranging from 300 kW to 1 MW) by 2030, with a target of 500 systems by 2026. In terms of cooperation with the private sector, the government is promoting industrial parks where most, or all, of the energy needs are met by renewable sources (RE100 Industrial Complexes). In South Korea, there are tax incentives for investments in low-carbon manufacturing and photovoltaic installations that meet specific carbon dioxide emission thresholds. Carbon-neutral industrialization. Additionally, initiatives to co-locate renewable energy generation with industrial centers are supported to improve efficiency. An important direction for changes in South Korea’s RES policy is the 12th Basic Power Supply Plan (2026–2040): phasing out old coal-fired power plants while balancing them with nuclear energy.	[81]
Spain	European NextGenerationEU Funds & Direct Subsidies & RENOVAL 2 Program & PERTE-EHRA Initiative & IRPF & REER and others	The European Program is the cornerstone of Spain’s solar energy expansion, providing direct subsidies for rooftop and utility-scale photovoltaic installations, with a particular focus on battery energy storage. Meanwhile, the €355 million RENOVAL 2 Program (launched in January 2026) supports the production of renewable technology components. The PERTE-EHRA initiative, meanwhile, aims to mobilize over €12.25 billion in public–private investments by 2026 for green hydrogen and electricity storage. The country also offers tax incentives, including: Royal Decree-Law 18/2022, which allows for accelerated depreciation of renewable energy installations for self-consumption; an income tax (IRPF) deduction of up to 40% of the value of photovoltaic installations in residential buildings that reduce non-renewable energy consumption; property tax (IBI) relief of 25–50% for a period of 3–5 years; the REER (pay-as-bid) auction system, ensuring long-term revenue stability through variable guaranteed premiums, often referred to as contracts for difference (CfDs).	[82]
Sweden	Green Technology Deduction (Grön teknik-avdraget) & Sweden’s Climate Social Plan and green projects	Green Technology Tax Credit (Grön teknik-avdraget): Starting in early 2026, this will remain the primary form of support for private individuals, providing a 20% tax credit on the costs of materials and labor associated with the installation of solar panels. This program also includes a 50% tax credit for the installation of home batteries and electric vehicle (EV) charging stations. Additionally, to accelerate the development of wind energy, the Swedish government provides municipalities with financial support equal to property tax revenue for approving new wind farms. The Swedish Energy Agency is offering grants for green projects focusing on grid reliability, bio-combined heat and power, and energy storage.	[83]
Thailand	Power Development Plan (PDP) 2026–2050 & FiT & Rooftop Solar Incentives & Royal Decree (No. 805) & The Utility Green Tariff (UGT) program & others	Thailand aims to achieve a share of over 50% clean energy in its energy mix by 2037 (solar, wind, and bioenergy). The government is supporting structural reforms to increase the private sector’s participation (feed-in tariffs (FiTs)) in the transition. A new royal decree (No. 805), effective 3 March 2026, allows homeowners to deduct up to a certain amount of the costs of installing rooftop solar panels. In addition, individual users can sell surplus electricity back to the grid (MEA/PEA). The Community Solar Program (A 1500 MW “Quick Big Win”) enables public utilities to procure energy from local community projects. The Utility Green Tariff (UGT) for businesses allows corporate buyers to purchase renewable energy, particularly from hydroelectric power plants, through the grid.	[84]
Vietnam	National Power Development Plan VIII (PDP8)	The country aims to have renewable energy account for approximately 47% of total installed capacity by 2030, with a particular focus on offshore wind farms and rooftop solar panels. The government supports renewable energy projects and offers incentives for such projects: a 10% corporate income tax (CIT) rate, a four-year tax holiday, and a 50% CIT reduction for nine years. Additionally, the government provides incentives for rooftop photovoltaic installations on key buildings.	[85]

). The programs cited (examples) focus on combining, scaling, and updating incentives to go beyond simply installing power generation capacity and effectively integrate it into the power grid. Key strategies include transitioning from guaranteed tariffs to competitive bidding, mobilizing vast amounts of private capital through green bonds, and streamlining permitting processes. The table provides examples of RES support programs and instruments. Due to the broad scope of these programs, not all EU and RCEP countries under review are included; only selected ones are listed. The criterion used to select countries for the table was regional location (Central Europe and other geographic regions; for RCEP, examples of large regions—such as China, South Korea, and Japan—and medium-sized—such as Malaysia, Vietnam, Thailand—were provided).

3. Materials and Methods

The purpose of dynamic clustering is to classify countries based on the temporal evolution of total renewable energy installed capacity (Total RES) from 2015 to 2024, as opposed to a snapshot from a single year. The multi-level perspective (MLP) was used for the purposes of this analysis. This perspective has proven to be a useful tool for analyzing socio-technical transitions toward sustainable development (the energy transition). This approach is used in research and is subject to constructive evaluation [13,14]; it is particularly useful in the theory of ecological modernization, as presented by scholars such as Mol and Spaargaren [15,16].

The input data are derived from two Total RES spreadsheets: 15 RCEP countries (including China, Japan, South Korea, Australia, and ASEAN members) and EU-27. The two spreadsheets are combined into one dataset with countries as observations and annual capacity as time-indexed variables from 2015 to 2024. Before running any analysis, the dataset is preprocessed by removing non-country rows such as descriptive text, units, and total rows, as well as coercing all yearly values into a numeric data type. Only countries with complete observations across all ten years are used.

To deal with the strong right skewness and scale differences between large and small energy systems, each capacity series for a given year was variance stabilized using the transformation: ln(1 + x). This helps to reduce the effect of outliers (e.g., large economies). Then, the variance-stabilized time series vectors were standardized (z-normalization) to ensure that distances in the clustering stage were based on relative temporal pattern differences, not absolute magnitude differences as seen in the original data, which span six to seven orders of capacity.

Hierarchical agglomerative clustering was then performed using Ward’s minimum variance method on the standardized trajectory vectors, with Euclidean distance as the dissimilarity measure. Ward’s criterion iteratively merges clusters to minimize the increase in within-cluster sum of squares, which is well-suited for identifying compact groups of countries with internally similar growth paths. The clustering structure was inspected using a dendrogram, and a three-cluster solution (k = 3) was selected as a parsimonious partition capturing the main structural breaks in the agglomeration process without over-fragmenting the sample. Cluster membership was ultimately determined by applying a cut to the dendrogram, yielding three distinct groups. The classification was then presented in the form of a country-level table, combining annual capacity trajectories with the assigned cluster labels. It was not only descriptive, but it also enabled a more structural interpretation—countries to be positioned within differentiated growth regimes, namely accelerating expansion, moderate but stable growth, and, in some cases, stagnation or outlier-like developmental paths. The boundaries between these regimes were not always sharp, though the clustering procedure imposed a certain analytical clarity.

To complement this typological perspective, the Compound Annual Growth Rate (CAGR) was employed as a quantitative measure of long-term dynamics in the total installed capacity of renewable energy sources (Total RES) over the period of 2015–2024. The indicator reflects the geometric mean rate of growth across the entire time horizon, which—importantly—reduces the influence of short-term volatility observed between individual years. In this sense, it captures a more underlying trend rather than episodic variation. CAGR was computed according to the standard formulation (Formula (1)), providing a synthetic measure of expansion that remains comparable across countries despite differences in initial capacity levels:

$$\mathrm{C}\mathrm{A}\mathrm{G}\mathrm{R}={\left({\displaystyle \frac{{\mathrm{X}}_{2024}}{{\mathrm{X}}_{2015}}}\right)}^{{\displaystyle \frac{1}{9}}}-1$$

(1)

where

• ${\mathrm{X}}_{2015}$—installed renewable energy capacity in 2015;
• ${\mathrm{X}}_{2024}$—installed renewable energy capacity in 2024;
• 9—number of years of growth in the analyzed period (2015–2024).

The use of the exponent 1/9 follows from the fact that the period includes nine annual increments between 2015 and 2024. CAGR is interpreted as a constant annual growth rate that would lead from the initial level to the final level under the assumption of geometric growth.

In the empirical analysis, the CAGR indicator was calculated separately for each country and subsequently used to compare dynamic clusters identified using Ward’s method based on the 2015–2024 trajectories. To characterize differences between clusters, the arithmetic mean of CAGR, the median, and the standard deviation were computed within each group. The mean reflects the overall growth intensity within a cluster, the median indicates the rate representative of a “typical” country, while the standard deviation allows for the assessment of internal heterogeneity in growth dynamics.

The choice of Ward’s hierarchical clustering method was informed by several theoretical factors. As opposed to methods like k-means, where one needs to know beforehand how many clusters they will generate, Ward’s method does not have this constraint, and the number of clusters can be determined through examination of the dendrogram. Also, Ward’s algorithm, based on minimizing the variance criterion, is especially appropriate for analyzing trajectory-type data since it produces tightly bound clusters. Hierarchical clustering is much less sensitive to initialization compared to k-means and is unlikely to get stuck in local optima. Lastly, model-based clustering procedures were not applied because the sample size was insufficient (42 observations) and the assumption about underlying distributions did not exist for capacity trajectories over time.

It should be emphasized that CAGR is a relative measure and is sensitive to the initial level. In countries with very low baseline capacity, even small absolute increases may generate high percentage values. For this reason, the interpretation of the indicator was conducted in parallel with the analysis of installed capacity levels and the structure of dynamic clusters. CAGR does not replace the analysis of complete time trajectories; rather, it provides a synthetic representation of them, enabling comparison of the intensity of the energy transition across countries and groups of countries.

Methodologically, the CAGR analysis plays a complementary role to dynamic clustering: clustering identifies similarity in the shape of growth paths, whereas CAGR measures the average speed of expansion over the analyzed period. Combining both approaches makes it possible to distinguish the analyzed systems.

To assess changes in the dynamics of renewable energy development before and after 2020, an Acceleration Index was applied. The construction of this indicator is based on the comparison of two sub-periods, 2015–2019 (the pre-2020 period) and 2020–2024 (the post-2020 period), which allows the potential effects of the energy shock, geopolitical changes, and the intensification of climate policies to be captured. The pre-2020 period (2015–2019) is defined to end immediately before the structural shock, while the post-2020 period (2020–2024) explicitly includes 2020 as the first year capturing the effects of the energy shock, the COVID-19 disruption, and the intensification of climate policy.

First, for each country, the Compound Annual Growth Rate (CAGR) was calculated for both sub-periods using the geometric growth Formulas (2) and (3):

$$\mathrm{C}\mathrm{A}\mathrm{G}{\mathrm{R}}_{2015-2019}={\left({\displaystyle \frac{{\mathrm{X}}_{2019}}{{\mathrm{X}}_{2015}}}\right)}^{{\displaystyle \frac{1}{4}}}-1$$

(2)

$$\mathrm{C}\mathrm{A}\mathrm{G}{\mathrm{R}}_{2020-2024}={\left({\displaystyle \frac{{\mathrm{X}}_{2024}}{{\mathrm{X}}_{2020}}}\right)}^{{\displaystyle \frac{1}{4}}}-1$$

(3)

where

• First, ${\mathrm{X}}_{\mathrm{t}}$ denotes the total installed renewable energy capacity in year t;
• Next, 4 denotes the number of annual increments in each sub-period.

The Acceleration Index (Equation (4)) was then defined as the difference between the growth rate in the second sub-period and that in the first sub-period.

$$\mathrm{A}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}=\mathrm{C}\mathrm{A}\mathrm{G}{\mathrm{R}}_{2020–2024}-\mathrm{C}\mathrm{A}\mathrm{G}{\mathrm{R}}_{2015–2019}$$

(4)

A positive value indicates an acceleration of development after 2020, a negative value indicates a slowdown, while a value close to zero suggests a stable growth rate across both periods.

Although the analysis identifies 2020 as the turning point in the structure, it needs to be stated that there are other external factors that can affect the dynamics of the transition to renewable energy as well. The geopolitical events, such as trade wars and disruptions in the supply chain, as well as advancements in technology, especially concerning energy storage and grid management, can be seen as the turning points. These aspects are not considered in the current study, which only focuses on capacity growth trends. Nevertheless, the effect of these aspects can be seen indirectly from the growth trend after 2020, especially when it comes to larger energy infrastructures.

The analysis was conducted individually for each country, and the results were subsequently aggregated at the level of dynamic clusters (k = 3) previously identified using Ward’s hierarchical clustering method based on the full 2015–2024 trajectories. At the cluster level, the mean value of the Acceleration Index was calculated, allowing for comparison of energy system response regimes to post-2020 changes.

A cluster validity check based on silhouette scores yielded values of 0.34 (k = 2), 0.36 (k = 3), and 0.33 (k = 4). The elbow method shows a strong decrease in within-cluster variance up to k = 3, after which the marginal improvement becomes substantially smaller. This indicates that k = 3 represents a reasonable compromise between model simplicity and the ability to capture heterogeneity in trajectory patterns.

Methodologically, the acceleration indicator is relative in nature and sensitive to the initial level as well as to the low-base effect. In systems with very low initial capacity, even small absolute increases may generate high percentage values. For this reason, the interpretation of the Acceleration Index was carried out jointly with the analysis of installed capacity levels and cluster membership.

The application of this procedure makes it possible to distinguish systems that entered a phase of accelerated transformation after 2020 from those that maintained their previous growth rate or exhibited a slowdown. The Acceleration Index thus constitutes a synthetic measure of the change in the intensity of energy transition within a two-period framework. Figure 3 presents the methodological workflow.

Data preprocessing, statistical analysis, and visualization procedures were conducted using Python version 3.11 (Python Software Foundation, Wilmington, DE, USA). The analysis employed pandas version 2.2.0, NumPy version 1.26.4, scikit-learn version 1.4.0, SciPy version 1.12.0, and Matplotlib version 3.8.2 libraries for data processing, clustering procedures, statistical computations, and graphical visualization.

4. Results

This section outlines the clustering results derived from the applied analytical procedure. The emphasis is on how countries group when their renewable capacity trajectories are considered jointly, not in isolation.

Table 6. Summary of dynamic clustering results (k = 3).

Cluster	Countries	Number of Countries	General Pattern of Trajectory
Cluster 1	Australia, Austria, Belgium, China, Denmark, Finland, France, Germany, Greece, Indonesia, Italy, Japan, Laos, Malaysia, Netherlands, New Zealand, Philippines, Poland, Portugal, Romania, South Korea, Spain, Sweden, Thailand, Vietnam	25	Sustained expansion with post-2020 acceleration
Cluster 2	Bulgaria, Cambodia, Croatia, Cyprus, Czech Republic, Estonia, Hungary, Ireland, Latvia, Lithuania, Luxembourg, Malta, Myanmar, Singapore, Slovakia, Slovenia	16	Moderate growth, mostly linear, limited acceleration
Outlier case	Brunei	1	Stagnation, no structural change

and Appendix A summarize these outcomes, presenting the dynamic clustering of total installed renewable energy capacity (Total RES) for the period 2015–2024. The classification was obtained using Ward’s hierarchical method, with the solution constrained to three clusters (k = 3). This is a simplification, perhaps, but analytically useful.

The structure that emerges is somewhat asymmetrical. A large share of both major and medium-scale economies—across Europe (such as Germany, France, Italy, Spain, Poland, Sweden) and Asia (including China, Japan, South Korea, Indonesia, Vietnam, Malaysia)—is concentrated in the first cluster. This group reflects trajectories marked by sustained and relatively strong expansion of capacity over time. It is not perfectly uniform, but still, the overall pattern remains consistent. Growth is not only present but persistent.

The second cluster is composed mainly of smaller European countries, along with selected smaller Asian economies. Here, the growth paths appear more moderate, sometimes uneven. Often they originate from relatively low initial levels, which matters. The scale of transformation remains limited, at least when compared with the first cluster, and the dynamics suggest a slower structural adjustment rather than rapid expansion. The third cluster is specific in nature (Brunei), representing a nearly stagnant trajectory.

The structure of the table (Table 6 and Appendix A) confirms that the dynamics of renewable energy development are more strongly associated with economic scale and investment capacity than with continental affiliation. The largest systems exhibit the most consistent and accelerating growth patterns over the analyzed decade.

Figure 4 presents the trajectories of total installed renewable energy capacity over the period 2015–2024 for 42 countries grouped into three clusters based on the similarity of their growth paths. A clear divergence is visible between clusters, with Cluster 1 showing sustained and often accelerating growth, Cluster 2 exhibiting moderate and mostly linear increases, and the outlier case remaining nearly flat. The vertical reference line around 2020 indicates a structural break, after which acceleration becomes more pronounced in a subset of countries, particularly within Cluster 1.

Figure 5 presents the dendrogram of the dynamic clustering of growth trajectories in total installed renewable energy capacity over the period 2015–2024, indicating a clear separation of three groups of countries characterized by distinct development paths.

4.1. Description of the Clusters

4.1.1. Dynamic Cluster 1: Sustained Expansion and Structural Acceleration

The first dynamic cluster consists of most large- and medium-sized economies in both Europe and Asia. This cluster comprises the leading EU countries such as Germany, France, Italy, Spain, Poland, Sweden, Austria, and the Netherlands, in addition to the energy powerhouse countries in Asia, namely China, Japan, South Korea, Indonesia, Vietnam, Malaysia, and Australia. The identifying characteristic of the first cluster is the evident upward trend in the total installed renewable energy capacities between 2015 and 2024. In most of these countries, the upward trend is not only evident but also accelerating, especially after the period of 2019–2020, indicating an even more vigorous support for the deployment of renewable energy, the decreasing costs of the technologies, and the massive mobilization of capital.

From a structural point of view, this cluster corresponds to a set of systems that have surpassed a threshold of critical mass. Renewable energy development is integrated into the nation’s energy policies and reinforced through the development of the grid infrastructure and the markets. Noticeably, the geographical scope of this cluster is diverse in the sense that the economies of Europe and Asia show similar patterns of development despite their institutional differences.

4.1.2. Dynamic Cluster 2: Moderate Growth and Constrained Transformation

The second cluster comprises predominantly smaller European economies (e.g., Slovenia, Slovakia, Malta, Latvia, Luxembourg, Estonia, Hungary, Croatia, Bulgaria, and Czechia), together with several relatively small Asian energy systems (e.g., Cambodia, Myanmar, and Singapore). These countries exhibit positive yet comparatively moderate growth trajectories over the period of 2015–2024. Expansion is observable and, in many cases, stable; however, it does not display the marked acceleration identified in Cluster 1.

From the systemic point of view, the identified cluster can be characterized by low transformation potential. This limitation is not caused by the reluctance to adopt new technologies but is related to structural factors such as the economy’s scale, investment intensity, grid infrastructure, and market size, etc. In the case of the European Union member states in the identified cluster, the presence of regulatory conformity in the context of the EU’s climate and energy package is present, but the level of infrastructural development and capital mobilization is moderate. In the case of the Asian subgroup, the presence of small economies is related to conditions such as restricted internal demand, resource availability, and spatial factors, etc. Accordingly, the identified pattern implies an incremental adjustment in the energy mix, not a fundamental transformation of the energy system architecture.

4.1.3. Dynamic Outlier Case: Stagnation and Minimal Structural Change

The third cluster contains only one case, namely Brunei. The characteristic feature of its trajectory is the extremely low and flat level of renewable capacities over the entire period from 2015 to 2024. Unlike the other clusters, there is no indication of accelerating growth in the level of renewable capacities in the country.

The isolated position of Brunei is not the result of statistical fluctuations but is related to the structural specifics of the energy system in the country, where the level of fossil fuels used, the lack of necessity for diversification, and the high level of hydrocarbon reserves could have affected the need for the development of renewable energy sources. From the point of view of the dynamic clustering method, the isolated position of Brunei in the third cluster, in terms of the speed of transformation, can be regarded as an outlier, thereby proving that the differences in the level of transformation, in addition to the level of capacities, also involve differences in the patterns of evolution in the trajectories of the sample countries.

4.2. Comparative Analysis and CAGR

Dynamic clustering of growth trajectories in total installed renewable energy capacity over the period 2015–2024 suggests a pattern that does not align with simple geographic divisions. Not Europe versus Asia, at least not in a straightforward sense. Instead, what emerges is a transcontinental configuration in which countries exhibiting similar expansion paths tend to group together, regardless of location. This points toward a deeper structural logic. The pace of renewable energy transition seems embedded in systemic conditions—capital availability, institutional capacity, infrastructure readiness—rather than geography alone.

The first dynamic cluster brings together the largest and several medium-sized economies from both the European Union and East as well as Southeast Asia. Within Western Europe, this includes countries such as Germany, France, Italy, Spain, Austria, Sweden, and the Netherlands, while Poland and Romania represent the Central European component. On the Asian side, major actors include China, Japan, South Korea, Indonesia, Vietnam, and Malaysia, alongside Australia and New Zealand. It is a diverse set, geographically dispersed, yet internally coherent in terms of growth dynamics. The common denominator across these economies is a relatively high, and in many cases accelerating, rate of expansion in renewable energy capacity over the analyzed decade.

This acceleration becomes particularly visible after the 2019–2020 period, not uniformly, but with a noticeable shift in trajectory. It may be partially associated with the tightening of climate policy frameworks, the declining cost curves of photovoltaic and wind technologies, and also with broader geopolitical pressures influencing energy security strategies. The interaction of these factors does not produce identical outcomes across all countries—far from it—but the direction of change shows a certain convergence. Growth intensifies, sometimes abruptly, then stabilizes, or continues.

From an “energy geography” point of view, this group brings together countries whose natural conditions are extremely heterogeneous, ranging from industrialized Northern European countries to tropical Southeast Asian countries. What they share is not climate or resource availability, but rather the ability to mobilize capital and technology. Western Europe corresponds to the regulatory-institutional model of support systems and markets, while East Asia corresponds to an industrial model of investments. Beyond these differences, the growth patterns are similar. This may indicate a kind of global convergence between large countries that have overcome a threshold for the development of renewable energy.

The second cluster comprises mainly small European countries and some relatively small Asian economies. In Europe, these are mainly Central and Southern European countries (Slovenia, Slovakia, Malta, Latvia, Luxembourg, Estonia, Hungary, Croatia, Bulgaria, and Czechia), while in Asia, these are Cambodia, Myanmar, and Singapore. Their growth is positive but clearly more moderate. Their expansion is characterized by linearity without acceleration. Transformation is still underway, but it is not a breakthrough.

In some of the European Union countries included in this group, the development of renewables happens within the framework of a common European climate policy. At the same time, the scale of investment in this sector remains limited due to the smaller domestic markets. The smaller countries in the Asian group are characterized by isolated infrastructure projects rather than a wide range of technological diversification. The main characteristic of this group is the absence of the acceleration effect after 2020, as in the case of the first group. From a regional development perspective, this can be seen as a moderate level of adaptation rather than restructuring.

The third is only Brunei. Moreover, its trajectory is almost flat throughout the period, which means that there was little growth in renewables. From an economic geography point of view, this is a country that is highly dependent upon its fossil fuel income and where the economic structure faces little demand for diversification of its energy systems. Moreover, its geographic location in the hydrocarbon-rich region of Southeast Asia also contributed to its stagnation. The separation of this single-country cluster again confirms that Ward’s method identified its trajectory as being statistically and structurally different from all other countries.

From the point of view of continental regions, European countries are more heterogeneous than those of Asia. European Union countries are in both the first and second clusters, reflecting differences between large western economies and smaller central and Baltic countries. Asian countries, however, are more polarized, with large transforming economies of China, Japan, South Korea, and Indonesia on one side, and smaller systems with low growth dynamics on the other. This may indicate more profound developmental differences between countries of Asia than those of the European Union.

From the point of view of energy transition theory, it is possible to identify that the results support the idea of the existence of two main development pathways: an acceleration pathway (large- and medium-sized economies with increasing growth dynamics) and an incremental pathway (smaller states with stable but moderate growth rates). Although geographical aspects may be seen as influencing initial conditions, they do not directly impact development pathways. On the other hand, structural aspects are crucial for defining development pathways, such as market size, investment intensity, infrastructure, absorptive capacity, and potential for systemic integration.

Dynamic clustering shows that, although the global renewable energy transition is unevenly distributed, it is not specifically concentrated along continental lines. Large economies, whether located in Europe or Asia, show similar development pathways with accelerated growth rates. On the other hand, smaller states, despite being in different geographical areas, show similar structural constraints. These results support that renewable energy dynamics are essentially determined by scale and systemic potential rather than regional specificity.

As shown in Table 6, countries’ Compound Annual Growth Rate (CAGR) of total installed renewable capacity for the period from 2015 to 2024 is provided, as well as their allocation to dynamic clusters (k = 3). A clear differentiation of clusters from each other in terms of their growth regimes is observed. Countries of Cluster 1, including the largest economies of Europe and Asia, demonstrate a systematic pattern of growth; their CAGR values often reach several or even ten percent annually, which, given their high initial capacity base, results in high absolute growth rates. Cluster 2 includes mostly small countries that, in some cases, demonstrate high CAGR values. An outlier case includes a single country with a particular pattern of growth; thus, it represents a particular growth model.

As seen from

Table 7. CAGR 2015–2024 by country and dynamic cluster (k = 3).

Country	CAGR (%)	Cluster
Netherlands	22.85	1
South Korea	18.97	1
Poland	18.73	1
China	16.04	1
Australia	15.30	1
Vietnam	12.49	1
Finland	11.15	1
Belgium	11.03	1
Laos	9.90	1
Greece	9.38	1
Japan	7.87	1
Denmark	7.42	1
Spain	7.10	1
Germany	6.99	1
France	6.35	1
Portugal	5.96	1
Indonesia	5.84	1
Philippines	5.59	1
Sweden	5.19	1
Austria	5.08	1
Thailand	5.32	1
Italy	4.10	1
Romania	2.92	1
New Zealand	2.27	1
Malaysia	2.16	1
Hungary	26.14	2
Lithuania	23.50	2
Singapore	22.89	2
Estonia	15.68	2
Cyprus	15.54	2
Luxembourg	15.30	2
Malta	13.31	2
Ireland	10.24	2
Cambodia	9.63	2
Slovenia	6.78	2
Bulgaria	6.32	2
Croatia	5.53	2
Czechia	4.85	2
Latvia	2.94	2
Slovakia	1.36	2
Myanmar	0.94	2
Brunei	19.58	3

, renewable energy systems’ development demonstrates a difference not only in scale but also in tempo; moreover, it is more closely related to the level of development and capacity of systems than to their geographical location.

Table 8. Renewable energy capacity and share in global total (2024).

Country	Capacity (GW)	Share in Global (%)
Australia	63.51	1.43
Brunei	0.01	0.00
China	1827.27	41.08
Philippines	9.30	0.21
Indonesia	14.30	0.32
Japan	132.32	2.97
Cambodia	2.29	0.05
South Korea	33.78	0.76
Laos	10.37	0.23
Malaysia	9.47	0.21
Myanmar	3.54	0.08
New Zealand	8.93	0.20
Singapore	1.37	0.03
Thailand	12.60	0.28
Vietnam	49.02	1.10
Austria	28.86	0.65
Belgium	16.28	0.37
Bulgaria	7.18	0.16
Croatia	4.40	0.10
Cyprus	0.90	0.02
Czechia	6.45	0.15
Denmark	13.54	0.30
Estonia	2.20	0.05
Finland	15.90	0.36
France	74.34	1.67
Greece	18.23	0.41
Spain	88.50	1.99
Netherlands	36.64	0.82
Ireland	6.61	0.15
Lithuania	4.62	0.10
Luxembourg	0.86	0.02
Latvia	2.31	0.05
Malta	0.24	0.01
Germany	178.66	4.02
Poland	32.42	0.73
Portugal	20.48	0.46
Romania	14.53	0.33
Slovakia	2.69	0.06
Slovenia	2.56	0.06
Sweden	42.36	0.95
Hungary	8.71	0.20
Italy	72.12	1.62

presents the absolute renewable energy capacity and its share in global capacity for all analyzed countries in 2024, providing a structural perspective complementary to the trajectory-based analysis. The results reveal a highly concentrated global market, with China accounting for over 40% of total capacity, while most countries contribute marginal shares below 1%.

The aggregated CAGR table by clusters (

Table 9. CAGR 2015–2024: cluster summary (k = 3).

Cluster	Mean CAGR (%)	Median CAGR (%)	Std Dev (%)
1	9.04	7.1	5.56
2	11.31	9.93	8.0
3	19.58	19.58	-

) demonstrates that the groups identified based on 2015–2024 trajectories differ not only in installed capacity levels but also in the pace of renewable energy expansion. Cluster 1, comprising the largest and medium-sized economies of Europe and Asia, is characterized by a stable and moderate average growth rate, combined with relatively low intra-group variability. This pattern indicates a mature and systematic model of transformation.

Cluster 2 records a higher mean and median CAGR values, yet simultaneously exhibits greater standard deviation. This suggests more heterogeneous growth paths, often driven by a low initial capacity base and reflecting earlier stages of transition. The outlier, Cluster 3, represented by a single case, confirms its structurally distinct nature. The table (Table 9) confirms that renewable energy transition dynamics vary across clusters and are shaped both by system scale and by the developmental stage of the respective energy systems.

The CAGR estimates calculated for the period of 2015–2024 show that the pace of growth in installed renewable energy capacity differs quite markedly across the analyzed set of countries. The differentiation appears both between the identified dynamic clusters and also within them, which complicates any overly simplified interpretation. Still, one general pattern remains clear. All observed cases exhibit positive growth rates. This suggests that the entire decade can be interpreted as a phase of consistent expansion in renewable energy systems, though it is not uniform—far from it. What varies is not the direction of change, since that remains broadly aligned across countries, but rather its intensity, tempo, and underlying structural features. Some trajectories indicate gradual scaling, others a more rapid acceleration, occasionally uneven. The Compound Annual Growth Rate functions here as a synthetic indicator of these processes, capturing the average annual pace of transformation over the full-time horizon. By construction, it reduces the influence of short-term volatility—year-to-year deviations, temporary slowdowns, sudden spikes—and instead highlights the longer-term developmental pathway. A kind of smoothed signal, one could say, though still sensitive to initial conditions and end-period effects.

In the first cluster, which includes the largest and medium-sized economies of Europe and Asia, economic growth was moderate but stable. The mean CAGR values are lower than those of the second cluster; however, it should be kept in mind that these values must be interpreted regarding the high base effect. Countries such as Germany, France, Japan, and China start with already substantial installed capacity. In these cases, even single-digit growth rates represent very large absolute values of capacity additions. This cluster represents a mature transformation model with systematic growth, high investment volumes, and a somewhat predictable trajectory. Low intra-cluster variability indicates that the transition process is already institutionally embedded and structurally consolidated.

Cluster 2 has a larger mean and median CAGR, which means that the percentage growth is larger. Yet, to a considerable extent, it is because of the lower starting point in 2015. Small economies, especially those from Central and Eastern European countries and some Southeast Asian ones, developed renewable energy capacity dynamically, although from a limited starting point. While relative growth is considerable, absolute growth is lower than that of Cluster 1. A larger standard deviation is seen here, too, which shows that while some of these countries clearly accelerated after 2020, others display a more linear development path.

The third cluster, for which there is only one case, has the highest CAGR recorded. This result must be interpreted with caution. The extremely high rate is the result of an extremely low starting point in terms of capacity level. This is a classic case of the low-base effect, where even a small increase in the numerator results in an extremely high CAGR.

The results for the period 2015–2024 indicate that the rate of transition to renewable energy is dependent on the level of development of the energy system, in addition to its structural scale. Large economies tend to grow at a steady rate, resulting in high increases in absolute terms, but low increases in relative terms. Conversely, small economies tend to grow at a faster rate, but the impact is not so high in terms of infrastructure development. The general implication is that the results obtained through the CAGR must be viewed in the context of the starting point in terms of capacity level and the overall scale of the energy system, after which it can be used to make an accurate assessment of the depth of the energy transition.

4.3. Analyzing by Acceleration Index

The Acceleration Index compares the differences between the growth rates of renewable energy in the pre-2020 period (2015–2019) and the post-2020 period (2020–2024).

The findings reveal that for Cluster 1, where the largest European and Asian economies are included, the acceleration effect is moderate but systematic. This indicates the reinforcement of an established growth trajectory, but it did not trigger a structural change. In other words, the post-2020 period simply reinforced an established growth trajectory, rather than initiating a new one.

The findings for Cluster 2 reveal more variability. While some countries show clear signs of acceleration in the post-2020 period, others exhibit stable growth rates. This may be explained by differences in adaptive capacity among smaller energy systems. Institutional readiness, flexibility in the grid, financial space for investments, and previous levels of penetration may all affect the response to the shock. The post-2020 period may be seen as differentiating, but it did not trigger uniform acceleration.

An outlier case, as represented by a single case, displays a particular dynamic largely governed by the low-base effect. In this context, percentage acceleration can appear substantial, yet it does not necessarily imply a new systemic restructuring of the energy mix.

Table 10. Acceleration Index (2015–2019 vs. 2020–2024) for k = 3. Country-level results.

Country	CAGR_Early	CAGR_Late	Acceleration	Cluster
Australia	15.97	13.28	−2.69	1
Austria	2.98	8.08	5.11	1
Belgium	10.32	9.91	−0.41	1
China	12.18	19.49	7.3	1
Denmark	6.6	8.82	2.22	1
Philippines	4.76	7.05	2.28	1
Finland	6.38	16.2	9.82	1
France	5.46	7.8	2.34	1
Greece	4.87	13.97	9.1	1
Spain	3.43	11.46	8.03	1
Netherlands	21.59	18.38	−3.21	1
Indonesia	4.7	8.11	3.41	1
Japan	10.2	5.22	−4.98	1
South Korea	25.62	11.2	−14.42	1
Laos	9.54	6.09	−3.45	1
Malaysia	0.28	2.55	2.27	1
Germany	6.19	8.31	2.12	1
New Zealand	0.36	4.59	4.23	1
Poland	7.74	27.9	20.16	1
Portugal	3.93	9.39	5.47	1
Romania	−0.1	6.91	7.01	1
Sweden	3.77	7.31	3.54	1
Thailand	10.31	1.57	−8.74	1
Vietnam	11.31	6.3	−5.01	1
Italy	1.87	6.91	5.04	1
Bulgaria	1.09	13.24	12.16	2
Croatia	3.12	7.84	4.72	2
Cyprus	7.18	22.3	15.12	2
Czechia	0.79	10.03	9.24	2
Estonia	4.78	27.69	22.91	2
Ireland	13.45	8.35	−5.11	2
Cambodia	11.51	7.22	−4.28	2
Lithuania	6.41	45.98	39.57	2
Luxembourg	12.73	18.96	6.22	2
Malta	19.3	6.39	−12.92	2
Myanmar	1.19	0.95	−0.24	2
Singapore	19.4	29.14	9.74	2
Slovakia	0.49	3.15	2.66	2
Slovenia	1.58	12.28	10.7	2
Hungary	20.73	30.27	9.54	2
Latvia	0.61	6.08	5.47	2
Brunei	0.0	49.53	49.53	3

supports the idea that the post-2020 response was not regionally uniform. The patterns of acceleration were largely influenced by the capacity of the system, economic scale, and the preceding stage of transition.

The Acceleration Index results—defined as the difference between renewable energy growth rates in the sub-periods of 2015–2019 and 2020–2024—provide a way to trace shifts in the dynamics of the transition process under changing external conditions. In particular, they reflect the influence of geopolitical disruptions, the energy crisis, and the post-2020 tightening of climate policy frameworks. The construction of the index is relatively simple, but its interpretation is not entirely straightforward. It indicates whether a given energy system has moved into a phase of faster expansion, or whether it continues along a previously established trajectory. In many cases, a tendency toward acceleration can be observed. Though uneven. The scale of this effect varies quite significantly across clusters, which suggests differentiated adaptive capacities.

In the case of Cluster 1—comprising the largest and several medium-sized economies in Europe and Asia—the acceleration pattern appears moderate, yet relatively stable across countries. Economies such as Germany, France, Poland, China, and Japan had already exhibited sustained growth in renewable capacity during the 2015–2019 period. So, the post-2020 phase does not introduce a clear structural break. Rather, it reflects a continuation, perhaps an intensification, of an already established trend: incremental, but persistent. This continuity may point to a deeper level of institutional and infrastructural embedding of the transition process within these systems. Investments initiated earlier seem to have been carried forward and, in some cases, accelerated under new policy and market conditions—not necessarily redesigned, more often scaled up. The observed pattern could therefore be interpreted as an indicator of systemic maturity—where energy systems are capable of absorbing external shocks and policy impulses without destabilizing the overall direction of transition. There is a certain resilience, though it is not uniform across all cases.

In Cluster 2, comprising mostly smaller European economies and some Southeast Asian countries, the post-2020 period shows a more heterogeneous trend. There are countries that show a clear acceleration trend. This could be seen as a catch-up effect or as a strengthened investment mobilization in the aftermath of the energy crisis. Other countries in this cluster show similar growth rates to those in the preceding period. This could be an indication of structural limitations in terms of market size, limited grid infrastructures, and investment potential. The heterogeneity of the acceleration in this cluster confirms our assumption of smaller systems having a more heterogeneous reaction to global stimuli.

An outlier case comprises a single case and shows a particular trend. The high acceleration value in this case can be largely attributed to the low-base effect. In this case, even relatively low increases in the period between 2020 and 2024 amount to substantial percentage increases. The interpretation of this trend requires caution. A high value of the index does not necessarily imply a fundamental structural change in the country concerned. It could be the result of isolated investment activities in a country of extremely limited scale.

4.4. Comprehensive Cluster Analysis

The analysis based on the Acceleration Index suggests that the period after 2020 did not generate a homogeneous breakthrough across the set of analyzed countries. Rather than convergence, what becomes visible is a reinforcement of already existing structural asymmetries. Larger economies tend to continue along relatively stable and coordinated expansion paths. Smaller systems, in contrast, respond in a more heterogeneous manner—sometimes adapting, sometimes lagging, occasionally shifting direction. And there are also cases that remain outside the dominant trajectory of transformation altogether and are not fully aligned.

This differentiated pattern implies that acceleration in renewable energy development is not simply triggered by global shocks or common policy impulses. It seems more deeply rooted. The key drivers appear to be systemic in nature—linked to the prior stage of transition, the availability and mobilization of capital, and the degree of infrastructural preparedness. These elements condition the ability of a given system to absorb and translate external stimuli into actual capacity growth. Geography alone does not explain much here, nor does a uniform reaction to post-2020 events. The process is more uneven and structurally embedded.

4.4.1. Cluster 1: Systems of Durable and Scalable Transformation

The first group is composed of the largest and several medium-sized economies spanning both Europe and Asia. It includes countries such as Germany, France, Italy, Spain, Poland, Sweden, as well as China, Japan, South Korea, Indonesia, Vietnam, and Australia. A geographically dispersed set, yet, analytically, they form a relatively coherent cluster. What brings these economies together is not location, but the pattern of their development in renewable energy capacity over the analyzed decade. The trajectories show a systematic increase—consistent and sustained, and, in many cases, also accelerating, especially in the later years of the period. This is not identical across all countries, of course, but the general direction remains aligned. Expansion is not incidental here. It reflects a more embedded, ongoing transformation process.

The paths of these economies show a nearly monotonic character without any strong signs of stagnation or regression. In the post-2020 period, their growth tends to strengthen, suggesting their potential to respond to political and market stimuli such as the energy crisis, the rise in fossil fuel prices, and the acceleration of climate policies. From a systemic viewpoint, these economies have reached the scale effects, possessing well-developed grid infrastructures, access to capital investment, and well-developed regulatory regimes.

From a geographical viewpoint, this group is transcontinental in nature, thus supporting the notion that renewable energy patterns depend primarily on system size and structural potential rather than geographical characteristics.

4.4.2. Cluster 2: Incremental and Structurally Constrained Systems

The second cluster includes smaller EU member states (Slovenia, Slovakia, Malta, Latvia, Lithuania, Estonia, Hungary, Czechia, Bulgaria, and Croatia) and smaller Asian economies (Cambodia, Myanmar, and Singapore).

In Cluster 2, the trend is positive but less steep and less homogeneous than that of Cluster 1. The growth is linear and moderate without any structural breaks after 2020. In some of these states, there is no strong acceleration effect, which may be because of infrastructural or financial issues. There may be issues of scalability for smaller energy infrastructures, which may face difficulties like a lower market size, lower potential for integrating variable renewable energy sources, or lower investment potential.

4.4.3. Outlier Case

The third cluster comprises one case with a uniquely different pattern (Brunei). It is defined by an almost flat pattern of development during most of the period analyzed, with rare exceptions of increasing values. This cluster also verifies that the algorithm not only detects level differences but also temporal patterns.

Such an economy is characterized by high dependence on fossil fuels and low transformation pressures. High relative values of indicators like CAGR or the Acceleration Index may be caused by an extremely low starting value.

4.5. Summary of the Clustering Results

The juxtaposition of dynamic clusters with the CAGR and Acceleration Index analysis reveals three distinct development models (Figure 6 and Figure 7):

• Mature expansion model (Cluster 1)—moderate and stable growth rates combined with a very high-capacity base; strong ability to respond to post-2020 impulses.
• Incremental model (Cluster 2)—higher relative growth in selected cases, but with limited systemic scale; greater variability in post-2020 response.
• Marginal model (Outlier case)—absence of sustained dynamics, with only sporadic or point-based increases.

The key observation is that differences between clusters are systemic rather than regional. European and Asian countries appear within the same dynamic groups, suggesting a global convergence of growth mechanisms among economies of comparable scale.

The results of the dynamic clustering suggest that the transformation toward renewable energy tends to follow a logic shaped by scale effects and the broader notion of systemic absorptive capacity. Larger economies exhibit expansion patterns that are relatively stable, in many cases also predictable. This is not accidental. It is supported by existing infrastructure, institutional continuity, and the ability to coordinate investments over time. Smaller countries, on the other hand, sometimes record higher growth rates when expressed in percentage terms. But this faster growth—fragmented, occasionally volatile—does not necessarily translate into a deeper structural reconfiguration at the regional scale. The effect remains limited.

At a more general level, the findings indicate that renewable energy dynamics cannot be explained primarily through geographic determinants alone, such as resource endowment or climatic conditions. These factors matter, but they do not dominate. More decisive appears to be the capacity of a system to mobilize financial resources, to integrate new technologies into existing energy structures, and to sustain policy direction over longer time horizons. Not always visible at first glance, yet persistent. In the end, what differentiates countries is less their location and more their systemic weight—critical mass, institutional maturity, and the coherence of the transition process itself. These elements shape whether a given economy enters a path of accelerating transformation or remains within a segment characterized by constrained, and often discontinuous, dynamic capacity.

From the standpoint of policy, one could conclude that success in transitioning towards renewable energy is associated less with regionality than with a combination of other factors, such as the systemic capabilities associated with the intensity of investments, infrastructure readiness, and coordination of institutions. As far as the European Union is concerned, the research indicates the appropriateness of employing the model of regulation and institution-building through binding targets and integrated markets, as well as coordinated policies, which enable sustainable development paths in terms of renewables’ growth. In the case of RCEP countries, one could point out the existence of a more heterogeneous model focused on production and characterized by the use of industrial policies and investments for the rapid growth of economies. Small systems, in their turn, do not possess the necessary capabilities. The above means that a policy transfer will be rather difficult across different regions. From a broader perspective, one should say that increasing the speed of the transition process involves not only the implementation of technologies but also enhancing systemic and governance capabilities to ensure sustainable investments in renewables.

5. Discussion

The results of the clustering analysis can be interpreted from the viewpoint of the theory of energy transition, particularly the theory of multi-level perspective (MLP) on socio-technical transitions proposed by Geels [13,14]. It is emphasized that transitions are a result of the interplay between niche innovations, socio-technical regimes, and landscape pressures [15,16,86]. The division of countries into clusters of accelerated, incremental, and stagnant transitions can be attributed to the differences in the configurations of regimes rather than the technologies used. The large- and medium-sized systems have already assimilated renewable energy sources into their dominant regime structures; thus, the scaling-up of transitions was achieved after 2020. In contrast, the small systems are still in the initial stages of transition, characterized by a restricted niche technology diffusion due to intra- and institutional lock-ins.

From the point of view of the ecological modernization theory, which has been propounded by scholars such as Mol and Spaargaren [15,16], the results provide support for the idea that environmental change can be integrated with economic development, rather than being opposed to it. In the case of the economies found in Cluster 1, it has been shown that the expansion of renewables can be integrated with industrial competitiveness and capital accumulation. In such economies, environmental policy instruments, markets, and technological change are mutually reinforcing. The smaller acceleration found for the other economies could be interpreted as indicating that ecological modernization is not a spontaneous process, but that it needs sufficient institutional density, financial resources, and infrastructural readiness; otherwise, it would be incremental rather than structural change [87,88,89,90,91,92,93,94,95].

The results can be justified based on the theories of convergence and divergence in economic development. The classical theory of economic development and growth assumes that the lower the development level of an economic system, the higher its growth rate. This assumes the “catching up” effect. This assumption is partially supported by the findings, as indicated by the higher values of the CAGR for smaller economies. The results of the dynamic clustering model indicate that the “catching up” effect is not supported by the absolute convergence of economic capacities. This is in accordance with the assumptions of the endogenous growth theory, which assumes that the basis of economic divergence is the absorptive capacities of technology and human capital [96,97]. The large economies, despite lower growth rates, have sustained their positions due to the large absolute sizes of the economies, which provide the basis for continuous economic growth. The transition of the renewable energy industry is an example of conditional convergence [98,99,100,101,102]. Achieving success in energy transformation is a long process; many industries in the EU and RCEP countries have undergone transformation for years, e.g., the steel industry [103,104], power industry [105], etc.

These patterns may be explained by the knowledge accumulated from the institutional theory and the path dependence theory, particularly the concept of carbon lock-in presented by Unruh [106]. The countries that are experiencing stagnation and acceleration seem to be locked into the old fossil fuel infrastructures and economic systems. The fact that these patterns exist proves that there is a long-term institutionalization of the energy systems. This shock, which occurred after 2020, may be explained as an amplifier of the pre-existing conditions rather than a transformative event per se. Renewable energy systems may be explained as not being dependent on the shock effects but rather on the historical development of the institutional capacities and infrastructural maturity, and political continuity, as presented in [107,108,109].

6. Conclusions

The results of the conducted dynamic analysis indicate that the development of renewable energy capacity in the years 2015–2024 follows three clearly distinct transformation regimes. The application of hierarchical clustering made it possible to distinguish a group of systems characterized by sustained and scalable expansion, a group undergoing incremental and moderate transformation, and a case of structural stagnation. This means that countries differ not only in the level of installed capacity, but above all in the dynamics and shape of their growth trajectories. Thus, the first research question receives a clear answer: it is empirically possible to identify repeatable patterns of renewable energy development that reflect different stages of energy transition maturity.

The analysis further demonstrates that this differentiation is not regional in a geographical sense. European and Asian countries appear within the same dynamic clusters, which implies that membership in the EU or RCEP does not determine the pace or character of transformation. Structural factors are decisive: the size of the economy, investment capacity, the development of grid infrastructure, and institutional stability. The response to the second research question is therefore unambiguous—the dynamics of renewable energy development are primarily a function of systemic scale and capacity, rather than a simple continental division.

According to the results of the post-2020 acceleration analysis, it seems that this period mainly reinforced existing growth paths in many systems, but it did not mark a new “structural breakthrough” for all systems. Large and mature systems have continued their expansion in a quite stable way, sometimes even with a moderate acceleration, while smaller systems reacted differently, from a clear “catch-up effect” to a maintenance of existing growth rates. Therefore, it is necessary to consider 2020 as a “reinforcing factor” of existing tendencies, which did not mark a “structural breakthrough.” Thus, it is possible to give a “conditioned” answer to the third research question: yes, acceleration occurred, but it depended on the previous maturity of the energy system.

With regard to practical relevance, it can be assumed that the results of the research will help to develop the concept of the strategy for renewable energy transition at the country and regional level. It is apparent that since development clusters vary substantially, it would be necessary to develop the instruments of influence according to their structure. Specifically, in terms of large-scale energy systems with significant growth rates (Cluster 1), policy-makers could consider measures aimed at improving the infrastructure of the network, its expansion, and capacity increase. On the contrary, for small energy systems characterized by structural restrictions (Cluster 2), measures aimed at accelerating the transition process will have to include investments, infrastructure creation, and institutional improvement, among others. Finally, the results of the research also imply a possibility to use them for benchmarking the performance of a particular country’s energy system with respect to other states, helping to determine feasible development ways.

The major scientific contribution of the present paper is its dynamic trajectory-based approach in the analysis of renewable energy transformation in diverse economic systems. Unlike the traditional cross-section-based approach in the analysis of renewable energy transformation in different economic systems, the present paper introduces a unique combination of hierarchical dynamic clustering, CAGR analysis, and an innovative Acceleration Index in the analysis of renewable energy transformation in different economic systems. By revealing the fact that the patterns of renewable energy expansion are primarily dependent on the scale and absorptive capacity of the economic systems rather than the geographical location of the economic systems, the present paper makes a unique contribution in the field of energy transition analysis. The present paper also makes a unique contribution in the field of methodology by introducing a unique combination of dynamic clustering and CAGR analysis in the analysis of renewable energy transformation in different economic systems.

The limitation of the present study is a fact which is based on the exclusive use of the total installed capacity of renewable energies. The expansion of renewable energies is quantitatively measured in terms of the installed capacity without considering the qualitative characteristics of the expansion. Installed capacity is a structural variable, but the expansion in installed capacity does not measure the effectiveness of the expansion in terms of the performance of the system. The clustering is also based on standardized trajectories and the use of Euclidean distances in Ward’s method, which may also be dependent on the type of transformations used in the data (logarithmic scale and z-score standardization) and the number of clusters used (k = 3). The analysis is based on IRENA renewable energy capacity statistics, which provide a consistent and internationally comparable dataset. However, it should be acknowledged that cross-country data may be subject to differences in national reporting practices, data revisions, and classification standards, which may introduce minor inconsistencies in the time series. To ensure comparability, only countries with complete observations for all years (2015–2024) were included in the analysis. No imputation procedures were applied, and incomplete cases were excluded at the preprocessing stage.