Towards 2030 and beyond: assessing future energy efficiency policies and trends using ODEX methodology

Building upon a diverse body of empirical and theoretical literature, this study advances the hypothesis that future progress in energy efficiency is best achieved through a systemic and integrated approach that combines technological innovation, targeted policy design, and behavioural change mechanisms. This hypothesis reflects growing recognition in the academic and policy domains that isolated interventions are insufficient to close the persistent efficiency gap observed across sectors and countries, particularly EU Member States. The European Union acts as a central driver of energy-efficiency progress by setting binding targets, harmonizing policy frameworks, and coordinating measures that guide and support national actions across all member states. According to Mah et al. (2025) six of the ten most energy-efficient countries worldwide are EU member states, underscoring the Union’s significant influence in shaping ambitious and effective efficiency policies.

Energy efficiency (EE) is widely recognized as a central pillar of the clean-energy transition and one of the most cost-effective strategies for reducing greenhouse gas emissions, strengthening energy security, and supporting long-term economic resilience as recognised by Filippini et al. (2014), Horowitz and Bertoldi (2015), Cahill et al. (2010) and Boonekamp (2006). Within the European Union (EU), this recognition has been operationalized through the Energy Efficiency Directive and the Energy Efficiency First principle (von Malmborg, 2023a, 2023b)), both aimed at accelerating efficiency improvements across major sectors and governance levels (Economidou et al. (2022); Malinauskaite et al., (2019, 2020)).

Despite this strong policy architecture, most Member States continue to exhibit a persistent efficiency gap, defined as the difference between technically and economically achievable savings and their more modest real-world uptake as reported by Bukarica and Robić (2013) and Pusnik et al. (2017).

A considerable body of research attributes this gap to behavioural, informational, and institutional barriers. Information asymmetries and bounded rationality limit the adoption of efficient technologies as reported in Bukarica and Tomšić (2017), while behavioural biases and habitual consumption patterns further hinder cost-effective investments (Reuter et al. (2020)). Changing consumption trends driven by digitalisation, and evolving mobility practices add new complexities to demand-side behaviour and the effectiveness of traditional policy interventions as shown in Brugger et al. (2021). Additional challenges emerge at the level of public and commercial buildings where high upfront costs, fragmented regulatory responsibilities, and slow renovation cycles continue to impede progress, as shown in recent assessments of large public buildings by Zhou et al. (2024). These findings resonate with broader concerns of Papantonis et al. (2022) about governance fragmentation and uneven policy implementation across Member States and align with Mandel and Pató (2024) analysis showing that demand-side measures grounded in the Energy Efficiency First principle can effectively compete with supply-side investments.

At the energy-system level, Europe’s high import dependency and exposure to volatile global markets further underline the strategic value of reducing demand as reported by Guarascio et al. (2025). At the same time, rebound effects, where efficiency improvements induce additional consumption, continue to undermine expected savings as highlighted by Karakaya et al. (2024). Chu et al. (2024) report that broader economic structures, including the size of the shadow economy and national economic sophistication, also shape energy-intensity trajectories. Recent research further highlights those divergences in local energy planning (Palermo et al. (2024)), inconsistencies in energy performance certification methodologies (Sesana et al. (2024)), and persistent municipal-level barriers (Rivas et al. (2022)). Aforementioned studies collectively point to the need for more coherent multi-level governance. In parallel, studies on behavioural change and prosumerism underscore the importance of integrating sufficiency-oriented strategies into the efficiency agenda (Korsnes et al. (2024)). Complementary evaluation studies confirm the need for more robust methods to assess whether policies truly deliver measurable energy savings as highlighted by Bertoldi and Mosconi (2020).

Methodological innovation has attempted to keep pace with these complexities. Approaches such as stochastic frontier analysis provide insights into the gap between actual and optimal energy demand (Filippini and Hunt (2012)), while hybrid top-down and bottom-up indicators improve sector-level evaluation (Reuter et al. (2021)). Meta-modelling efforts offer simplified yet comprehensive tools for evaluating efficiency policy impacts across multiple dimensions as reported in Bashmakov et al. (2024). Policy-learning research further demonstrates that iterative feedback and stakeholder interaction shape the effectiveness and longevity of efficiency policies as reported by von Malmborg (2024). Together, these developments underscore a clear message: energy-efficiency outcomes reflect the interaction of technical, behavioural, economic, and institutional dynamics, and thus require evaluation frameworks capable of integrating these dimensions over time.

Over the past two decades, considerable effort has gone into developing accounting frameworks and indicator-based tools for tracking energy-efficiency trends. Ang et al. (2010) review major accounting systems and emphasize the need for more uniform concepts and decomposition structures. At the EU level, indicator work increasingly seeks to connect observed consumption patterns with policy interventions. Horowitz and Bertoldi (2015) for example, propose a harmonized model combining bottom-up indicators with top-down data to estimate realized policy impacts. More recently, decomposition studies have extended this logic across countries and sectors. Trotta (2020) uses ex-post index decomposition to quantify Finland’s efficiency-related energy and climate benefits and warns against relying solely on ex-ante engineering estimates that often overstate savings. Jain (2025) applies decomposition to 144 countries, illustrating both the strengths and limitations of ex-post approaches for understanding long-term efficiency trends.

Sector-specific analyses similarly highlight the close link between indicators and policy objectives. Tsemekidi Tzeiranaki et al. (2023) examine tertiary-sector trends in relation to EU climate targets, while Rodriguez et al. (2020) stress that indicators are not neutral but embed assumptions about productivity, decoupling, and policy priorities. Together, these contributions show substantial progress in developing ex-post indicator frameworks but also reveal that they are largely used for retrospective attribution or communication rather than integrated with forward-looking scenario exercises.

Despite progress in measurement and modelling, a major methodological gap remains. Ex-post and ex-ante assessments of energy-efficiency developments are generally conducted using incompatible analytical frameworks grounded in different assumptions, variables, and sectoral boundaries. The ODEX indicator is widely used for retrospective assessment of sectoral efficiency trends, yet it has rarely been extended to forward-looking contexts. Conversely, ex-ante scenario modelling relies on conceptual structures that cannot be directly compared with historically observed ODEX trends. This fragmentation limits policymakers’ ability to judge whether projected efficiency pathways are consistent with empirical evidence or whether planned measures reflect realistic trajectories.

The present study addresses this methodological gap by applying the ODEX approach to both historical data and forward-looking policy scenarios. Using a single indicator framework for ex-post and ex-ante assessments creates a coherent basis for comparing past performance with expected future outcomes, enabling policymakers to evaluate whether projected efficiency gains align with observed trends. The novel contribution of this study lies in extending ODEX beyond its conventional retrospective role to serve as a tool for scenario analysis, thereby providing a consistent and transparent framework for assessing long-term policy strategies. By integrating ex-post evidence and ex-ante expectations within one analytical structure the study strengthens the empirical foundation on which future EU energy-efficiency planning can rely.

The overarching objective of this research is to develop and apply a robust, standardized framework for assessing energy efficiency (EE) policy impacts using the ODEX methodology, focusing on both ex-ante (forecasted) and ex-post (historical) evaluations. By bridging the gap between observed performance data and prospective scenarios, this research aims to offer a comprehensive tool for policymakers, analysts, and other stakeholders to better understand how EE measures contribute to national and EU-level climate and energy targets. Furthermore, it explores how integrating technical and gross indicators can more accurately capture the influence of technology-driven improvements and behavioral changes on overall energy consumption trends.

The scope of this study extends across the major final energy consumption sectors, namely industry, transport, households, and services, covering two European countries, namely Slovenia and Croatia. Sectoral disaggregation serves two purposes: it helps identify which areas offer the greatest potential for further EE improvements and clarifies how policy interventions can be tailored to achieve maximum impact. In doing so, the research addresses existing data and methodological disparities among Member States, illustrating how a harmonized, flexible approach can accommodate differences in data availability and granularity.

Ultimately, the framework and findings presented here are intended to guide evidence-based policy design and evaluation, highlighting best practices in data harmonization, projection modelling, and cross-country benchmarking. By providing actionable insights, the study contributes to ongoing efforts to scale up energy savings, meet the ambitious targets set out in the European Energy Efficiency Directive, and support broader climate objectives, including decarbonization and sustainable economic growth.

The remainder of this paper is structured as follows. The following section outlines the purpose of the study and the motivation for applying the ODEX methodology in both ex-ante and ex-post evaluations. The subsequent section, Methodology, details the analytical framework, data sources, and calculation procedures employed to assess energy efficiency trends. The section Results and Discussion presents empirical findings from the case studies of Slovenia and Croatia, focusing on sectoral dynamics, policy impacts, and comparative insights. Finally, the paper concludes with the section Conclusion, which summarizes the main findings and outlines recommendations for future research and policy development.

The ODEX (ODYSSEE Energy Efficiency Index as presented in Lapillonne (2020) methodology serves as a standardized, top-down framework for monitoring and evaluating energy efficiency (EE) policies within the European Union. In Energy Efficiency Directive (EU) 2018/2002 (European Parliament and the Council (2018)) and Directive (EU) 2023/1791 (European Parliament and the Council (2023)) stipulates binding energy-saving obligations for Member States (Article 4, 5 and Article 8), noting the necessity of reliable evaluation methods to ensure that efficiency measures deliver tangible benefits. The subsequent European Green Deal (European Commission (2019)) further emphasizes energy efficiency as an essential strategy in the drive toward climate neutrality by 2050. By providing harmonized data and transparent benchmarks, the ODEX is recognized within these initiatives for its utility in verifying whether actual efficiency gains align with policy objectives.

In addition, the ODYSSEE-MURE Project established the ODEX as an integral component of a broader EU-wide system for policy evaluation and cross-country comparisons. Through this platform, policymakers and stakeholders gain access to in-depth analyses and methodological guidance, ensuring that energy-saving achievements can be tracked. As a result, the ODEX serves not only as a diagnostic measure of progress, but also as an instrument for informed decision-making, contributing to the EU’s overall endeavour to reduce energy consumption and greenhouse gas emissions.

By isolating technology-driven efficiency gains from structural or behavioural shifts, ODEX offers a robust measure of genuine improvements across industry, transport, residential, and service sectors. ODEX cannot separate policy induced savings from autonomous progress, but provides information on total aggregated savings. Unlike conventional energy intensity indicators (e.g., energy consumption per unit of GDP), ODEX focuses on changes in specific energy consumption expressed in physical units, thus providing a more accurate assessment of energy efficiency advancements (Lapillonne (2020)) as energy consumption is more directly dependent on physical production than economic value.

A key strength of ODEX lies in its disaggregation capability, enabling detailed examinations of sub sectoral trends (e.g., residential heating vs. industrial processes). Additionally, the methodology distinguishes between gross (apparent) and technical efficiency gains, allowing analysts to account for non-technical influences on energy use. This distinction underpins a more nuanced interpretation of observed changes and the identification of genuine technological improvements as reported in Lapillonne (2020).

ODEX also enhances cross-country comparability by harmonizing data and methodologies, thereby facilitating benchmarking of EE measures among EU member states. Through its mathematical framework, the index can be converted into aggregated energy savings, supporting monitoring obligations under directives such as the Energy Efficiency Directive (EED). As a result, ODEX provides policymakers with a reliable tool for ex-post evaluation, strategic planning, and ensuring alignment with broader EU decarbonization objectives.

The ODEX as described in Lapillonne (2020) is a sectoral composite indicator designed to quantitatively evaluate the progress of energy efficiency across major final energy consumption sectors namely, industry, transport, households, and services, within the European Union. Developed as part of the ODYSSEE-MURE initiative, the ODEX facilitates top-down policy monitoring by integrating a harmonized framework for measuring sectoral performance in terms of energy efficiency improvements.

Methodologically, ODEX is constructed as a weighted aggregate of sub-sectoral indices, each representing variations in specific energy consumption over time. These specific consumption indicators are expressed in physical units (e.g., kWh/year per appliance, toe/m² for heating, litres per 100 km for transport modes) and are selected to serve as proxies for energy efficiency changes, independent of macroeconomic distortions. The weighting structure is based on the relative share of each sub-sector in the total final energy consumption of the corresponding sector, ensuring representativeness and comparability across different energy uses.

To mitigate short-term volatility due to exogenous factors such as climatic variability, economic downturns, or statistical inconsistencies, ODEX is calculated using a three-year moving average. This approach enhances the stability of the indicator and better captures underlying trends driven by technological and behavioural transformations.

ODEX exists in two principal forms: the gross (or regular) ODEX and the technical ODEX. The gross ODEX reflects observed trends in specific consumption, including variations resulting from operational inefficiencies, user behaviour, and structural shifts. In contrast, the technical ODEX isolates pure technological advancements by neutralizing instances of increased specific consumption that are not attributed to efficiency improvement (e.g., under-utilization of industrial equipment during economic recessions or increased appliance size in households). This distinction is essential for disentangling apparent efficiency regressions from genuine technological stagnation or decline. Quantitatively, an ODEX value of 90 implies a 10% improvement in energy efficiency relative to the base year, conventionally set to 100. Energy savings attributable to efficiency gains can be derived from the ODEX using the expression (Lapillone (2020)):where \({ES}_{t}\) denotes energy savings and \({E}_{t}\) is the observed energy consumption in year t. This formulation permits the translation of index variations into absolute energy saving metrics, aligning with top-down evaluation methodologies. In line with ODYSSEE methodology the calculation is based on a sliding base year, which means that energy efficiency gains are measured in relation to the previous year.

Compared to conventional energy intensity indicators (e.g., energy per unit GDP or value added), the technical ODEX offers a more accurate and policy-relevant representation of energy efficiency, as it excludes structural and economic effects not directly associated with technical efficiency improvements. As such, ODEX represents one of the analytical tools for evidence-based policymaking, enabling the assessment of historical trends, benchmarking across member states, and informing future energy efficiency strategies in support of EU climate and energy objectives.

The analysis presents ex-ante estimates of energy savings, which are subsequently compared with ex-post savings data drawn from the ODYSSEE-MURE project, thereby illuminating the level of ambition anticipated for the 2022–2030 period.

For this study, average yearly energy savings in the industry, transport, household, and services sectors for the period 2015–2022 were obtained from the ODYSSEE database for both Slovenia and Croatia, fostering the ex-post component of the proposed approach. Since the database reports cumulative annual savings relative to the base year 2000, the total savings for the 2015–2022 period were calculated by subtracting the cumulative value from 2014 from that of 2022. This method offers a more accurate view of how energy efficiency policies implemented in the past have contributed to actual savings and serves as a foundation for evaluating the potential impact of future measures.

Estimating the effects of energy efficiency policies demands access to comprehensive and granular datasets. In particular, the ODEX index calculation relies on detailed energy consumption data that differentiate consumption by industrial subsectors, transport modes, specific end uses within the residential sector (e.g., space heating, water heating, and appliances), and various subsectors within the services sector. Alongside these energy consumption data, robust sectoral activity indicators are needed to calculate specific energy consumption and, ultimately, estimate the influence of efficiency measures on energy use trends.

For the industrial sector, production indices or value-added metrics can serve as viable proxies for activity. In transport, disaggregated data on passenger and freight activity per mode (e.g., road, rail, air, and maritime) are necessary to delineate energy consumption patterns. Residential energy consumption is typically correlated with building attributes (e.g., floor area, number of buildings, and size of households), while in services, indicators such as total floor area or employee counts are commonly used to capture variations in energy demand.

These requirements pose significant data collection and harmonization challenges, particularly when different countries employ a range of modeling approaches with varying levels of disaggregation and computational complexity. Within this context, the streamSAVE + project aims to improve the granularity and reliability of data used for national energy savings estimates. Slovenia and Croatia provided detailed data in support of this initiative, drawn in part from the modeling work conducted during the preparation of their National Energy and Climate Plans (NECPs). By adopting the ODEX methodology which permits flexibility in the selection of specific energy and activity indicators researchers can accommodate the heterogeneity in data availability across countries while preserving methodological consistency. Although variations in national datasets inevitably influence the results, the application of a harmonized methodology enables these effects to be more clearly identified and interpreted through the analysis of intermediate ODEX components.

Slovenia and Croatia were selected as case studies due to the availability and quality of the necessary input data. In addition, these countries provide a compelling comparative context given their shared historical background. This common starting point allows for a meaningful examination of their divergent energy efficiency trajectories and future policy plans within the framework of EU climate and energy targets.

To construct bottom-up estimates of energy savings, models must derive consumption levels from activity data in conjunction with assumptions about the effectiveness and coverage of various policies and measures. This bottom-up approach enables a fine-grained analysis of how sector-specific interventions (e.g., industrial process optimizations, building refurbishments, or modal shifts in transport) contribute to aggregate energy consumption trends.

In this study, the research team employed a standardized data template based on the ODYSSEE framework, specifying the necessary disaggregated consumption and activity data. This template also outlined various options for populating specific energy consumption indicators, granting countries flexibility to incorporate whichever indicators best reflected their available data.

The estimation of energy savings involved combining actual 2022 data with forward-looking projections for 2030, 2040, and 2050. Actual data for 2022 were extracted from the ODYSSEE database, while the projections stemmed from national modeling exercises (The Ministry of the Environment, 2024) and official planning documents. Detailed data on projected energy consumption and associated activity indicators were collected within the framework of the streamSAVE + project by national partners the Energy Institute Hrvoje Požar (EIHP) in Croatia and the Jožef Stefan Institute – Centre for Energy Efficiency (IJS-CEU) in Slovenia. The primary data sources consist of national energy projections developed for the preparation of the most recent National Energy and Climate Plans (NECPs), submitted in 2024/2025.

Merging these datasets can introduce discrepancies because the baseline years for projections differ, the underlying data collection methods vary in detail, and modelling conventions often diverge between ex-post observations and ex-ante forecasts. For instance, we identified differences in household energy consumption and related activity data for Croatia when comparing projections with statistical sources. However, since ODEX is based on intensities and structural shares rather than absolute values, the influence of these discrepancies is minimised. In the transport sector, detailed historical data on energy consumption by transport mode were not available; this was addressed by assuming the same modal structure in the projections as in the base year. To minimise inconsistencies and avoid unrealistic outcomes small adjustments were made either to the projections or to the historical data to improve consistency. This calibration step is important to ensure that the resulting ODEX estimates and assessed policy impacts reflect real-world trends and plausible expectations for future energy use.

The ODEX methodology enables the isolation of structural effects from energy savings calculations, thus providing more precise estimates of the aggregate impacts of energy efficiency policies and measures. It is true that ODEX includes the effect of price changes, autonomous technical progress and other market forces, but as scope and number of policies is broadened with more and more demanding targets, they influence also on these factors. The calculation procedure used in this study is outlined below:

Where \({UC}_{i,t}\) is the unit consumption index for subsector i at time t, and \({UC}_{i,t-1}\) is the corresponding index from the previous year. By inverting this ratio, it is possible to derive value at year t.

In this expression, \({ES}_{t}\) represents the estimated energy savings achieved between years t-1 and t, and \({E}_{t}\) denotes the total sectoral energy consumption at year t.

Data from Slovenia and Croatia were gathered to illustrate the application of the ODEX-based analytical framework under real-world conditions. Specifically, country-specific energy consumption and activity data were extracted to enable both cross-country comparison and sectoral analysis. The activity data used in this analysis were originally developed to support the preparation of the National Energy and Climate Plans (NECPs), based on national energy modelling frameworks. For Slovenia, the energy projections for NECP were produced by the Jožef Stefan Institute using the MESAP REES-SLO2 model, while for Croatia, the projections were developed by the Energy Institute Hrvoje Požar (EIHP) employing the LEAP (Long-range Energy Alternatives Planning) model. Table 2 provides detailed activity input data for Slovenia, encompassing multiple economic sectors such as industry, transport, households, and services. The values, expressed as a production index (industry), floor area in square meters and vehicle/transport kilometers serve as essential indicators for modelling and forecasting energy demand. Notably, certain industrial subsectors (e.g., food, beverage and tobacco) show missing figures, reflecting the ongoing challenges in gathering comprehensive and standardized data. Meanwhile, subsectors like paper, pulp and printing products or non-metallic minerals present clear upward trajectories, suggesting steady growth in production activities over the coming decades. Beyond industry, Table 2 illustrates a dynamic transport landscape, evidenced by the projected changes in the number of vehicles and vehicle-kilometers travelled. Similarly, the data point to moderate but consistent increases in housing surface area and the number of dwellings highlighting the pace of new construction and the potential need for energy efficiency upgrades. In services, expanding floor areas may indicate broader economic shifts and a growing demand for commercial or public-sector facilities.

Table 3 shows the activity input data for various sectors in Croatia. These figures reflect base year (statistical data for 2022) and projected values (2030, 2040 and 2050), thereby enabling analysts and policymakers to model energy consumption trends over time. The data in Table 3 also underscore the importance of tailoring policies to sectoral needs, highlighting, for instance, the distinct activity patterns in industry, or the contrasting transport modes of cars, buses, and motorcycles. Such insights are particularly valuable when integrating top-down approaches (e.g., macroeconomic analyses) with bottom-up data (e.g., energy audits at the facility level). By combining these perspectives, decision-makers gain a nuanced view of how each sector’s structure and growth trajectory can influence overall energy demand. This, in turn, assists in designing targeted interventions such as fiscal incentives, regulatory mandates, and information campaigns to stimulate greater uptake of efficiency solutions. Moreover, these projections inform national and EU-level planning, ensuring that energy and climate policies remain both responsive to evolving circumstances and aligned with long-term decarbonization goals.

Table 4 lists the energy input data for Slovenia’s industrial, transport, household, and services sectors, thereby offering an overview of current and projected energy consumption through 2050. The table is organized by subsector, illustrating both the diversity of energy requirements and the potential gaps in data availability (denoted as “NA”). For instance, while subsectors like paper, pulp and printing products and chemicals have clearly defined trajectories, other subsectors, such as textiles or machinery & metal products, show missing values for the near and long term. Such data gaps highlight the challenge of gathering sector-specific energy usage information, often due to methodological or reporting inconsistencies. In the Slovenian case, energy intensive industrial branches are modelled separately while other branches are modelled in an aggregate way, hence the lack of data for those industries. Beyond the individual figures, Table 4 also underlines several important trends. In the transport sector, for example, the transition away from traditional fuel use is visible in the declining total energy consumption for cars, from 1.26 Mtoe in 2022 down to just 0.24 Mtoe by 2050, suggesting widespread adoption of alternative mobility solutions. Meanwhile, the household sector demonstrates a noticeable shift in heating demands when climate corrections are taken into account, indicating potential efficiency improvements or a switch toward less energy-intensive heating technologies. For the services sector, although the total remains fairly constant, the data imply an ongoing need to optimize energy use as the economy evolves.

Table 5 presents Croatia’s sectoral energy consumption and projections through 2050. Industry shows reduced demand in several subsectors, like non-metallic minerals and primary metals, implying efficiency gains or structural changes.

In transport, cars drop from 1.46 Mtoe in 2022 to 0.50 Mtoe by 2050, a consequence of a shift toward cleaner mobility.

Household heating significantly decreases, due to more efficient technologies, while rising appliance and cooling needs could offset these savings. The services sector remains relatively stable, reflecting retrofits and updated building standards.

To reduce inconsistencies between historical (ex-post) data and future (ex-ante) projections, a technical ODEX was adopted. Although attempts were made to reconcile and adjust the 2022 actual data with the baseline assumptions in the projections, discrepancies could not be fully eliminated. In particular, negative energy savings could arise if an increase in specific consumption was recorded for a given subsector.

The technical ODEX addresses this by holding the unit consumption constant in cases where an increase is observed between consecutive years, thereby avoiding unrealistic negative savings values and providing a more stable estimate of technical energy efficiency gains.

Projections for Slovenia and Croatia (Fig. 1) indicate an overall decrease in total final energy consumption, with Slovenia exhibiting a marginal increase in 2025 relative to 2022. A more granular analysis of individual sectors reveals notable differences between the two countries. In Slovenia, industrial energy consumption steadily rises throughout the projection horizon, whereas transport consumption increases until 2025 and subsequently declines. By contrast, households and services both show continuous reductions. In Croatia, transport and household energy consumption decrease persistently across the full timeline, while the service and industrial sectors experience a modest increase between 2022 and 2025 before entering a downward trend.

Full size image

Taking into account the Article 4 of the Energy Efficiency Directive, Slovenia and Croatia have set their respective 2030 final energy consumption targets at 4320 ktoe (The Ministry of the Environment Climate and Energy, 2025) and 5880 ktoe (The Ministry for the Economy, 2025), as outlined in their National Energy and Climate Plans (NECPs). These projections suggest that both countries will meet these targets, a conclusion further illustrated by the sectoral energy consumption trajectories shown in the figures below.

In accordance with the Energy Efficiency Directive (EED), the final energy consumption target under Article 4 encompasses all energy delivered to industry, transport (including international aviation), households, public and private services, agriculture, forestry, fishing, and other final users. By contrast, it excludes consumption associated with international maritime bunkers, ambient energy, deliveries to the transformation and energy sectors, and losses from transmission and distribution, as specified in Annex A of Regulation (EC) No 1099/2008. Consequently, ambient heat was excluded in the calculations for Slovenia, whereas for Croatia, consumption related to ambient heat is included in the results. This was done because no separate data on ambient heat was available for Croatia, so results for Croatia on energy savings are mildly underestimated. Total and sector-specific energy savings for Slovenia and Croatia are presented in Fig. 2, revealing notable differences across the projection horizon. Figure 2 illustrates cumulative energy savings over time, benchmarked against the reference base year 2022, to highlight the aggregated impact of energy efficiency measures across sectors.

Full size image

In Slovenia, the industry sector shows the lowest total savings by 2050, followed by the services sector. In industry, savings increase gradually throughout the entire period, suggesting sustained implementation of energy efficiency measures over time. Conversely, the majority of savings in services occur after 2030, indicating a later surge in efficiency-related interventions or technological improvements in that sector. In the residential sector, over half of the total 2050 savings are achieved by 2030, reflecting early and aggressive measures such as building renovations and appliance upgrades. However, the rate of additional savings between 2040 and 2050 is comparatively small, largely because most buildings are already renovated by 2040, leaving limited potential for further efficiency gains. Meanwhile, the transport sector exhibits the largest absolute savings by 2050, with the majority of those gains concentrated between 2030 and 2040, due to widespread adoption of more efficient vehicles and shifts toward low-carbon transport modes during that interval. These findings underscore the importance of tailoring policies and measures to the specific dynamics and technological readiness of each sector.

In Croatia, the household sector is projected to achieve the highest overall energy savings by 2050, followed by transport, industry, and services. Notably, most household sector savings occur between 2030 and 2040, indicating a later onset or slower initial uptake of efficiency measures compared to Slovenia.

In the transport sector, although Croatia exhibits higher overall energy consumption, it achieves lower absolute energy savings compared to Slovenia. This discrepancy may indicate that Croatia’s policy measures or implementation strategies are either less ambitious or encounter more substantial barriers. These differences are further reflected in the underlying activity data. In Slovenia, the projected activity of public transport modes—particularly buses and rail—shows a marked increase over time, suggesting a shift toward more efficient mobility solutions. In contrast, public transport activity in Croatia remains largely constant throughout the projection period. Additionally, while Slovenia projects a decline in car usage after 2025, car activity in Croatia remains relatively stable, highlighting divergent approaches to modal shift and transport demand management. A comparison of savings as a percentage of sectoral consumption further emphasizes these differences. By 2050, the transport sector in Croatia realizes approximately 105% savings relative to its consumption in 2050, whereas Slovenia reaches 135%, underscoring Slovenia’s wide-ranging policy initiatives. Croatia mainly relies on technological solutions as no structural change is visible in transport activity, while Slovenia envisages implementation of measures changing modal split, reducing need for transport and also technological measures.

Conversely, Croatia’s industrial sector exhibits substantially higher projected energy savings, amounting to approximately 61% of its sectoral consumption by 2050, compared to Slovenia’s 16%. This notable divergence may, at least in part, be attributed to differences in the choice of activity indicators used in the ODEX calculation. For Croatia, value added is used as a proxy for industrial activity, whereas Slovenia employs a production index. Since value added typically grows faster than production indices—particularly in economies undergoing structural transformation—this methodological choice can significantly influence the magnitude of calculated energy savings. Selection of activity indicators can affect the interpretation of efficiency trends, as higher growth in the denominator (activity) leads to more pronounced reductions in specific energy consumption, and thus greater apparent efficiency gains. In addition to methodological factors, actual energy consumption trajectories also differ between the two countries. In Slovenia, industrial energy use is projected to increase by 16% between 2022 and 2050, whereas in Croatia it is expected to decline by approximately 18% over the same period. This reinforces the importance of considering both modelling inputs and real consumption patterns when interpreting cross-country comparisons of energy efficiency outcomes.

In the services sector, both countries exhibit similar savings levels when measured against overall consumption, indicating a comparable degree of policy penetration and effectiveness in that domain.

Table 6 presents the sectoral relative ODEX index for Slovenia and Croatia. In Slovenia, energy efficiency by 2050 is expected to improve by 64% in transport, 47% in households, 41% in services, and 15% in industry compared with 2022. In Croatia, the projected improvements are 61% in households, 52% in transport, 40% in industry, and 38% in services. The notably low improvement in Slovenian industry stands out, with very limited progress expected between 2030–2040 and 2040–2050. The higher projected efficiency gains in Croatian industry relative to Slovenia can be partly explained by the choice of activity indicators: Croatia uses value added, whereas Slovenia relies on a production index.

Important insight can also be gained by comparing historical savings (ex-post) with projected ones (ex-ante). Figure 3 provides a comparison between historical and projected energy savings in Slovenia and Croatia.

Full size image

Average annual energy savings were calculated by dividing the total savings over each period by the number of years within that period. For the historical period 2015–2022, savings were derived from the ODYSSEE database by subtracting the cumulative energy savings recorded in 2014 from those reported in 2022.

This comparison shows a substantial increase in the required efficiency improvements, indicating that both countries would need to accelerate their energy-efficiency gains significantly in order to meet their 2030 final energy-consumption targets. Achieving such increases will likely require the implementation of additional policies and measures beyond those currently in place. The results of this study indicate that, for both Slovenia and Croatia, achieving the projected reductions in final energy consumption would require energy-efficiency gains to roughly double compared with the current historical levels.

An exception to this trend is observed in Slovenia during the 2041–2050 period, where the rate of annual savings declines, likely reflecting the saturation of cost-effective energy efficiency potential in most sectors.

In both countries, the transport sector is projected to contribute significantly to future savings, contrasting sharply with the minimal savings achieved in this sector during the historical reference period.

In Slovenia, average annual savings in the household sector gradually decline over time, primarily due to the near-completion of building renovation efforts. In contrast, savings in the services sector remain relatively stable, with the exception of the 2023–2025 interval, which reflects inconsistencies between projected values and actual 2022 data.

For Croatia, the household sector emerges as the dominant contributor to average annual savings throughout the forecast period, apart from a slight anomaly in 2023–2025, which is also attributed to data inconsistencies between observed and projected baseline values.

The ODEX-based assessment of future energy efficiency policies and measures offers a comprehensive view of potential energy consumption trajectories in the absence of these interventions. By comparing projected energy use with and without savings (policy and non-policy induced, e.g. autonomous technical progress, market induced progress), one can quantify how significantly policy actions influence energy demand. In Slovenia, for instance, failing to implement the planned measures would elevate energy consumption by approximately 19% in 2030, 48% in 2040, and 67% in 2050 as shown in Fig. 4. For Croatia, the corresponding increases would be around 17%, 50%, and 88% in the respective years.

Full size image

These findings highlight both the pivotal role and the substantial impact of energy efficiency strategies in meeting national targets under Article 4 of the Energy Efficiency Directive (EED). Notably, 2030 consumption projections for both countries closely track established target values, underscoring the necessity of effective and timely policy execution. Any delay or underperformance in implementing energy efficiency measures could jeopardize compliance with those targets and diminish the broader contributions to EU climate objectives.

Moreover, these results illustrate the scale of potential missed opportunities, both in terms of energy savings and emissions reductions, if policy ambitions are not fully realized. Ensuring robust financing mechanisms, enhancing stakeholder engagement, and aligning measures across multiple sectors will be critical to achieving, and possibly exceeding, the modelled outcomes. As such, sustained commitment to ongoing monitoring, data refinement, and iterative policy development remains essential to maximize the effectiveness of energy efficiency policies and to maintain alignment with the National Energy and Climate Plans (NECPs).

As shown in Fig. 5, the projected energy consumption in the industrial sector for Slovenia (left) and Croatia (right) from 2022 to 2050 is contrasted under scenarios with and without policy measures. The results suggest that targeted efficiency improvements, such as process optimizations and technology upgrades can yield substantial savings, though the extent of these reductions varies between the two countries.

Full size image

Figure 6 highlights the transport sector, where projections point to significant energy-saving opportunities through more efficient vehicles, electrification, and modal shifts. Particularly notable gains are observed by the mid-2030s, underscoring the value of continued policy support and technological advancement in this sector.

Full size image

Figure 7 focuses on the household sector, illustrating how building renovations, higher appliance efficiency, and awareness campaigns collectively drive considerable savings. While Slovenia demonstrates earlier gains, Croatia exhibits a sharper decline in consumption between 2030 and 2040, suggesting differences in policy timing or market uptake across the two countries.

Full size image

Whereas, Fig. 8 depicts energy consumption in other sectors, such as services, comparing scenarios both with and without policy measures. Although absolute savings here are smaller than those in the transport or household sectors, the gradual yet substantive reductions emphasize the importance of sustained, sector-specific interventions in conjunction with broader efficiency policies.

Full size image

The Slovenian and Croatian case studies provide valuable insights into both the strengths and limitations of implementing energy efficiency (EE) measures under real-world conditions. On the one hand, the results highlight that strong policy commitments, such as those embedded in National Energy and Climate Plans, can drive meaningful gains in multiple sectors. For instance, the gradual yet consistent improvements in Slovenia’s industrial sector and Croatia’s pronounced savings in industry demonstrate how targeted interventions (e.g., process optimizations, technology upgrades, and financial incentives) can yield substantial efficiency dividends. Similarly, the robust implementation of residential EE measures anticipated in both countries underscores the effectiveness of building renovation programs, appliance standards, and awareness campaigns.

On the other hand, several challenges emerge from these case studies, emphasizing the nuanced and context-dependent nature of EE policies. Chief among these are data availability and consistency issues, which can impede accurate ODEX calculations and generate discrepancies between ex-post and ex-ante estimates. In Croatia, for example, the delayed realization of household-sector savings relative to Slovenia suggests that local market barriers, institutional capacity, and consumer preferences may slow policy uptake. Meanwhile, Slovenia’s comparatively modest industry savings may point to gaps in its industrial policy framework or implementation effectiveness, notable divergence could also be attributed to differences in the approach to the national energy projections, where the use of value added as a leading model parameter for industrial energy projections can yield significantly higher energy savings. These differences underscore the importance of tailoring EE strategies to sector-specific realities, strengthening policy coordination, and improving data granularity. By addressing these challenges, policymakers and stakeholders can enhance the accuracy of ODEX-based evaluations and ultimately foster more ambitious, evidence-based energy efficiency interventions.

One of the key advantages of the ODEX-based methodology lies in its adaptability and relevance to a broad range of European countries, each characterized by distinct economic structures, policy priorities, and data-reporting practices. By employing standardized metrics for specific energy consumption, the method ensures that observed or anticipated efficiency improvements can be meaningfully compared, even when differences in climate, industrial composition, and transport patterns exist. This comparability facilitates both cross-country benchmarking and the possibility of aggregating national results into broader regional or EU-wide analyses, thus supporting the formulation and evaluation of coherent, pan-European strategies.

Nevertheless, the successful application of the approach across diverse national settings hinges on the availability of sufficiently granular and high-quality data. Many EU Member States face challenges in collecting, harmonizing, and maintaining comprehensive energy and activity datasets, particularly at the subsector level (e.g., detailed transport modes or specific manufacturing branches). Data gaps can lead to incomplete ODEX indicators and may undermine the reliability of resulting energy savings estimates. In addition, variations in modelling approaches, reporting conventions, and institutional capacities introduce further complexity, necessitating careful reconciliation of input assumptions and methodological choices.

Addressing these issues requires improved data-sharing platforms, standardized reporting guidelines, and enhanced technical assistance to countries with less extensive data infrastructures. As the EU intensifies its efforts under the Energy Efficiency Directive and the European Green Deal, ongoing collaboration and capacity-building among national statistical agencies, research institutions, and policy stakeholders will be crucial. By continually refining data collection processes and aligning analytical methods, the ODEX-based framework stands to offer a scalable, consistent tool for monitoring progress and informing evidence-based policies across diverse European contexts.