Measuring the Circular Economy at the Macro Level: Do Developed and Developing Countries Differ?

2.1 Conceptual foundations of the circular economy

The CE provides a framework for rethinking production and consumption under resource, climate, and waste pressures. It challenges the linear model of extraction, production, consumption, and disposal, and promotes the retention of material, component, and product value for as long as possible. Despite its growing policy relevance, CE remains conceptually diverse because different academic, policy, and business communities emphasize different objectives and mechanisms [27-29].

A common conceptual foundation is that CE is not limited to recycling. Kirchherr et al. [8] showed that CE definitions usually combine reduce, reuse, and recycle strategies and are often linked to sustainable development. These strategies can slow resource loops by extending product lifetimes, close loops through recycling and recovery, and narrow loops by reducing material intensity. The hierarchy embedded in R frameworks also matters: reduction is usually preferred to reuse, and reuse is preferred to recycling. Recycling remains important, but it may be less circular when it leads to quality loss or downcycling [30].

The systems perspective is equally important. CE requires structural transformation rather than isolated efficiency improvements. It operates across micro, meso, and macro levels, involving changes in product design, firm practices, industrial networks, national infrastructure, and policy architecture [8, 31]. This multi-level nature means that circular performance must be measured in a way that matches the level of analysis.

CE is also closely connected to sustainable development. The sustainable development tradition emphasizes the integration of economic progress, social equity, and environmental protection [32-34]. CE can be viewed as an economic model for translating this ambition into resource and production systems. However, CE definitions often emphasize environmental and economic dimensions more strongly than social equity [8]. This creates a measurement challenge because a narrow resource-efficiency index may overlook wider distributional and welfare implications.

For this study, the relevant implication is clear. A macro level CEI should not be a simple recycling measure. It should capture system-level conditions that reflect resource efficiency, ET, carbon pressure, and material dependence. It should also be comparable across countries with different development levels. These conceptual foundations guide the choice of indicators and the construction of the index.

2.2 Existing approaches to measuring the circular economy

CE measurement is commonly organized across micro, meso, and macro levels [31, 35, 36]. Micro-level approaches examine products, firms, business models, and life-cycle stages. Meso-level approaches focus on industrial sectors, eco-industrial parks, supply chains, and regional systems. Macro level approaches assess national or regional circularity through resource productivity, waste intensity, recycling, emissions, EE, and material flow indicators.

Micro-level indicators provide detailed tools for product and firm assessment. Examples include the DEI, End-of-Life Index, Recycling Indicators (RI) for electronic products, the MCI, the Circular Economy Indicator Prototype (CEIP), and life-cycle based economic-environmental measures [9-12, 37-39]. These tools are useful for design, recovery, and firm level decision making, but they require detailed technical or corporate data and are difficult to scale to national comparison.

Meso-level indicators capture interactions among firms, sectors, and regions. Studies of chemical enterprises, eco-industrial parks, resource productivity, wastewater systems, agri-food systems, and value-chain circularity show how circular gains can emerge from coordination and industrial symbiosis rather than from isolated firms [13, 20, 40-44]. These approaches are policy relevant, but they remain sectoral or territorial and cannot directly answer how countries compare at the macro level.

Macro level measurement has developed through two main streams. The first stream constructs composite indicator systems for countries, provinces, or regions, as shown in the work of Qing et al. [18], Geng et al. [19], Claudio-Quiroga and Poza [23], and Drofenik et al. [24]. The second stream relies on economy-wide material flow accounting and related approaches to track domestic material consumption, processed outputs, stock accumulation, and circularity rates [25, 45-47]. These studies provide important foundations but are often tied to specific regions, data systems, or material flow accounting requirements.

Recent reviews confirm that CE indicators remain diverse and fragmented. Moraga et al. [48] showed that indicators differ by system function, product focus, material flow boundary, energy focus, and scenario design. Parchomenko et al. [49] identified separate clusters of resource-efficiency, material-flow, and product-oriented measures. De Pascale et al. [50] classified 61 indicators by spatial level and circular principles, while de Oliveira and Oliveira [51] found that many indicators do not fully reflect circular strategies or sustainability goals. These findings suggest that indicator choice strongly shapes what is understood as circular performance.

The literature therefore supports the need for a parsimonious macro level index that can be applied across countries with heterogeneous data systems. Table 1 summarizes key indicators and measurement systems from previous empirical studies. Rather than reproducing every prior approach, the present study builds on this literature by selecting a small set of internationally comparable indicators that capture systemic circular performance at the national level.

2.3 Measurement gaps in cross-country comparison

Table 1 shows that CE measurement has expanded rapidly, but the increase in indicators has not solved the problem of cross-country comparability. Many indicators are designed for products, firms, sectors, or regions. They are useful within their own contexts, but they are not designed to compare developed and developing economies under a common macro level framework.

At the micro and meso levels, tools such as the MCI, the DEI, and life-cycle based assessment methods help identify improvement opportunities within firms and products [9, 14, 37]. Evidence from Vietnam also shows that such tools can diagnose firm level material flows [16]. However, they cannot show where a national economy stands in circular transition or how it compares with other countries.

At the macro level, existing approaches are often region specific or data intensive. European studies benefit from Eurostat and harmonized monitoring frameworks [23]. Material flow accounting studies provide rich information on inputs, outputs, stocks, and recycling [25, 46, 47]. Yet these approaches are difficult to convert into a simple, standardized benchmark for countries with different development levels and uneven data quality.

A further problem is inconsistency in measurement logic. CE indicators may represent resource productivity, material flows, product design, recycling, waste treatment, innovation, or social sustainability [48-50]. Without a clear objective and comparable data structure, composite indices may become mechanical aggregations rather than analytically meaningful measures.

This review reveals a clear gap. There is no macro level CEI explicitly designed to compare developed and developing countries within a unified framework. Such an index must balance conceptual relevance and data feasibility. It must capture the essence of circularity, standardize indicator direction and scale, use transparent weighting, and recognize structural differences in technology, infrastructure, finance, and institutional capacity.

2.4 Structural differences between developed and developing economies

A comparison between developed and developing economies must account for structural heterogeneity. Circular transition reflects economic structure, institutional capacity, technology, infrastructure, and resource management systems. It is not only a matter of recycling or waste treatment [8, 47]. Ignoring these differences may distort cross-country interpretation.

Developed economies generally have more diversified industries, stronger environmental regulation, formal waste management systems, better secondary material markets, and more reliable statistical systems [23, 46, 52]. These conditions support higher circular performance and make national indicators easier to measure consistently.

Table 1. Summary of circular economy (CE) measurement indicators

Developing economies often depend more heavily on primary resource extraction, rapid infrastructure expansion, and informal recovery systems [25, 47]. Informal recycling may be important but poorly recorded. Financial constraints, weaker regulatory enforcement, and limited technological capacity can also reduce the effectiveness of material recovery systems [49].

These differences mean that the same indicator may carry different meanings across country groups. High material inflows in developing economies may reflect industrialization and urbanization, while high material consumption in developed economies may reflect saturated but resource-intensive lifestyles. For this reason, a macro level CEI must use broadly available indicators, apply common normalization and aggregation procedures, and remain conceptually aligned with CE principles [8, 47, 48, 50].

Recognizing structural heterogeneity is therefore central to the present study. The objective is not merely to generate a ranking, but to examine whether developed and developing economies display systematically different circular patterns when assessed under a unified and transparent macro level framework. By embedding the measurement design within an explicit understanding of development heterogeneity, this study aims to produce a more balanced and analytically meaningful comparison of CE performance across country groups.

4.1 Descriptive statistics and preliminary patterns

Table 2 reports the summary statistics of the four CE indicators for 109 countries. The results show considerable cross-country variation, suggesting that countries differ strongly in their level of CE performance.

Table 2. Summary statistics of circular economy (CE) indicators

EE has the highest mean value, at 12.631, and also the largest dispersion, with a standard deviation of 5.8574. Its values range from 1.8850 to 48.2422. This wide range indicates large differences in the ability of countries to generate economic output from energy inputs. Some economies use energy much more efficiently, while others remain highly energy intensive.

ET records a relatively low mean value of 0.0989, suggesting that renewable energy still accounts for a small share of total energy consumption in most countries. However, the maximum value of 0.5305 shows that some countries have made stronger progress in shifting toward renewable energy systems.

CAE has a mean value of 0.1940, with values ranging from 0.0157 to 1.4950. This variation indicates that carbon intensity differs substantially across countries. Economies with higher CAE values generate more carbon emissions per unit of output, while those with lower values show better CAE.

RE also shows strong variation across countries. Its mean value is 2.5178, with a standard deviation of 2.1736, and values ranging from 0.1029 to 16.6011. This suggests that the material intensity of production differs widely. Some economies depend heavily on material inputs to generate output, while others use materials more efficiently.

Figure 2 reports the pairwise correlations among the CE indicators. The results suggest moderate to strong associations between the variables, indicating that they capture related but distinct dimensions of CE performance.

A strong negative correlation is observed between EE and ET (r = -0.7722), suggesting that countries with higher energy productivity do not necessarily exhibit higher shares of renewable energy. This may reflect structural differences between economies, where improvements in EE are driven by technological optimization within existing energy systems rather than a transition toward renewables.

CAE is moderately positively correlated with ET (r = 0.4451), indicating that countries with higher renewable energy adoption tend to exhibit lower carbon intensity. This relationship is consistent with CE principles, as the shift toward cleaner energy sources contributes to reducing environmental pressure.

In contrast, CAE shows a moderate negative relationship with EE (r = -0.4519), suggesting that improvements in energy productivity may not always coincide with reductions in carbon intensity. This highlights the possibility that efficiency gains are achieved within carbon-intensive systems, particularly in developing economies.

RE exhibits relatively weak to moderate correlations with the other variables. Its near-zero correlation with CAE (r = 0.0300) suggests that material use intensity and carbon intensity may evolve independently across countries. Meanwhile, the moderate positive correlation between RE and ET (r = 0.3940) and the negative correlation with EE (r = -0.3743) indicate that material use patterns are partially linked to both ET and efficiency dynamics, but remain a distinct dimension of CE performance.

Table 3 reports the Kaiser-Meyer-Olkin (KMO) measure and Bartlett's test of sphericity. The overall KMO value is 0.762, indicating good sampling adequacy. Bartlett's test is also highly significant (chi-square = 3641.76, p < 0.001), rejecting the null hypothesis that the correlation matrix is an identity matrix. These results confirm that the variables are suitable for PCA.

Table 3. Kaiser-Meyer-Olkin (KMO) and Bartlett’s test of sampling adequacy

4.2 Principal Component Analysis results

Table 4 and Figure 3 report the PCA results. The first principal component (PC1) has an eigenvalue of 2.3149 and explains 57.87% of total variance. It is the only component with an eigenvalue greater than 1, and the scree plot shows a clear elbow after PC1. PC1 is therefore retained as the composite CEI.

Table 4. Principal Component Analysis (PCA) results

Figure 3. Scree plot of eigenvalues

Table 5 presents the factor loadings for PC1. All four indicators load positively on the retained component, confirming that PC1 captures a common CE dimension.

ET and EE have the highest loadings, at 0.5939 and 0.5919. CAE* and RE* also contribute positively, with loadings of 0.4126 and 0.3559. This means that the index is driven mainly by energy transition and energy productivity, while also capturing lower carbon intensity and lower material pressure.

Table 5. Factor loadings from Principal Component Analysis (PCA)

4.3 Classification of countries by circular economy performance

Figure 4 presents the CEI heatmap for 109 countries over 2002-2024. The index ranges from 0 to 1, with higher values indicating stronger circular performance. The color pattern shows a clear hierarchy across countries, with high-performing economies concentrated in the upper range and low-performing economies concentrated in the lower range.

The first thing the figure makes clear is that cross-country differences in CE development are large. The top part of the heatmap is dominated by advanced economies such as Sweden, Denmark, Norway, the United Kingdom, Switzerland, the Netherlands, Israel, Finland, France, Germany, Luxembourg, and Austria. These countries remain in dark green shades for most years. This suggests that their CE performance is not only high, but also stable over time. This pattern may be partly associated with stronger technological capacity, more developed environmental regulation, higher EE, and greater progress toward green growth in many developed economies.

The middle group looks more mixed. Countries such as Spain, Belgium, New Zealand, Ireland, Singapore, Slovenia, Italy, Hungary, Greece, Estonia, China, Poland, Croatia, Chile, Romania, Brazil, and Malaysia mostly appear in light green to pale yellow shades. This means they perform at moderate to fairly good levels, but they do not belong to the leading group. In practical terms, these countries seem to have made some progress in energy use, resource management, and environmental pressure reduction, but the transition is still less deep and less consistent than in the top tier.

Another useful insight from the figure is that it captures not only the level of CE performance, but also how it changes over time. In several countries, the colors gradually move from yellow or orange to lighter green in later years. This points to some improvement in the CEI. But the pace of improvement is uneven. A few countries improve only slightly, while many others keep roughly the same relative position throughout the sample period. This suggests that the shift toward a CE is likely to be a long-term structural process. Its pace may be influenced by underlying economic and institutional conditions.

At the lower end of the heatmap, countries such as Tanzania, Cameroon, Myanmar, Mozambique, Ethiopia, Zimbabwe, Mongolia, Uganda, Burkina Faso, and Madagascar remain in red or dark red shades for much of the study period. This indicates low CE performance and slow progress over time. From a practical point of view, this is not surprising. Many of these economies face common structural constraints, including weak technological capability, poor infrastructure, heavy dependence on raw material extraction, lower EE, and limited institutional capacity. In simple terms, these countries do not yet have the conditions needed to support a strong circular transition.

The figure also reveals some short run fluctuations in specific countries, shown by isolated color changes that differ from the overall pattern of a row. These shifts may reflect temporary changes in energy policy, economic shocks, changes in production structure, or adjustments in resource and CAE. Still, these short-term variations do not change the broader picture. Relative positions remain fairly stable. The leading group stays near the top, and the lagging group stays near the bottom for most of the period.

The figure offers three main insights. First, CE performance at the country-level is strongly stratified, with developed economies holding a clear advantage. Second, improvement does occur, but it is generally slow, which shows that circular transition depends on existing institutional quality, technological capability, and economic structure. Third, the gap between the leaders and the laggards remains wide, which suggests that the global transition toward a CE is still uneven and closely tied to differences in development levels.

Figure 5 presents the classification of 109 countries into three groups based on their CEI in 2024, namely high CE, medium CE, and low CE. The classification is based on tertiles of the 2024 CEI distribution. Countries in the top third of the distribution are classified as high CE economies, those in the middle third as medium CE economies, and those in the bottom third as low CE economies. This tertile-based approach provides a transparent and reproducible way to divide countries into relative performance groups. The figure shows a clear ranking pattern, with countries arranged from the highest to the lowest CE score. This visual structure makes it easy to observe both the distribution of countries across groups and the size of the gap between leading and lagging economies.

At the top of the ranking, the high CE group is dominated by advanced economies such as Sweden, Denmark, Norway, the United Kingdom, Switzerland, the Netherlands, Finland, France, Germany, and Austria. These countries record CE scores close to the upper bound of the index, indicating strong CE performance. Their position suggests that they have built more mature systems for resource efficiency, environmental governance, and green transition. In practical terms, these countries are better able to combine economic activity with lower resource pressure and stronger sustainability outcomes.

The medium CE group includes a broad set of countries with moderate but uneven performance. This group covers economies such as Latvia, Thailand, Mexico, the United Arab Emirates, Colombia, Viet Nam, South Africa, India, Uruguay, and Panama, among others. Their CE scores indicate that some progress has been made, but the transition remains incomplete. These countries are neither at the frontier nor at the bottom. Instead, they represent transitional cases where CE development is present but not yet deep, stable, or comprehensive.

Figure 4. Heatmap of circular economy index (CEI) across 109 countries, 2002–2024

Figure 5. Distribution of 109 countries across three circular economy (CE) groups in 2024

At the bottom of the distribution, the low CE group consists mainly of developing and lower income economies such as Angola, Nigeria, Bolivia, Honduras, Pakistan, Bangladesh, Nepal, Rwanda, Tanzania, Cameroon, Mozambique, Ethiopia, Zimbabwe, Uganda, Burkina Faso, and Madagascar. Their low CE scores suggest limited progress in circular transition. This pattern may reflect structural constraints such as weak infrastructure, lower technological capability, heavy dependence on primary resource use, and weaker institutional capacity to support circular practices.

An important feature of the figure is the clear separation between groups. The high CE group occupies the upper part of the ranking with relatively compressed but high scores, while the medium CE group forms a broad middle layer. In contrast, the low CE group is concentrated at the lower end of the scale, with several countries recording very low index values. This indicates that CE performance across countries is not evenly distributed, but strongly stratified.

The figure suggests three main points. First, CE development remains highly uneven across countries. Second, advanced economies continue to dominate the top positions, indicating that economic development, institutional quality, and technological capability likely matter for CE performance. Third, the wide gap between the high and low groups implies that the global transition toward a CE is still incomplete and marked by substantial cross-country disparities.

4.4 Comparative analysis: Developed vs developing economies

Table 6 shows a clear difference in CE performance between developed and developing economies. The mean CEI is 0.864 for developed economies, compared with 0.413 for developing economies. The gap of about 0.45 points is large. This suggests that developed economies, on average, have moved further toward more resource-efficient and sustainable production systems.

Table 6. Descriptive statistics of the circular economy index (CEI) by development group

The standard deviation also differs between the two groups. Developed economies have a lower standard deviation of 0.110, while developing economies record a higher standard deviation of 0.216. This means that developed economies are not only stronger on average, but also more consistent in their CE performance. By contrast, developing economies show wider variation. Some developing countries have made visible progress, while others remain at a much lower level of circular transition.

The minimum and maximum values further support this pattern. In the developing group, the CEI ranges from 0.000 to 0.920, showing a very wide spread in performance. This indicates that CE development within developing economies is highly uneven. In the developed group, the index ranges from 0.301 to 1.000. Although the weakest developed economy still records a relatively low score, the group as a whole remains much stronger, as shown by its higher mean and lower dispersion.

The table indicates that developed economies have achieved a higher and more stable level of CE performance, while developing economies remain more fragmented. One possible explanation is that developed countries generally have stronger institutions, better environmental regulation, more advanced technology, and greater financial capacity to support circular practices. Developing economies, by contrast, often face deeper constraints in infrastructure, governance, financing, and formal resource management, which may slow their transition toward a CE.

4.5 Robustness of group differences

The group comparison confirms a strong difference between developed and developing economies. Levene's test rejects equality of variances at the 1 percent level, indicating different dispersion across groups and supporting the use of Welch's t test (Table 7).

Table 7. Robustness tests for differences in the circular economy index (CEI) between developed and developing economies

Welch's t-test shows a highly significant mean difference, with a statistic of -14.111 and a p-value below 0.001. The average CEI is 0.413 for developing economies and 0.864 for developed economies, implying a gap from 0.466 to 0.451. This is both statistically significant and substantively large.

The Mann-Whitney U test gives the same conclusion, with a p-value below 0.001. The result is therefore robust even without relying on normality assumptions. CE performance is systematically higher in developed economies than in developing economies.

Figure 6 shows that developed economies remain ahead of developing economies throughout 2002-2024. The CEI for developed countries declines slightly from around 0.88 at the beginning of the period to nearly 0.85 in 2024, while the index for developing countries falls from about 0.44 to about 0.39. The gap remains large and shows no clear sign of narrowing.

79 developing countries

30 developed countries

Figure 6. Circular economy index (CEI) over time, 2002–2024

4.6 Discussion

The findings support the central argument that CE performance differs substantially between developed and developing economies. This result is consistent with previous studies showing that circular transition is shaped by structural conditions rather than by recycling performance alone. For example, Kirchherr et al. [8] emphasized that CE requires systemic changes in production, consumption, reuse, and resource efficiency, while Haas et al. [47] showed that material flows and resource dependence remain central barriers to achieving a more circular global economy. In the same direction, Mayer et al. [46] argued that economy-wide circularity depends on the ability to close material loops, reduce input pressure, and improve resource productivity. The present study extends these insights by showing that such structural differences are not only visible in individual indicators, but also appear clearly in a composite macro level CEI.

The higher CEI scores of developed economies are broadly consistent with studies on CE measurement in Europe and other advanced economies. Claudio-Quiroga and Poza [23] found large differences in CE performance among European countries, reflecting differences in institutional capacity, technology, and policy implementation. Similarly, Drofenik et al. [24] showed that national CE progress depends on the alignment between policy targets and implementation capacity. The results of the present study support this view, but they also add a wider comparative perspective by showing that the gap is even more visible when developed and developing economies are assessed under the same measurement framework. Developed economies tend to perform better, possibly because they often have stronger technological capacity, more advanced environmental regulation, better data systems, and stronger institutional support for circular transition.

At the same time, the results should not be read as a simple success story for developed economies. Many developed countries still have high levels of material consumption. This point is consistent with previous research arguing that high-income economies may perform well on recycling, EE or environmental regulation while still maintaining resource-intensive consumption patterns. In this sense, stronger CEI scores reflect relative advantages in efficiency and transition capacity rather than a complete circular transformation. The index developed in this study helps reveal this tension by combining CAE, EE, ET, and RE in one framework. This is important because a country can perform well in one circular dimension while remaining weak in another.

For developing economies, the findings are also consistent with prior studies suggesting that CE implementation may be constrained by infrastructure, finance, technology, and institutional quality. Emami et al. [25], in the ASEAN context, showed that CE progress remains uneven because material consumption, waste systems, and circularity rates differ strongly across countries. Parchomenko et al. [49] and De Pascale et al. [50] also noted that CE indicators often fail to capture informal recovery systems and local circular practices. This is particularly relevant for developing economies, where informal reuse, repair, and recycling may be significant but poorly recorded in official statistics. Therefore, lower CE scores do not necessarily mean that these economies lack circular practices. Rather, they indicate that informal practices have not yet been transformed into formal, scalable, and measurable circular systems.

Compared with previous studies, the main contribution of this study is that it provides a unified macro level comparison between developed and developing economies using the same indicators, the same normalization procedure, and the same PCA-based weighting structure. Earlier studies have often focused on specific regions, specific material flows, firm-level indicators, or European data systems. This study moves the discussion forward by offering a simple and internationally comparable index that can identify broad development gaps in CE performance. The evidence suggests that developing economies need stronger circular infrastructure, better environmental data systems, targeted green finance, cleaner technology transfer, and institutional reforms that can transform informal circular practices into more efficient and scalable circular systems.