The research, led by Abdikafi Hassan Abdi and colleagues, addresses a long-standing puzzle in the environmental-development literature. The dominant paradigm, the Environmental Kuznets Curve, predicts that environmental degradation rises during early development before technological progress, structural transformation, and regulatory maturity reverse the trend. But tests of this proposition typically use emissions or the ecological footprint as the outcome of interest. These indicators track the pattern of environmental pressure without establishing whether the demand that remains is sustainable relative to available biocapacity. Declining footprints may therefore coexist with continued ecological overshoot, a distinction that reflects the divide between weak and strong sustainability described by Daly.

The load capacity curve addresses this limitation through the load capacity factor, defined as biocapacity divided by ecological footprint. A value above unity indicates biocapacity-reserve, while a value below unity indicates ecological deficit. The load capacity curve predicts a U-shaped relationship between income and this factor. Following the scale, composition, and technique decomposition described by Grossman and Krueger, scale effects dominate at low income, as industrialization, urbanization, and resource extraction outpace efficiency gains. Rising income then shifts composition and technique toward recovery, through cleaner energy, innovation, and stronger institutional capacity. Yet the empirical evidence for this nonlinear relationship is consistent in direction but not uniform in shape or timing across countries, with U-shaped, inverted-U, and country-specific patterns reported across the literature.

The authors contend that these inconsistent conclusions may reflect more than differences in samples or estimators. They may result from combining economies operating under fundamentally different ecological constraints. In deficit economies, additional production and consumption occur within an existing condition of ecological overshoot. When expanding resource use, energy demand, and waste outweigh improvements in technology and efficiency, economic growth can deepen that imbalance. Restoring balance consequently requires environmental gains sufficient to offset both new pressure and the existing excess of demand over biocapacity. Surplus economies retain a regenerative margin, but continued growth in ecological demand or deterioration in biocapacity can progressively erode that reserve. Ecological status can therefore change over time and should not be treated as a permanent national characteristic.

To test this proposition directly, the study employs an annual unbalanced panel of 133 countries covering 2013 to 2021. Countries are classified as ecological-deficit or biocapacity-reserve based on their annual load capacity factor, with country-year observations below unity classified as deficit and those at or above unity as surplus. Because classification is annual rather than fixed, six countries, Botswana, Ecuador, Honduras, Kazakhstan, Sudan, and Ukraine, change regime at least once during the sample period and are represented in both groups. The regression sample comprises 95 deficit and 43 surplus countries. The variables are transformed into natural logarithms to reduce scale differences, stabilize variance, mitigate heteroskedasticity, and permit the estimated coefficients to be interpreted as elasticities.

The econometric strategy is deliberately multi-estimator. Second-generation panel unit-root testing permits cross-sectional dependence among globally connected economies, which the Pesaran CD and power-enhanced CDw+ statistics confirm for every variable at the 1 percent level. Slope homogeneity is assessed using the Pesaran-Yamagata test and its heteroskedasticity- and autocorrelation-consistent counterpart, neither of which rejects slope homogeneity. Baseline fixed-effects and random-effects models provide the initial estimates, while panel-corrected standard errors and feasible generalized least squares assess robustness to cross-sectional dependence, heteroskedasticity, and serial correlation. Full-sample difference GMM incorporates lagged load capacity factor and internal instruments to examine persistence and potential endogeneity, with the Hansen test and Arellano-Bond serial correlation tests supporting instrument validity and model specification.

The results reveal marked differences in the determinants of load capacity across ecological regimes. In deficit economies, GDP is negative and statistically significant, whereas its squared term is positive and significant across all baseline specifications. This coefficient pattern describes a convex income-load capacity relationship. However, the implied turning points exceed the maximum income observed in the deficit sample. The results therefore do not indicate that these economies have entered the recovery phase of the load capacity curve. Within the observed income range, further economic expansion remains associated with declining load capacity. For surplus economies, GDP and its squared term are statistically insignificant under fixed effects, while the random-effects estimates produce a negative GDP coefficient and a positive squared term, with turning points inside the observed income range.

Under the regime-specific corrected estimators, the asymmetry becomes even more striking. The panel-corrected standard errors estimates show that GDP and squared GDP are insignificant in deficit economies, meaning the baseline nonlinear relationship is not robust to the correction of the error structure in this regime. In biocapacity-reserve economies, GDP remains negative and its squared term positive and significant, with turning points inside the observed income range. The corrected estimates consequently support the load capacity curve only among surplus economies. Education and ICT are positive and significant across the deficit specifications under both corrected estimators. In biocapacity-reserve economies, education is insignificant under PCSE but negative under FGLS, while ICT is consistently negative. Industrialization is negative in deficit economies under PCSE and FGLS but insignificant in the surplus regime. Natural-resource rents become negative and significant in the full-sample GMM estimates after potential endogeneity is addressed, though the disagreement between the regime-specific corrected estimators prevents this adverse result from being generalized across ecological regimes.

The principal conclusion is that ecological status conditions both the shape of the income-load capacity relationship and the effects of its structural drivers. Previous load capacity curve studies have predominantly estimated common relationships for geographical or income-based groups, rather than dividing countries according to whether ecological demand exceeds biocapacity. This regime-based division reveals asymmetries that pooled samples conceal. Income thresholds for ecological recovery diverge sharply between the two regimes. Structural and capability drivers also operate differently across regimes, and at times in opposite directions. The same development driver may strengthen environmental awareness, technological absorption, monitoring, and resource efficiency where regenerative constraints are already binding, while productivity growth, digital infrastructure, electricity demand, electronic waste, and rebound consumption may erode an existing biocapacity margin faster than efficiency gains can preserve it.

These findings carry direct policy implications. Because income growth has not yet reached the threshold at which it improves ecological capacity in deficit economies, governments there should not rely on growth alone to relieve ecological pressure, and should instead pair growth with efficiency gains and low-impact sectoral targeting. Industrialization policy should be prioritized in deficit economies specifically, where its adverse ecological effect is concentrated, through cleaner production standards and energy-efficient technologies. Education policy in deficit economies should explicitly incorporate green skills and sustainability-oriented capacity, given its consistent contribution to load capacity there. Digitalization strategy should be differentiated by regime, with ICT investment in deficit economies targeting resource-monitoring and efficiency infrastructure, while biocapacity-reserve economies guard against rebound effects that let expanding digital infrastructure erode existing ecological capacity. Because ecological performance responds relatively quickly to changing conditions rather than following a fixed pattern, targeted interventions in either regime are likely to yield measurable effects within a policy-relevant timeframe.

Subject of Research: The load capacity curve across ecological regimes: The roles of industrialization, natural resources, education, and ICT in ecological-deficit and biocapacity-reserve economies

Article Title: The load capacity curve across ecological regimes: The roles of industrialization, natural resources, education, and ICT in ecological-deficit and biocapacity-reserve economies

Article References: Abdi, A. H., Patel, N., Mohamed, F. H., Mehta, D., & Ahmed, A. M. (2026). The load capacity curve across ecological regimes: The roles of industrialization, natural resources, education, and ICT in ecological-deficit and biocapacity-reserve economies. Environmental and Sustainability Indicators, 32, Article 101566. https://doi.org/10.1016/j.indic.2026.101566

Image Credits: AI Generated

DOI: 10.1016/j.indic.2026.101566

Keywords: load, capacity, curve, across, ecological, regimes, roles, industrialization, natural, resources, education, ecological-deficit