Disentangling Structural Breaks from Nonlinearity in the Energy–Growth Nexus: A Regime-Based Panel Analysis for Service Economies

Energy and electricity are foundational inputs for logistics networks, digital services, and production systems, making accurate modelling of the energy-growth nexus essential for service-system planning and infrastructure investment.This study revisits the energy-growth relationship in six Middle Eastern economies over 1970-2022 and examines whether widely reported nonlinearities reflect intrinsic economic dynamics or instead arise from unmodelled structural breaks.Adopting a sequential empirical strategy, the analysis combines baseline panel estimators (pooled OLS, fixed effects, and random effects), quadratic-form diagnostics, and Bai-Perron multiple structural break tests.The results show that apparent nonlinear patterns in the full-sample panel largely disappear once multiple breaks are endogenously identified and regime-specific models are estimated.Break selection based on the Bayesian Information Criterion reveals four statistically significant breakpoints, partitioning the sample into five regimes.Within these regimes, the relationship between energy use and economic output is well approximated by linear specifications with timevarying coefficients.Regime-wise estimates indicate that per capita energy use has a statistically significant and economically meaningful positive effect on output only during the 2003-2011 period, associated with rapid demand expansion and investment cycles.In contrast, per capita electricity consumption exhibits a consistently positive association with GDP across baseline and regime-specific models, underscoring the critical role of reliable power infrastructure for logistics, services, and digital activities.The findings caution against interpreting nonlinear energy-GDP relationships without explicitly accounting for structural instability.By demonstrating how unmodelled breaks can generate false nonlinearity, the study contributes a transparent, regime-aware empirical workflow that is well suited for energy, logistics, and service-system analytics and supports more adaptive, context-sensitive policy and infrastructure planning.