This study endeavors to elucidate the divergent conclusions encountered in empirical research regarding the interplay of Economic Growth (EG) and Energy Consumption (EC). For this purpose, we employ the Panel Smooth Threshold Regression (PSTR) model to intricately examine the non-linear impacts of independent variables on EC and EG within a dataset encompassing 46 countries over the period from 1996 to 2021. The outcomes of our investigation can be summarized as follows: First, the findings underscore the positive impact of the logarithm of net capital formation on EG. This impact is particularly pronounced at low levels of Research and Development (R&D), gradually waning beyond a certain threshold. Second, the ratio of capital to labor exhibits a negative influence on EC at lower R&D levels. Notably, these detrimental impacts become more pronounced as R&D levels increase. Third, trade openness contributes positively to EG, particularly evident at low R&D levels. However, with increasing R&D levels, the incremental benefits from trade diminish. Finally, our findings lend support to the feedback hypothesis. Nevertheless, the impact of R&D expenditures in countries moderates these positive effects. Policymakers should strategically balance resource allocation between capital formation and research endeavors, considering diminishing returns at elevated levels of R&D spending, to ensure sustained EG.

中文翻译：

---

探索不同研发制度下经济增长与能源消耗之间的非线性因果关系：来自 PSTR 模型的跨国证据

本研究致力于阐明有关经济增长（EG）和能源消耗（EC）相互作用的实证研究中遇到的不同结论。为此，我们采用面板平滑阈值回归 (PSTR) 模型来复杂地检查 1996 年至 2021 年期间涵盖 46 个国家的数据集中自变量对 EC 和 EG 的非线性影响。我们的调查结果可以概括如下：首先，研究结果强调了净资本形成对数对EG的积极影响。这种影响在研发 (R&D) 水平较低时尤其明显，超过一定阈值后会逐渐减弱。其次，在较低的研发水平下，资本与劳动力的比率对 EC 表现出负面影响。值得注意的是，随着研发水平的提高，这些有害影响变得更加明显。第三，贸易开放对EG有积极贡献，在研发水平较低的情况下尤其明显。然而，随着研发水平的提高，贸易带来的增量收益逐渐减少。最后，我们的研究结果支持了反馈假设。然而，各国研发支出的影响减弱了这些积极影响。政策制定者应战略性地平衡资本形成和研究工作之间的资源分配，考虑到研发支出水平提高带来的回报递减，以确保持续的EG。