Scholarlink

<p>The‍‌‍‍‌ Just Energy Transition (JET) in the Global South is a challenging and complex issue, being at the energy-poverty, sovereign financing, and geopolitical fragmentation intersection. This paper argues against the generally accepted view that geopolitical stability (GS) helps decarbonization in energy-import-dependent (EID) countries. We develop a hybrid Autoregressive Distributed Lag (ARDL) and Double Machine Learning (DML) framework to estimate conditional causal effects to investigate a panel of 29 Global South countries (1999–2024). Conventional dynamic models (ARDL) depict a positive association (+ 0.070) for the GS × EID interaction. Conventional dynamic models (ARDL) depict a positive association (+ 0.070) for the GS × EID interaction. In contrast, our Double Machine Learning (DML) conditional estimates for aggregate renewable consumption indicate a substantial sign reversal (estimate: -0.102). However, we reveal that this ‘Stability Trap’ is highly metric-dependent. When we exclude traditional biomass and estimate the model using strictly modern renewable electricity output, the conditional estimate reverses to a positive direction (+ 0.121), and ARDL confirms a significant positive long-run effect (+ 0.021, <i>p</i> &lt; 0.05). This demonstrates that the negative penalty in aggregate metrics is largely a statistical artifact of the ‘Biomass Paradox.’ When modern green grid infrastructure is isolated, geopolitical stability is associated with a positive transition effect. We interpret this as a measurement artifact rather than a uniform failure of modern green investment.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The stability trap and the Biomass Paradox: conditional evidence from a hybrid ARDL-DML analysis of just energy transitions in the Global South

  • 期刊论文

  • 发表日期：

    2026年7月30日

查看全文
  • Saqib Munir,
  • Mushab Rashid,
  • Haider Ali Shams

摘要

The‍‌‍‍‌ Just Energy Transition (JET) in the Global South is a challenging and complex issue, being at the energy-poverty, sovereign financing, and geopolitical fragmentation intersection. This paper argues against the generally accepted view that geopolitical stability (GS) helps decarbonization in energy-import-dependent (EID) countries. We develop a hybrid Autoregressive Distributed Lag (ARDL) and Double Machine Learning (DML) framework to estimate conditional causal effects to investigate a panel of 29 Global South countries (1999–2024). Conventional dynamic models (ARDL) depict a positive association (+ 0.070) for the GS × EID interaction. Conventional dynamic models (ARDL) depict a positive association (+ 0.070) for the GS × EID interaction. In contrast, our Double Machine Learning (DML) conditional estimates for aggregate renewable consumption indicate a substantial sign reversal (estimate: -0.102). However, we reveal that this ‘Stability Trap’ is highly metric-dependent. When we exclude traditional biomass and estimate the model using strictly modern renewable electricity output, the conditional estimate reverses to a positive direction (+ 0.121), and ARDL confirms a significant positive long-run effect (+ 0.021, p < 0.05). This demonstrates that the negative penalty in aggregate metrics is largely a statistical artifact of the ‘Biomass Paradox.’ When modern green grid infrastructure is isolated, geopolitical stability is associated with a positive transition effect. We interpret this as a measurement artifact rather than a uniform failure of modern green investment.