Introduction

The pollution haven hypothesis postulates that dirty industries migrate from high-income countries to low and middle-income countries through the trading of goods and foreign direct investment. Foreign direct investment (FDI) inflows remain one of the main sources of external funding for developing countries, yet, the relocation of carbonintensive and energy-intensive industries from jurisdictions with more stringent environmental regulation to weak locales results in pollution haven. The transfer, dissemination, and diffusion of FDI inflows with polluting technologies, goods, and services to developing countries become the most important part of the challenge to achieve the sustainable development goals (SDGs). On the contrary, the environmental Kuznets curve (EKC) hypothesis postulates that the initial growth of a country’s economic development leads to gradual deterioration of environmental quality and improves environmental conditions after reaching a threshold in economic development (Grossman and Krueger, 1991). Thus, both the pollution haven hypothesis and the EKC hypothesis are important policy derivatives for developing countries. Considering the importance of climate change mitigation and its impacts, as accentuated in SDG 13, the effect of FDI inflows, economic development, and energy consumption on greenhouse gas emissions in developing countries needs further attention to be able to alleviate the impacts. Studies on pollution haven hypothesis (Zakarya et al. (2015) Behera and Dash, 2017; Solarin et al. (2017) Sun et al. (2017), support the validity of this hypothesis. Solarin et al. (2017) validated the pollution haven hypothesis for Ghana using the autoregressive distributed lag (ARDL) bounds testing approach. Sun et al. (2017) examined the impact of FDI inflows, economic growth, energy use, economic freedom, urbanization, financial development, and trade openness on CO2 emissions using the autoregressive distributed lag model. The study confirmed the validity of the pollution haven hypothesis in China and that the positive effect of FDI inflows stems from the large contribution of manufacturing, mining and electricity shifted from the developed countries. Using the fully modified ordinary least squares (FMOLS) and dynamic ordinary least squares regression, Behera and Dash (2017) found a positive impact of FDI inflows and energy consumption on CO2 emissions in 17 south and southeast Asian countries, thus, confirming the pollution haven hypothesis. Zakarya et al. (2015) found a long-run effect of FDI inflows and energy consumption on CO2 emissions in Brazil, Russia, India, and China, thus, validating the pollution haven hypothesis via panel causality and FMOLS regression. On the contrary, studies like Zhu et al. (2016) and Zhang and Zhou (2016), rejected the pollution haven hypothesis. Zhu et al. (2016) employed panel quantile regression to examine the heterogeneous effect of FDI inflows, economic growth, and energy consumption on CO2 emissions in Indonesia, Malaysia, Philippines, Singapore, and Thailand from 1981 to 2011. The study found insufficient support for the pollution haven hypothesis but rather found the halo effect hypothesis in high emission countries. Zhang and Zhou (2016) argue that FDI inflows of modern technologies contribute to CO2 emissions reduction in China rather than environmental deterioration. Dasgupta et al. (1999, 2001) and Dean et al. (2004) revealed that developing countries depend on sophisticated technology transfer through FDI inflows from developed countries as their primary source of acquiring technology. Hence, clean and upgrading from vintage to modern technologies help in the reduction of emission levels.