Urbanization and carbon emissions: looking at the role of mobile phone adoption in Sub-Saharan African countries.

Despite the plethora of studies on urbanization-carbon dioxide emissions relationship, studies that consider the role of mobile phone adoption are limited in the ecological literature. This study relied on the stochastic impacts by regression on population, affluence and technology (STIRPAT) analytical framework for modelling environmental impacts and adopted fixed effects ordinary least squares with Driscoll and Kraay standard errors (FE-DK) and the novel method of moments quantile regression (MM-QR) estimation techniques to examine the role of mobile phone adoption in the urbanization-carbon dioxide emissions link for 21 SSA economies, spanning 1995-2017. Results of estimation based on FE-DK statistically provide support for population size, per capita income, energy intensity, urbanization and mobile phone adoption as determinants of the two forms of carbon dioxide emissions (consumption-based carbon dioxide emissions and production-based carbon dioxide emissions). Distributional effects of these factors explain that (i) urbanization has heterogeneous positive effect on the two forms of carbon dioxide emissions, with higher impact in economies with relatively lower level of carbon dioxide emissions and (ii) mobile phone adoption has heterogeneous negative effect on the two forms of carbon dioxide emissions, with greater impact in economies with relatively higher level of carbon dioxide emissions. The study discussed the policy implications of these results in the context of SSA countries.

Assessing the distributional effects of financial development on consumption-based carbon emissions in Sub-Saharan Africa: a quantile-based analysis.

This study assessed the role of financial development (FD) and its distributional effects in explaining consumption-based carbon (ConCO2) emissions, in a framework that also examined the environmental Kuznets curve (EKC) hypothesis, in the context of 19 Sub-Saharan African countries. A composite index was used as measure of FD in a set of data spanning over the period 1995-2017, while controlling for population size (PS), energy intensity (EI) and natural resource rents (Nrr). Given that the variables deviate from expected normal distribution as adjudged by results of pre-estimation tests, the method of moments quantile regression (MM-QR) estimation technique was used to account for distributional effects of FD on ConCO2. Results of the fixed-effect regression based on Driscoll-Kray standard errors (FE-DK) which was validated by three other estimators (fully modified ordinary least squares (FMOLS), dynamic ordinary least squares (DOLS), canonical cointegration regression (CCR)) statistically provided support for FD, PS and EI as drivers of ConCO2. Distributional effects of this show that FD exerts significant positive effect on ConCO2 among countries in the higher quantiles, but insignificant positive effect among those at the lower quantiles. The model provided no support for the EKC hypothesis for SSA; policy implications of these results were presented.