Nonlinear sustainable outcomes of digitalization in Türkiye: A modified cross quantile regression perspective
Author :
Abstract
This study examines the effects of different types of digitalization on sustainable environmental management using quarterly data for Türkiye covering the period 1993-2023. Internet use, mobile subscriptions, and fixed line subscriptions are employed as the main indicators representing digitalization, while carbon emissions serve as the measure of environmental quality. To capture the relationships between variables across different points of the distribution, the Cross-Quantile Regression (CQR) method is applied, and the robustness of the findings is further assessed using Modified Quantile Regression (MQR) and Quantile on Quantile Kernel Regularized Least Squares (QQ-KRLS). The results indicate that internet use negatively affects environmental quality particularly in the lower carbon emission quantiles. However, as emission levels rise, the impact of internet penetration weakens considerably and becomes statistically insignificant in the higher quantiles. Mobile subscriptions are found to deteriorate environmental quality most notably in the middle quantiles, whereas the effects at the lower and upper quantiles remain limited. In contrast, fixed line subscriptions generally reduce carbon emissions, with this beneficial effect becoming more pronounced in the medium and high emission quantiles. These findings demonstrate that the environmental effects of different digitalization types are not homogeneous; rather, they vary depending on the emission level and display a nonlinear structure.
Keywords
Abstract
This study examines the effects of different types of digitalization on sustainable environmental management using quarterly data for Türkiye covering the period 1993-2023. Internet use, mobile subscriptions, and fixed line subscriptions are employed as the main indicators representing digitalization, while carbon emissions serve as the measure of environmental quality. To capture the relationships between variables across different points of the distribution, the Cross-Quantile Regression (CQR) method is applied, and the robustness of the findings is further assessed using Modified Quantile Regression (MQR) and Quantile on Quantile Kernel Regularized Least Squares (QQ-KRLS). The results indicate that internet use negatively affects environmental quality particularly in the lower carbon emission quantiles. However, as emission levels rise, the impact of internet penetration weakens considerably and becomes statistically insignificant in the higher quantiles. Mobile subscriptions are found to deteriorate environmental quality most notably in the middle quantiles, whereas the effects at the lower and upper quantiles remain limited. In contrast, fixed line subscriptions generally reduce carbon emissions, with this beneficial effect becoming more pronounced in the medium and high emission quantiles. These findings demonstrate that the environmental effects of different digitalization types are not homogeneous; rather, they vary depending on the emission level and display a nonlinear structure.
