REVISITING TRADE LIBERALIZATION AND ECONOMIC GROWTH IN PAKISTAN: A NON-LINEAR MACHINE LEARNING APPROACH

Abstract

The role of trade liberalization in fostering economic growth is widely recognized. However, its impacts differ from one economy to another. This research analyzes the effect of trade liberalization on the economic growth of Pakistan from 2001 to 2024, considering important control variables such as FDI, remittances, gender development, and governance effectiveness. Using the Kernel-based Regularized Least Squares Machine Learning Modeling approach, the study captures non-linear and complex relationships that are often missed by linear econometric techniques. The findings suggest that both trade liberalization and other variables like FDI, remittances, gender development, and governance effectiveness have a positive and significant impact on economic growth. Also of importance is the discovery of a non-linear relationship between trade liberalization and growth, which demonstrates that too much liberalization becomes counterproductive. The study reinforces the notion that even though maximization of trade liberalization enhances economic performance, it should be exercised with caution to safeguard the long-term economic stability of Pakistan.