Testing the Validity of the Feldstein-Horioka Puzzle for G20 Countries with Panel Data Method

Abstract

Domestic and foreign savings are viewed as the foundation of economic growth and development in most modern economic studies, whether in developed or developing countries. Since the 1980s, with the increasing capital mobility, many studies have been carried out in this field. Recently, capital mobility has been frequently discussed based on the Feldstein-Horioka (FH) puzzle, which assumes the existence of a strong relationship between domestic savings and investment. This study aims to examine the validity of the FH puzzle for the G20 countries in the 1995-2020 period. The variables of the ratio of domestic savings and investments to gross domestic product used in the study were obtained from the World Bank's Development Indicators database. The study utilized innovative data analytical procedures for panel data analysis, taking into account the methodological problems of the existing literature. In this context, the current study considers panel data's cross-sectional dependence and heterogeneity. The FH panel coefficients from the augmented mean group (AMG) and the common correlated effects mean group (CCEMG) estimators are 0.261 and 0.301, respectively. The results show that the FH puzzle is not valid for G20 countries. This study presents essential policy and practice implications for future studies based on savings and investment relationships.