Re-Analysis of the EU Public Debt Crises with NARX

Abstract

This paper estimates the public debt of Southern EU countries by adapting the public debt law of motion equation, which includes the variables that represent the competitiveness of a country such as past public debt, GDP, primary balance, real exchange rate, real interest, and inflation, to Southern EU countries. The analysis is performed with the Non-linear Autoregressive Exogenous Artificial Neural Network, which is a dynamic and robust statistical method for the post-monetary union period consisting of quarterly data between 2005Q1 and 2021Q4. The results of the analysis reveal that the public debt in the Southern EU countries is robustly explained by the public debt law of motion equation with a confidence ratio of over 95%. This result implies that the public debt problem in Southern EU countries is associated with the competitiveness of these countries. In addition, the analysis goes beyond parametric analyzes that relate it to economic growth or a few variables in the estimation of public debt and uncovers the significance of inclusive variables and non-parametric analyzes included in the public debt law of motion equation in solving the problem.