A Comparative Study on Consumption Functions: The Case of the European Union

Abstract

The consumption function describes the relationship between consumption expenditures and income. As is well known,the distribution of macroeconomic data such as income and consumption is unequal. Accordingly, estimators derivedfrom linear models may be inefficient. This study attempted to reach efficient estimators, using gamma distribution,within the limits of this study. The main purpose of this study is to estimate parameters of mainstream consumptionfunctions using the panel-data of EU members and negotiating countries. Using data from the World Bank, the EuropeanUnion Macroeconomic Database (AMECO), and the Bank of International Settlement (BIS), consumption functionswere estimated by the Generalized Linear Model (GLM) approach. The reliability of the estimators was tested withthe Generalized Moments Method (GMM). Furthermore, the study uses GLM, based on machine learning, to obtainrobust estimators for overfitting. Findings of all three methods are compatible with each other and the “PermanentIncome” hypothesis verified.