THE FREQUENCY-DOMAIN CAUSALITY RELATIONSHIP BETWEEN GOVERNMENT SPENDING AND GDP FOR THE PERIOD 1967-2023: WHICH IS MORE VALID IN ALGERIA, THE KEYNESIAN HYPOTHESIS OR WAGNER'S LAW?

Abstract

The purpose of this paper is to re-examine the relationship between government spending and Gross Domestic Product (GDP) in Algeria during the period 1967-2023. Using a novel approach known as frequency-domain causality, this methodology combines classical time-series analysis with spectral analysis, offering the advantage of assessing the co-movement of series in the context of both time and frequencies. In the case of the Algerian economy, we find that frequency-domain causality is able to identify the relationship between the variables in the short, medium, and long run, and these results have significant implications for both the Keynesian hypothesis and Wagner's Law.

The Gregory-Hansen cointegration test results indicated the presence of a long-run relationship between government spending and GDP in Algeria for the 1967-2023 period. The findings within the frequency-domain causality framework revealed that the short, medium, and long-term fluctuations between the variables are time-linked. The analysis of the continuous frequency-domain causality results pointed to a unidirectional causal relationship running from GDP to government spending across different frequencies (short, medium, and long term). This leads to the validation of Wagner's Law as the more accurate hypothesis for Algeria during the study period, in contrast to the Keynesian hypothesis and the absence of a relationship in the opposite direction.