Page 36 - Azerbaijan State University of Economics

THE JOURNAL OF ECONOMIC SCIENCES: THEORY AND PRACTICE, V.82, # 2, 2025, pp. 32-60

In the context of natural-resource-dependent economies such as Algeria, the literature
highlights the importance of fluctuations in commodity prices. Mendoza (1995)
analyzed the impact of terms of trade on economic fluctuations, a crucial aspect for
oil-exporting countries. Oil revenues can influence the government's ability to spend,
and the volatility of these revenues can lead to macroeconomic instability. Chellai
(2021) specifically studied the impact of government spending shocks on
macroeconomic variables in Algeria using SVAR models, highlighting the relevance
of this approach to the Algerian context. Daoudi (2023) also examined the impact of
fiscal policy on economic growth in Algeria using a SVAR model, reinforcing the
relevance of this methodology for the Algerian case.

SVAR models have become a standard tool for macroeconomic policy analysis. They
overcome the limitations of reduced VAR models by imposing restrictions based on
economic theory to identify structural shocks. These restrictions can be short-term
(contemporaneous) or long-term. Lütkepohl and Velinov (2016) discussed methods
for identifying long-term restrictions via heteroskedasticity. Pfaff (2008) also
provided an implementation of VAR, SVAR and SVEC models in the R package
'vars', demonstrating their practical applicability.

Variance decomposition and impulse response functions (IRFs) are key tools in
SVAR analysis. IRFs visualize the dynamic response of one variable to a unit shock
in another, while variance decomposition quantifies the proportion of the variance in
the forecast error of one variable explained by shocks in other variables. These tools
are essential for understanding the transmission channels of shocks and the
interdependence between macroeconomic variables (Martin et al., 2013).

In summary, the literature review highlights the relevance of SVAR models for analyzing
the impact of public spending, particularly in resource-dependent economies. Existing
studies provide a solid basis for the proposed analysis, while highlighting the need for a
contextual approach to understand the dynamics specific to Algeria.

METHODS
This section details the methodological approach adopted to analyze the impact of
public spending on Algerian economic performance. We rely on the Structural Vector
Autoregression (SVAR) model, a robust econometric technique for analyzing
dynamic interdependencies between macroeconomic variables. The discussion will
include the theoretical aspect of the SVAR model, the data used, and the steps
involved in identifying structural shocks.

36