Trade-threshold Effect on Inflation in Tunisia: New Evidence Resulting from a Nonlinear Approach

ABSTRACT

Trade openness is an important determinant of the inflation process. The effect of trade openness on inflation, however, is still an issue of debate at both theoretical and empirical levels. This study tried to provide a contribution to the literature by examining the relationship between inflation and trade openness in Tunisia over the period 1975Q1-2015Q4 using a nonlinear model. The originality of this study stems from the fact that it is the first investigation considering both the Residual-Based Tests for Cointegration in Models with the Regime Shifts and Threshold Regression model. The linear model confirms the existence of a positive relationship between inflation and trade in Tunisia. Yet, considering the nonlinear model, trade openness growth and Consumer Price Index (CPI) inflation growth show a statistically significant negative link as long as the trade openness evolution does not exceed the threshold. Nevertheless, if the trade openness growth is higher than the threshold, integrating the trade positively affects CPI inflation. Furthermore, a positive influence of Money supply growth on this type of inflation was noticed in Tunisia in all the considered regimes proving the effect of monetary factors on inflation level. Consequently, trade openness could be used to control inflation in Tunisia.

KEYWORDS:

• CPI inflation
• trade openness
• nonlinearity
• cointegration with regime shifts
• threshold regression model

JEL CLASSIFICATIONS:

• E31
• F41
• O55

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Since 1962, Triffin and Grudel consider that openness is the ‘safety valve’ to lower inflation pressures of the open economies.

2. Nevertheless, using the same data set as Romer (1993), Lane (1997) suggests that nominal price rigidity and imperfect competition in the non-traded goods sector leads to inverse relationship between inflation and openness which means that openness is not just working through trade effect terms.

3. The study used Arellano and Bond (1991) difference GMM estimator.

4. Bowdler and Malik (2017) use the volatility of money supply growth as variable to captures the importance of the seigniorage, also they use terms of trade growth as a proxy of the effect of greater international price stability for high value added products.

5. Authors apply Vector Error Correction model and Impulse Response Function analysis. In addition, Unrestricted VAR model is applied to the analysis for Azerbaijan.

6. In other models, the natural logarithm of the GDP deflator was used as a proxy for inflation (Kim and Beladi 2005; Thomas, 2012).