Euro depreciation and trade asymmetries between Germany and Italy versus the US: industry-level estimates

ABSTRACT

Since April 2014 to March 2015, the European Central Bank expansionary monetary policy instigates a huge depreciation of the euro in terms of dollar. According to the mainstream monetary theory, these dynamics should make the exports cheaper and at the same time make the imports more expensive. Has real depreciation of the euro helped in the improvement of European countries’ trade balances? Following the main methodologies in the recent literature, our study analyses the effects of this depreciation both for Italy and Germany towards the US. We use industry-level data at monthly frequency. The results are different from each bilateral relationship. We find that 11 industries register a long-run improvement (8 for Italy and 3 for Germany). The J-curve effect is proven just in six cases, always for Italy. The inverted J-curve effect is proven in eight cases, four for Germany, and four for Italy. These results seem to be an indirect demonstration of the structural asymmetries between German and Italian economies: German economic system is more able to be competitive with a strong currency, than Italy.

KEYWORDS:

• Industry trade
• bounds testing
• J-curve
• Germany
• Italy
• US

JEL CLASSIFICATION:

• F31
• F32

Acknowledgments

We most sincerely address our thanks to an anonymous referee and to Paolo Buonanno for their suggestions that have contributed at improving a first version of the paper. We are very grateful to Giovanni Urga who gave us the opportunity to discuss our research during the 5^th International Conference in Memory of Carlo Giannini (Bergamo 25-26 November, 2016). The usual disclaimer applies. Thank you to Benedetta Pedrini for his attentive final linguistic revision.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ About the possible advantages of the euro depreciation, see the report published by Natixis (Artus 2012).

² In mid-2014, Mario Draghi, president of the ECB, announced that the bank plans to engage in a form of quantitative easing through the purchase of private sector credit, including asset-backed securities and covered bonds, in addition to a cut of the benchmark refinancing rate from 0.15% to 0.05% and the deposit rate from -0.1% to -0.2%. On 9 March 2015, the Public Sector Purchase Programme actually started. The ECB decided to buy €60 billion worth of bonds a month as a way of injecting cash into European banks. This was supposed to stop in September 2016, but in December 2015, it has been extended by 6 months. The ECB has also said it will start buying regional and local government debt.

³ From the seminal contributions of Magee (1973), an intense debate about the J-curve phenomenon has been developed. The contributions can be divided into two groups: (1) the first group collects the scientific papers that use aggregate data in order to estimate the phenomenon and (2) the second group collects the scientific papers that use bilateral data. Clearly, this division is based on the progress of the econometric techniques. The most relevant contributions in the first group are Miles (1979) and Himarios (1985); they are the first authors who proposed a precise definition of short and long run in this field. Miles used several tests involving both the seemingly unrelated and pooled cross section and time-series regression techniques, while Himarios provides a critique of Miles’ results. Among the first research studies in the second group, see Rose and Yellen (1989), that explained the reasons for the preferable use of bilateral data, and Bahmani-Oskooee and Brooks (1999).

⁴ The year 2010 represented a real turning point for the European economic policy, as shown by another important event: the Greek government accepted the bailout deal with IMF and ECB. See among others Goodhart (2013).

⁵ See also Halicioglu (2007) about Turkey versus its main trade partners, Bahmani-Oskooee and Zhang (2013) about China and the UK and Bahmani-Oskooee, Harvey, and Hegerty (2013) about Italy and the US.

⁶ Bahmani-Oskooee and Zhang (2013) used a maximum of four lags, but the fourth lagged level variable is never significant.

⁷ More precisely, ${\epsilon }_{t-1}$substitutes ${\theta }_2{Y}_{t-1}^{US}+{\theta }_3{Y}_{t-1}^{fc}+{\theta }_{4}RE{X}_{t-1}$ in Equation (2).

⁸ See the Appendix for the complete explanation about the data set.

⁹ This proposition is coherent with the so-called multiplier approach as theorized by Meade (1948, 1949) and described in many international economics handbooks (see, for instance, Gandolfo (2002), chap. 8).

¹⁰ We considered the sum between import and export in order to weigh the international trade share.