Globalization and the effects of changes in functional income distribution on aggregate demand in Germany

Abstract

Germany has experienced a period of extreme nominal and real wage moderation since the mid‐1990s. Contrary to the expectations of liberal economists, this has failed to improve Germany’s mediocre economic performance. However, Germany is now running substantial current account surpluses. One possible explanation for Germany’s disappointing performance is found in Kaleckian theory, which highlights that the domestic demand effect of a decline in the wage share will typically be contractionary, whereas net exports will increase (Blecker 1989). The size of the foreign demand effect will critically depend on the degree of openness of the economy. This paper aims at estimating empirically the demand side of a Bhaduri and Marglin (1990) type model for Germany. The paper builds on the estimation strategy of Stockhammer, Onaran, and Ederer (2009) and Hein and Vogel (2008, 2009). The main contribution lies in a careful analysis of the effects of globalization. Since Germany is a large open economy by now it is a particularly interesting case study.

Keywords:

• distribution
• demand
• investment
• consumption
• foreign trade
• macroeconomics
• Keynesian economics
• Germany
• globalization

JEL Classifications:

• E12
• E20
• E22
• E25
• E61

Acknowledgements

The authors are grateful to Stefan Ederer, Özlem Onaran and two anonymous referees for comments. Support from FWF Project nr. P18419‐G05 is acknowledged. The major part of the paper was written while Eckhard Hein was a visiting professor at Vienna University of Economics and Business Administration. The hospitality of the Vienna University is gratefully acknowledged. The usual disclaimers apply.

Notes

1. In traditional Kaleckian models, an increase in the wage share leads to an increase in aggregate demand. We use the term post‐Kaleckian for a generalization of these models that allow for profit‐led as well as for wage‐led demand regimes. Kaleckian and post‐Kaleckian models are both part of a broader class of post‐Keynesian models that highlight the central role of aggregate demand. The distinguishing feature of Kaleckian and post‐Kaleckian models is that they treat capacity utilization as an endogenous variable, also in the long run. For a survey, see Blecker (2002).

2. These values are synthethic values for Germany that are used in the econometric analysis. It consists of German data from 1991 onwards that are chained backwards with West German growth rates for the period 1970–1990. The West German export share in 1970 was 21%.

3. According to AMECO data, Germany’s nominal compensation grew by 0.9% annually compared to a 1.3% in the Euro area in the period 1996 to 2006. Over the same period productivity (per employee) grew by 1.8% and that of the Euro area by 1.3%.

4. Functional income distribution and its measure, the wage share, are used synonymously throughout this paper.

5. An extended version of the paper that includes unit root tests and some ECM results is available from the authors upon request.

6. , where e_CW is the elasticity of consumption with respect to wages.

7. As we have no reason to expect that globalization has influenced the consumption differential, no estimations for sub‐periods are reported for consumption. The Chow test found no evidence of a break point in 1987.

8. Note that a change in nominal unit labor costs can also occur because of a technology shock.

9. Note that RULC = Ω = ULC / P.

10. The elasticity of export prices with respect to real unit labor is e _PxΩ = e_PxULC e _ULCΩ. The respective values can also be read from Table 7.

11. Germany’s main trading partners are France, USA, UK, Italy, Netherlands, Belgium, Austria, Switzerland, Spain, Poland, Sweden, Japan, China, Czech Republic and Denmark. The weights were computed with the mean shares of exports and imports for two different periods 1969 to 1989 and 1989 to 2005, using trade data from 1969, 1979, 1989 1991, 1994, 1997, 2002 and 2005.

12. Note that the private excess demand effects discussed above are by definition partial (and disequilibrium effects). Here, however, the effects are general equilibrium effects. The assumption that other control variables are exogenous with respect to income and income distribution is therefore much more restrictive.

13. See the results in Hein and Vogel (2008) and in Stockhammer and Ederer (2008) for Austria and the Netherlands.