Japanese manufacturing facing post-Fukushima power crisis: a dynamic computable general equilibrium analysis with foreign direct investment

Abstract

The Great East Japan Earthquake and the subsequent tsunami that hit and severely damaged the Fukushima Daiichi Nuclear Power Station resulted indirectly in the shutdown of most of the nuclear power plants in Japan. To compensate for the lost nuclear power supply, more fossil fuels were used. People became concerned that this could be disadvantageous for domestic manufacturing industries and accelerated their offshoring to Asia, especially China, through foreign direct investment (FDI). We used a world trade computable general equilibrium (CGE) model with endogenous FDI from Japan to China to quantify the impact of the power crisis on the Japanese manufacturing sectors. We found that the power crisis as well as FDI would adversely affect several sectors that use power intensively, but would benefit the transportation equipment (TEQ), electric equipment (EEQ) and machinery sectors, despite the common expectation that these sectors would undergo a so-called ‘hollowing-out.’

Keywords:

• computable general equilibrium analysis
• foreign direct investment
• power crisis
• hollowing out of manufacturing

JEL Classification:

• D58
• F21
• Q43

Acknowledgements

The author gratefully acknowledges useful comments and suggestions provided by Kanemi Ban, Tom Hertel, Shiro Takeda and two anonymous referees. The usual disclaimer applies.

Funding

This work is supported by JSPS KAKENHI [grant number 25380285].

Supplemental data

Supplemental data can be accessed here.

Notes

¹ This is the result of their study with an assumption of full employment. They also employed an alternative assumption with wage rigidity and unemployment and predicted a more severe outcome.

² We conducted sensitivity analysis with respect to the first two parameters and confirmed that our results were affected little by their alternative assumptions.

³ See Hosoe (2012) for details.

⁴ Sensitivity analysis was conducted with respect to these elasticities. As shown in the ‘Appendix’ section in Supplementary Data, its results indicate that our simulation results are not affected qualitatively by this assumption. The details of the results are available upon request.

⁵ TEPCO’s fuel consumption during the period August 2011 to July 2012 increased by 118% for heavy OIL, 85% for crude OIL, 19% for LNG and 15% for COA compared with that during the period March 2010 to February 2011. URL: http://www.tepco.co.jp/tepconews/pressroom/consumption-j.html TEPCO’s financial report shows that nuclear capacity constitutes 5.7% of its total assets for fiscal 2010.

⁶ Details are shown in the ‘Appendix’ section in Supplementary Data, which is available upon request.

⁷ Indeed, we install difference of products between origins by Armington’s (1969) assumption in our CGE model but do not install any structure describing product differentiation à la Dixit and Stiglitz (1977).