Comment on “Estimating Non‐Keynesian Effects for Japan”

• Author: Naoyuki Yoshino
• DOI: http://doi.org/10.1111/j.1748-3131.2012.01240.x
• Published date: 01 December 2012
• Date: 01 December 2012

Comment on “Estimating Non-Keynesian

Effects for Japan”

Naoyuki YOSHINO†

Keio University

JEL codes: E62, H30, H62

Kameda (2012) is a very important piece of empirical work for the Japanese economy.

Japan has accumulated government debt to a level that exceeds more than 200% of gross

domestic product (GDP). Japan’s annual dependency on the issuing of government

bond is about 50% of the general expenditure of the central government. However,there

is strong opposition to increasing tax rates as it is believed that raising taxes would cause

the Japanese economy to stagnate even further.

Kameda (2012) is an empirical paper and focuses on providing some timely esti-

mates for the Japanese economy. Three of the important conclusions found by Kameda

are: (i) concerns about the negative effect of a hike in the consumption tax on Japanese

economy are unwarranted. The effect of raising the consumption tax on the economy is

neutral since both the Keynesian and non-Keynesian effects will offset each other. (ii) If

the Japanese government adopts an expansionary government spending policy, then the

non-Keynesian effects might depress the Japanese economy. (iii) The paper stresses the

importance of the deficit to GDP ratio rather than the debt to GDP ratio as the debt to

GDP ratio seems to provide less information about the future.

The empirical analysis in Kameda is based on the impulse response functions

obtained from a five variable vector autoregression (VAR). I have four comments on the

paper in the form of suggestions for future analysis. First, Kameda’s empirical analysis is

based on a reduced form. Therefore, whether the impact of government spending and

tax policies on the Japanese economy comes from aggregate supply side effects or aggre-

gate demand side.effects cannot be well explained in this model. Second, the model is a

closed economy model. The exchange rate will be affected by various fiscal policies, and

the results obtained by Kameda might be affected if the exchange rate is included as a

variable in the empirical estimation (McNelis & Yoshino, 2012). Third, if we look at the

components of government spending in Japan recently, social welfare spending is

increasing, while public works expenditure is diminishing. The effectiveness of an

increase in government spending will differ depending on whether it is as a result of an

increase of infrastructure investment or an increase of social welfare spending for elderly

people Fourth, financial institutions do not appear in the model at all. Japanese govern-

ment bonds are mainly held by domestic financial institutions which is very different

from the case of Greece where foreign ownership of government debt is high. The

†Correspondence: Naoyuki Yoshino, Faculty of Economics, Keio University,2-15-45 Mita, Minato-

ku, Tokyo108-8345, Japan. Email: yoshino@econ.keio.ac.jp

bs_bs_banner

doi: 10.1111/j.1748-3131.2012.01240.x Asian Economic Policy Review (2012) 7, 246–247

© 2012 The Author

Asian Economic Policy Review © 2012 Japan Center for Economic Research

246