The synchronization between Korea's and Japan's business cycles

• Published date: 01 December 2023
• Author: Keun Yeong Lee
• Date: 01 December 2023
• DOI: http://doi.org/10.1111/asej.12313

ORIGINAL ARTICLE

The synchronization between Korea’s and Japan’s business

cycles

Keun Yeong Lee

Sungkyunkwan University, Seoul, Republic of Korea

Correspondence

Keun Yeong Lee, Sungkyunkwan University, 25-2, Sungkyunkwan-ro, Jongno-gu, Seoul, 03063, Republic of

Korea.

Email: lky0614@skku.edu

Abstract

This article analyzes the evolution of the dynamic interactions between Korea’sand

Japan’s business cycles. The logarithmic industrial production is first decomposed into

trends and cycles using bounceback models. The estimation results of the two-state

Markov switching model show that the synchronization coefficient of Korea–Japan is

positive and time-varying. However, according to the estimation results of the

heteroscedasticity-based VAR model, the Japanese business cycle shock has a positive

effect on the contemporaneous Korean business cycle, but not vice versa. Based on

these results, I estimate a TVP-VAR model assuming Cholesky decomposition and

find that Japanese upward shocks do not have positive impacts on the Korean busi-

ness cycle in the period before the global financial crisis or the period after the global

financial crisis and before the COVID-19 outbreak. The response of Korea to the

Japanese shock is smaller in the three-variable TVP-VAR compared to the two-

variable TVP-VAR without the United States. The Korean business cycle upward

shock also has a similar effect on the Japanese business cycle, albeit smaller,

depending on the period. Overall, the size of the response seems to be closely related

to global events as well as changes in trade, FDI, and political conditions between two

countries.

KEYWORDS

bounceback model, heteroscedasticity, regime switching, synchronization, TVP-VAR

JEL CLASSIFICATION

C5, E3

DOI: 10.1111/asej.12313

© 2023 East Asian Economic Association and John Wiley & Sons Australia, Ltd.

Received: 6 February 2023; Accepted: 2 November 2023

Asian Econ. J. 2023;37:435–465. wileyonlinelibrary.com/journal/asej 435

1|INTRODUCTION

Since the inauguration of President Yoon Suk Yeol of the Republic of Korea in

2022, political and economic tensions between Korea and Japan have been

greatly easing. However, for a long time, Japan has been perceived by Koreans

as both a close and a distant neighbor, despite its geographical proximity, due

to unsavory historical events. From an economic point of view, Japan was

Korea’slargestimporterformorethan30 yearsuntil2006aseconomic

exchanges between Korea and Japan became active after the Korea–Japan

claim agreement that took effect in December 1965. However, this position has

been ceded to China as China has been Korea’s largest exporter and importer

since 2007. Nevertheless, Japan is still an important trading partner for Korea,

as reflected in conflicts over Japan’s export restrictions on materials, parts, and

equipment to Korea during the Moon Jae-in government. Because Korea’s

exports are still highly dependent on Japan’s basic capital goods, Korea has

long maintained an industrial structure in which imports from Japan also

increase when Korea’s exports increase. In addition, Japan joined the OECD in

1964, the first Asian country to do so, and has long served as a driving force for

the world economy along with the United States, Germany, France, and the

United Kingdom as one of the G5 countries. Therefore, it cannot be denied that

the Korean economy, a small open economy that relies heavily on trade, is

closely related to the Japanese economy. However, despite this close relation-

ship in economic terms, there seem to be not many studies on the macroeco-

nomic interconnection between the Korean and Japanese economies.

My aim with this study is to examine the correlations and causal relation-

ships between these business cycle estimates after decomposing logarithmic

monthly industrial production data of Korea and Japan from 1989 to recent

years into trends and cycles. First, in order to separate cycles from trends, unlike

previous researchers who used the HP (Hodrick–Prescott) filter, the bandpass

filter, and the BN (Beverage–Nelson) decomposition, I use nonlinear time-series

autoregressive (AR) models such as Hamilton’s(

1989) Markov switching model

and the bounceback models. The lag numbers of AR models are 0, 1, 2, and so

forth, and I examine the case of assuming the t-distribution as well as the nor-

mal distribution. Next, to see how the estimated industrial production fluctua-

tions in the two countries are synchronized over time, I estimate Leiva-Leon’s

(2017) model with a time-varying measure of synchronization. I also analyze

how this synchronization probability correlates with the proportion of exports

(imports) to Japan in Korea’s total exports (imports). Frankel and Rose (1998)

and Baxter and Kouparitsas (2005) showed that larger trade volumes were asso-

ciated with higher levels of business cycle synchronization. In addition, because

correlation does not necessarily mean causation, I estimate the causal relation-

ship at the same time based on heteroscedasticity (e.g., Ehrmann et al., 2005)

and then analyze the impulse response between variables. Finally, I examine

436 LEE