The Evolution of Income Risk and Consumption Insurance in South Korea over the Last Two Decades*
| Date | 01 April 2021 |
| Published date | 01 April 2021 |
| Author | Seonghoon Kim,Taehyun Ahn,Chung Gu Chee |
| DOI | http://doi.org/10.1111/obes.12404 |
328
©2020 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 83, 2 (2021) 0305–9049
doi: 10.1111/obes.12404
The Evolution of Income Risk and Consumption
Insurance in South Korea over the Last Two
Decades*
Taehyun Ahn,†Chung Gu Chee‡ and Seonghoon Kim§
†Department of Economics, Sogang University, Mapo-gu, Seoul, Korea
(e-mail: ahn83@sogang.ac.kr)
‡Economist, Bank of Korea, Jung-gu,Seoul, Korea (e-mail: chunggu.chee@bok.or.kr)
§School of Economics, Singapore Management University, Singapore (e-mail:
seonghoonkim@smu.edu.sg)
Abstract
Using data from the Korea Labor Income Panel Study, we study the evolution of income
risk and consumption insurance against transitory and permanent income shocks in South
Korea overthe last two decades. Wefind a decreasing trend in both income and consumption
risks. Furthermore, we estimate that 47.6% of permanent income shocks and 9.8% of tran-
sitory income shocks pass through to consumption. Wealso provide evidence of substantial
improvementsin consumption insurance among the less educated and the older cohor t near-
ing retirement. Our results suggest that recent developments of social safety net and welfare
system may have played an important role in insuring income shocks in South Korea.
I. Introduction
Much attention has been paid to income and consumption inequalities among researchers
and policymakers across many countries. Although the patterns of income inequality and
income risks are extensively explored, most of the existing studies focus on the United
States and European countries, mainly because of data availability and the size of their
economies (Blundell and Preston, 1998; Blundell, Pistaferri and Preston, 2008; Piketty,
2014).
Using nationally representative household panel survey data on consumption and
income, this study examines the trend of income and consumption inequalities in South
Korea (hereafter, Korea)and investigates the extent of consumption insurance with respect
to income shocks over the past two decades. We argue that exploring the case of Korea
JEL Classification numbers: E21, E24, J31.
*Jihye Park and Junxing Chayprovided excellent research assistance. The views expressed in this paper are those
of the authors and do not reflect the official views of the Bank of Korea.We are grateful for the financial support from
the Bank of Korea and the Ministry of Education and the National Research Foundation of the Republic of Korea
(NRF-2017S1A3A2066494). All errors are our own.
Consumption insurance in South Korea 329
is interesting and meaningful. Korea is one of the few countries that experienced rapid
economic growth since the 1960s. However, prior to the Asian financial crisis in 1997,
the government did not invest sufficiently in social insurance and means-tested welfare
benefits. In addition, due to the set of labour market reforms implemented after the finan-
cial crisis, temporary employment, including fixed-term contracts and temporary agency
work, has become widespread (Grubb, Lee and Tergeist, 2007). Hence, to improve insur-
ance against income shocks, the Korean government has gradually extended and adopted
social insurance and means-tested programs over the past two decades (Jo, 2008; Kim,
2010).1To the best of our knowledge, this study is the first of its kind to investigate the
degree of consumption insurance from developed countries other than the United States.
We argue that Korea can be considered a representative example among several industrial
countries in Asia that experienced unprecedented economic growth.
Recent research has documented that income inequality (typically measured by the
variance of log income) or income risk (estimated from unexplained component growth
of income) has been widening across developed countries (Attanasio, Hurst and Pistaferri,
2014; Attanasio and Pistaferri, 2014; Piketty, 2014). However, it is not clear whether the
increased income risk will translate to a lower level of welfare because of the availability
of consumption insurance. For example, severalf actors, such as intra-household transfers,
social insurance and means-tested welfare programs, and intertemporal optimization, can
affect howwell protected an individual or a household is from the impact of adverse income
shocks.
Empirical evidence on consumption insurance is relatively scant, compared with evi-
dence on income risks or wealth inequality, because of the lack of high-quality consumption
panel data. Most studies in the literature use data from the Panel Study of Income Dynamics
(PSID) in the United States (Blundell et al., 2008; Kaplan and Violante, 2010; Attanasio
and Pistaferri, 2014; Blundell, Pistaferri and Saporta-Eksten, 2018). These studies have
documented that, over the past decades, consumption inequality grows less rapidly com-
pared with income inequality.Moreover, these studies have shownsome evidence of partial
insurance of permanent shocks. However, it is difficult to find studies outside the United
States because of data limitations. The study of Santaeul`alia-Llopis and Zheng (2018)
which utilized a Chinese data set is an exception. They studied the consumption insurance
in China between 1989 and 2009 using data from the China Health and Nutrition Survey
(CHNS), a panel survey of nine provinces in China. Hence, the findings of our study would
contribute to the literature by expanding the understanding of consumption insurance in a
different country setting other than the United States and China. In addition, this study has
an advantage in terms of data quality over those existing studies because our survey data
collect comprehensive information on consumption spending (comparable with the US
Consumer Expenditure Survey) on an annual basis. The previous studies using the PSID
or CHNS data used imputed measures of total household consumption spending.
We firstly document the evolution of income inequality and consumption inequality,
which are measured byunconditional variances of log income and log consumption, respec-
tively. Figure 1 indicates that both income and consumption inequalities have decreased
1Examples include a means-tested cash transfer program (the National Basic Livelihood Security System) for
the needy introduced in 2000, non-contributory pension schemes for the poor and the elderly (the Basic Pension
Program) implemented in 2008, and gradual expansions of unemployment insurance, first introduced in 1995.
©2020 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial