The Chinese Economy and Income Inequality among East Asian Countries

Discussion Papers In Economics
And Business




The Chinese Economy and Income Inequality among
East Asian Countries

Sumie SATO Mototsugu FUKUSHIGE


Discussion Paper 10-06












Graduate School of Economics and
Osaka School of International Public Policy (OSIPP)
Osaka University, Toyonaka, Osaka 560-0043, JAPAN

---

The Chinese Economy and Income Inequality among
East Asian Countries

Sumie SATO Mototsugu FUKUSHIGE


Discussion Paper 10-06

February 2010


この研究は「大学院経済学研究科・経済学部記念事業」
基金より援助を受けた、記して感謝する。


Graduate School of Economics and
Osaka School of International Public Policy (OSIPP)
Osaka University, Toyonaka, Osaka 560-0043, JAPAN

---

The Chinese Economy and Income Inequality among
East Asian Countries

Sumie SATO

Graduate School of Economics, Kobe University
1-2, Rokkodai-cho, Nada-ku, Kobe 657-8501, Japan

Mototsugu FUKUSHIGE

Graduate School of Economics, Osaka University
1-7, Machikaneyama-cho, Toyonaka, Osaka 560-0043, Japan

Abstract
Using the Atkinson inequality measure of income distribution, we analyze the impact of China as a single country
and examine the effect of its domestic income inequality on total income inequality among East Asian countries.
First, we find that China‟s domestic income inequality exacerbated income inequality among East Asian countries
from the 1980s, and this effect became even more pronounced from 1990. Second, the growth of China‟s per
capita GDP had an equalizing effect on income distribution in a framework of ASEAN + China, but this effect was
reversed around 1997. However, relative to higher income countries such as Japan and South Korea, China‟s per
capita GDP remains low, and although China has contributed to income inequality in the area, it has recently had a
more equalizing effect.

Keywords: East Asia, China, income inequality, free trade agreement, harmonious society
JEL classification: O15, D31, F43

 Correspond to: Mototsugu FUKUSHIGE
Graduate School of Economics, Osaka University
1-7, Machikaneyama-cho, Toyonaka, 560-0043, JAPAN
Phone: & fax: +81-6-6850-5248
E-mail: mfuku@econ.osaka.u.sc.jp

- 1 -

---

1. Introduction
At the National People‟s Congress of 2007, Chinese Premier Wen Jiabao emphasized the
construction of a „balanced society‟, signaling that China had abandoned its „Get Rich First‟ policy, in
the words of Deng Xiaoping in 1978. This policy change was made to deal with dissatisfaction about
domestic policy; that is, the policy of „Get Rich First‟, which made economic growth its highest priority,
caused domestic income inequality. There are many empirical studies of recent trends in China‟s
domestic income inequality. For example, Tsui (1993), Chen and Fleisher (1996), Akita (2003) and
Kanbur and Zhang (2005) conducted a decomposition analysis of income inequality. Yao and Zhang
(2001), Zhang, Liu and Yao (2001) and Maasoumi and Wang (2008) investigated regional income
convergence. Liu (2006), Wan (2004) and Zhang and Wan (2006) measured China‟s inequality in rural
areas. Wang, Shi and Zheng (2002) measured urban inequality. Dong (2005), Gustafsson and Shi
(2002) and Sicular, Ximing, Gustafsson and Shi (2007) investigated differences in economic growth
rates in urban and rural areas. In addition, Kanbur and Zhang (1999) examined differences between
inland and coastal areas. Further, Meng, Gregory and Wang (2005) and Meng, Gregory and Wan (2007)
studied poverty problems brought about by increased inequality. In attempting to determine the causes
of rising inequality, Fujita and Hu (2001) and Wan, Lu and Chen (2006, 2007) analyzed the relationship
between globalization and economic growth, and Ma (2006) studied the role of foreign direct
investment. It is difficult to survey all the early work on China‟s income inequality because there are so
many empirical studies. However, most studies focus only on the trend and growth rate of inequality;
we cannot find studies of its international effect or of its effect on China‟s neighbors.
In terms of population and size, China‟s provinces (including five big cities) are similar in
scale to individual ASEAN countries. In this paper, we regard Chinese provinces as one country, and
compare China with ASEAN countries. For example, the per capita GDP of Shanghai is higher than that
of Malaysia, which has the second highest per capita GDP among the ASEAN countries and a larger
population. The per capita GDP levels of the Chinese cities of Tianjin and Beijing exceed that of
Thailand, which has the ASEAN region‟s third highest per capita GDP. China‟s per capita GDP
exceeds Indonesia‟s and, while falling short of the levels of Singapore, Malaysia and Thailand, exceeds
- 2 -

---

the ASEAN average. Hence, were there to be an economic union or a free trade agreement (FTA)
between China and ASEAN, China would be expected to be the leader because of its economies of scale
and presence in terms of both population and income. Regarding income inequality, although inequality
among Chinese provinces has exhibited a recent upward trend, it remains low by ASEAN standards. In
other words, in forming an economic union or FTA in this area, ASEAN countries would face a greater
income inequality challenge than would Chinese provinces. The existing studies of income inequality in
the ASEAN region by Mizoguchi and Yoshida (1998) and Datt and Walker (2004), who analyzed
inequality in 1985 and 1997, respectively, are somewhat dated. The problem of inequality became more
apparent in 2004, which is the final year covered by our study.
In this paper, using the Atkinson inequality measure of income distribution, we investigate the
impact of China‟s inequality in two ways. First, we investigate the impact of China‟s domestic
inequality when calculating inequality measures with and without considering Chinese provinces as a
country. Second, we measure the impact of China as one country when comparing income inequality
among East Asian countries including and excluding China.
A summary of the results follows. Since 1980, the inequality of China‟s domestic income
distribution has exacerbated income inequality in East Asia, and this effect has been even more
pronounced since 1990. On the other hand, this result depends on how income inequality is measured:
within a framework of ASEAN + China, the growth of China‟s per capita GDP has had an equalizing
effect on income distribution in the area. However, this effect has weakened since 1997. These
outcomes were a consequence of the economic stagnation of some ASEAN countries following the East
Asian financial crisis and the rapid growth of China‟s economy. When higher per capita GDP countries
such as Japan and South Korea are included, per capita GDP in China is relatively low, and although
China has contributed to income inequality in the area, it has recently had a more equalizing effect.
This paper is organized as follows. In Section 2, we explain the inequality measure used in this
study and propose indices to measure the effects of China as one country and its domestic inequality. In
Section 3, we define specific geographical areas for considering regional economic inequality and
investigate recent trends in per capita GDP and income inequality. In Section 4, we investigate the
- 3 -

---

impact of China‟s economic growth and its domestic income inequality on East Asian countries, using
the indices developed in Section 2. In Section 5, we summarize our results and discuss remaining issues.

2. The Inequality Measure and its Application
In this paper, we measure inequality in the income distribution by using the measure proposed
by Atkinson (1970). To analyze inequality, this measure uses a social welfare function that is based on
the evaluation of individual utility, which is defined as follows:


1 
u(x)  a 
x
b
for   ,
1 and   ,
0
1  
u(x)  ln x for   .
1

Then, the Atkinson measure is defined as:

1
1



1 n  x
1

A  1
i
  
 ,




for   ,
1 and   ,
0
n

i 1
 
 



 1 n
 x 
A  1  exp
log i ,





 for   ,1
n



i 1



where n is the total number of individuals, x is the income of the i-th individual, and  is average
i
income. The parameter  represents the degree of inequality aversion. Treating the social welfare
function in such a way is an advantage of the Atkinson measure.
We propose two methods for capturing the impact of China as one country and the impact of
its domestic income inequality using the Atkinson inequality measure. First, to measure the impact of
China as one country on income inequality in East Asia, we propose using two versions of the Atkinson
- 4 -

---

measure: one including and one excluding China. Second, to measure the contribution of China‟s
domestic income inequality, we propose two versions of the Atkinson inequality measure: one based on
considering China as one country and one based on considering Chinese provinces as individual
countries.
To measure income inequality among East Asian countries including China, with a population
weight for each country ( w ,
i  ,
1 ,...,
2
M ), the Atkinson measure is calculated as:
i

1
1



1 M
 x
1

A  1  
w i  ,




for   ,
1 and   ,
0
 N
i

i 1


 



 1 M
 x 
A  1  exp
w log i ,





 for   ,
1
N
i





i 1



in which N is the total population of the area under study. Defining
C
A as the Atkinson measure that
excludes China from the area under study, we propose the following index for measuring the impact of
China as one country:

DC  A
AC

.

To clarify the use of this index, consider the following simple numerical example. Suppose there is an
economy comprising five people ( j  ,
1
5
,...,
2
), whose incomes are 10, 20, 30, 40 and 50. Table 1
compares the Atkinson measure for all five people with that when one is excluded. Table 1 shows that
when the maximum and minimum incomes are excluded, the Atkinson measure is low, but when
average incomes are excluded, the Atkinson measure goes up. Hence, this index measures the
contributory effect on income inequality of a specific country. For our analysis, when China is
- 5 -

---

classified as a low income country in the area, DC is positive, and changes from positive to negative
according to China‟s economic growth. Further, DC is minimized when China‟s income is the average
for the area, and its value can only increase from that point. One would expect DC to become positive if
China‟s income grows sufficiently. It is clear from Table 1 that positive and negative changes in DC are
asymmetric; that is, DC falls rapidly but rises slowly. The changes are not symmetric around the
average.
To measure the contribution of China‟s domestic income inequality in the area under study, we
calculate the Atkinson measure considering China as one country ( A ), and considering the Chinese
provinces as countries (
D
A ). The difference between these two measures represents an index of the
contribution of China‟s domestic inequality, as follows:

DD  A
AD

.

Clearly, DD is zero if all the average incomes of the Chinese provinces are equal; otherwise, this index
takes a negative value. This index measures the decreasing effect of the domestic inequality of a
specific country on overall income inequality. Using the Atkinson measure, Das and Parikh (1982) and
Lasso de la Vega and Urrutia (2003) developed a decomposition method. However, instead of using a
decomposition method, because we focus on the contribution of China‟s domestic income inequality,
we construct an index from the difference in Atkinson measures. Thus, we investigate the impact of
China as one country and the impact of its domestic income inequality by using the DC and DD indices
described above.

3. China’s Regional Income Inequality versus ASEAN’s Income Inequality
To analyze the inequality of China‟s income distribution from the viewpoint of East Asia, we
consider the following specific groups of East Asian countries: the ASEAN countries; ASEAN + China;
ASEAN + 3 (namely China, Japan and South Korea); and all East Asian countries. The first three
- 6 -

---

groups relate to countries bound by FTAs or involved in negotiations. Table 2 shows the member
countries. Before calculating the DC and DD indices, we investigate the relative position of China‟s
economy and trends in per capita GDP and levels of inequality within the country groups. The data used
are population and GDP figures denominated in US dollars from 1980 to 2004. We obtained Chinese
data from the China Compendium of Statistics 1949–2004, compiled by the Department of
Comprehensive Statistics of the National Bureau of Statistics (2005). China‟s GDP figures are
converted from Chinese yuan into US dollars by using annual exchange rates from the World Economic
Outlook Database of the International Monetary Fund (IMF). All other data are from the World
Economic Outlook Database of the IMF. In this paper, the (eight) ASEAN countries are Indonesia,
Laos, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Vietnam; because the World
Economic Outlook Database of the IMF lacks the necessary data, we omit Brunei Darussalam and
Cambodia.
In 2004, because China‟s share of East Asia‟s total population was 62.9% and ASEAN‟s
(including Brunei Darussalam and Cambodia) was 27.1%, China and the ASEAN 10 accounted for 90%
of East Asia‟s population in that year. The percentages of East Asia‟s population accounted for by other
countries were: Japan, 6.3%; South Korea, 2.4%; Taiwan, 1.1%; and Hong Kong, 0.3%. Figures 1 and 2
compare levels of per capita GDP and economic growth rates for groups of countries with those of
China. To smooth out annual fluctuations, we used three-year moving averages for economic growth
rates. These figures confirm that China‟s recent economic growth rate exceeded the average growth rate
of the ASEAN countries, and China‟s level of GDP has almost reached that of ASEAN. Figure 3, which
shows each country‟s share of regional GDP, indicates that Japan accounted for more than 50% of East
Asia‟s GDP in 2004. However, the combined share of China and ASEAN is about 30%. Moreover,
Figure 3, which also shows GDP shares for 1980, 1990, 2000 and 2004, indicates that East Asia‟s GDP
for 2004 was almost five times its 1980 level.
In Figure 4, we compare the GDP per capita levels of the ASEAN countries with those of the
top five Chinese provinces (in terms of GDP per head), which are Shanghai, Tianjin, Beijing, Zhejiang
and Jiangsu. In 1980, the per capita GDP levels of these top five Chinese provinces were similar to
- 7 -

---

those of the ASEAN countries. In 2004, Shanghai‟s per capita GDP grew more rapidly than did that of
Malaysia, and China‟s fifth largest city, Jiangsu, matched Thailand‟s per capita GDP. These high
growth rates in coastal China probably result from China‟s „Get Rich First‟ policy.

4. China and Income Inequality in East Asia
In Figure 5, we compare the Atkinson income inequality measures, with  = 1, for ASEAN +
China, ASEAN + 3, the Chinese provinces and the ASEAN 8. Figure 5 shows that the Atkinson
measure for China has risen recently, but remains low, which means that China‟s inequality is relatively
low among ASEAN countries. For example, whereas China‟s Atkinson measure, with  = 1, is 0.11 for
2004, the corresponding figure for ASEAN is 0.30. This result is consistent with the findings of
Milanovic and Yitzhaki (2002), who “find that Asia is the most heterogeneous continent;
between-country inequality is much more important than inequality in incomes within countries.”
Figures 6 and 7 illustrate the DC and DD indices corresponding to the Atkinson measures with
 = 0.5 and  = 2, respectively. In both figures, the DD lines for ASEAN + 3 and East Asia are so similar
that it is difficult to distinguish between them.
The DC index with  = 0.5 within a framework of ASEAN + China implies reduced inequality
after 1988 and then a resurgence from the trough reached in 1998. This result means that, within an
ASEAN + China framework, around 1998 China‟s impact changed from reducing inequality to
increasing it. Moreover, China‟s entry into the higher income group of ASEAN countries around 1998
because of its rapid economic growth transformed China‟s effect on inequality. Further, in a framework
of ASEAN + 3 or an East Asian framework, China only began reducing income inequality around 2001.
This means that China‟s economic growth is expected to reduce income inequality within the ASEAN +
3 and East Asia. On the other hand, the DD index shows that the effect of domestic income inequality in
China increases income inequality in ASEAN + China, ASEAN + 3 and East Asia. However, the
contributions implied by DD are small relative to those implied by DC. In a framework of ASEAN +
China, DD indicates that China‟s domestic income inequality increased overall income inequality, but
- 8 -

---