Comments welcome.
Paper prepared for presentation at the Japanese Political Science Association Annual Meeting, Okayama, October 8-10, 2011
2
I. Introduction
Japan and Taiwan were praised for their growth with equity, and characterized as egalitarian countries (World Bank, 1993; Milanovic and Yizhaki, 2002). Recently, however, news such as school students are not able to pay tuition and their school lunch, or some people committed suicide because they could not be able to repay their debts, are often reported. Research reports also show that the income inequality is rising in Taiwan ( OECD, 2006, 2008; DGBAS, 20101).
While a certain degree of income inequality is useful in rewarding hard work, i.e.
different labor receives different income, if the income inequality gap is too big, it implies that the fruits of economic development are concentrated in too few hands.
In that case, income inequality may not be conducive to social mobility and even lead to social tension, eventually resulting in social unrest as we have seen in riots in London this summer. We expect Japan and Taiwan should be more sensitive to income inequality because the idea that fruits of economic growth were evenly shared across society was central to the political legitimacy of political regimes.
Equality is a key component of social contract. Rising inequality is expected to generate pressure on the government. How have the Japanese and Taiwanese government responded to rising income inequality? More importantly, to what extent, has income inequality affected Japanese and Taiwanese politics respectively?
What happen when income inequality increases? Based on the median voter hypothesis, Meltzer and Richard (1981) argue that greater inequality tends to produce greater generosity in redistribution policy. The lower the median income person/family is below the mean income, the more median income person/family would like to get benefits from the government. This median-voter model is intuitively convincing and some researches support this model (Kenworthy and Pontusson, 2005).
Meltzer and Richard’s argument that the government will respond with more generous programs when the income inequality rises implies that inequality has some impact on electoral politics. In the case of Japan and Taiwan, the recent development seems to be consistent with the median voter model. The current ruling party in Japan, the Democratic Party of Japan (DPJ) campaigned on expanding welfare services and defeated the long-term dominant Liberal Democratic Party (LDP) in 2009. As for Taiwan, major parties pledge to expand social welfare and
subsequently Taiwan’s welfare system has been expanded over past years. The ratio
1 Director –General of Budget, Accounting and Statistics (DGBAS, 2010), Executive Yuan, Taiwan
3
of social spending to government expenditure keeps increasing.
However, with a closer look at inequality measures (shown in section II), we find that an increase in inequality was continuous in Japan over time since the 1980s with the pace quickening in the late 1990s. In actual fact, back then, some scholars like Tachibanaki (1998) already pointed out that Japan may not be as equal as assumed, even more similar to the US. The inequality issue did not receive much attention until mid-2000s. What explains this timing puzzle? Besides, even though a majority of the Taiwanese thinks that income inequality is a serious problem, income inequality or welfare issues have not been a salient issues during past elections.
The causal relationship between rising inequality and more redistributive policy/a more expanded social welfare system should not be taken for granted. The median voter model assumes that public preferences will be expressed by voting. However, people’s voting behavior may be shaped by a variety of issues rather than mainly on redistributive policy. Moreover, inequality affects the public unevenly. Not everybody will treat income inequality with the same weight. In particular, social welfare policy benefits all as a public good. It is not so easy to use welfare policy to mobilize votes.
To have a better understanding of the relationship between income inequality and government response, we need to examine public perception of income inequality.
This paper does not discuss policy content per se but mainly focus on how people perceive income inequality and what the public considers most and wants the government to do. This paper uses Japan and Taiwan as case studies to proceed as follows. The following section reviews income inequality measurements and identify the trend for income inequality and possible driving forces in Japan and Taiwan. This is followed by an examination of perceptions of degree of inequality to evaluate the public response to inequality. Then, to evaluate the impact of income inequality on electoral politics, I analyze why income inequality and social welfare became a salient issue in Japan in 2009 but not in Taiwan during past elections despite of the similar trend for income inequality. And, finally, I conclude this paper with a discussion of this study’s implications.
II. Inequality in Japan and Taiwan
How serious is income inequality in Japan and Taiwan?
1) Income Inequality
According to the most often used inequality measure, the Gini coefficient, in 2008,
4
the coefficient for Japan was 0.3268 and 0.304 for Taiwan.2 If we use the OECD average as a benchmark, both the numbers are slightly above the OECD average, still more egalitarian than the US and Britain but less egalitarian than Sweden and
Germany.
In the case of Japan, in the 1970s, the Gini coefficient measure was below the OECD average as one of the most egalitarian countries. As the figure 1 shows, according the Income Redistribution Survey conducted by the Ministry of Health, Labor and
Welfare, since the 1980s, the Gini coefficient measure has continuously increased and accelerated in the second half of 1990s, up to 0.3814 from 0.3143 in 1981. In the 2000s, the coefficient went up and down little bit but did not change much and maintain around 0.36-0.38.3
Source: Income Redistribution Survey, Ministry of Health Labor and Wealth, Japan In Taiwan (as shown in Figure 2), there was also an upward trend in income inequality.
The Gini coefficient measure rose from 0.303 in 1989 to 0.35 (the record high) in 2001. Recently, we see a steady rise again but not over 0.35 yet. On average, between 1985 and 2008, Gini coefficient measure increase 0.5 percent annually in Taiwan and 0.4 percent annually (OECD, DGBAS, 2010). And from Figure 2, we can see that income inequality increased significantly in the 1990s.
2 Data are from Japan’s Ministry of Health and Welfare’s Income Redistribution Survey, Taiwan’s DGBAS, 2010 and OECD reports (2006, 2008). For comparison, equivalence-adjusted Gini coefficients are used.
3 Not equivalence-adjusted Gini coefficients 0
Figure 1 Gini coefficient (1975-2008, Japan )
5
source: Social Indicators (2010), DGBAS, 2010
If we take another measure, the Oshima index(R/P 20%), calculating the ratio of the top 20 percent average to the lowest 20 percent average income, we can find a similar upward trend. In Taiwan, in the 1980s, the share of household income accruing to the top 20 percent was 4.48 times that of the bottom 20 percent. It rose up to 5.5 in 1999 and 6.39 in 2001. The index went down little bit but re-rose to 6.34 in 2009. In Japan, in the 2000s, the share of the top 20 percent is also 6 times that of the bottom 20 percent, around 20 percent more than in the OECD average ( Mira d’Ercole, 2005). And researches find that in both Japan and Taiwan, both household incomes of the top 20 percent grew faster than those of the bottom 20%, and the gap between middle quintile and top quintile is stable, i.e., the gap between rich and poor is widening (OECD 2011; DGBAS, 2010)
2) Poverty Rate
If we use half of the median household income as a threshold to calculate the relative poverty, we find that the rate was 15.8 percent in 2008 in Japan and 7.7 percent in Taiwan in 2009. Between 1985 and 2008, the poverty rate increased 3.7 percent in Japan and 0.7 percent in Taiwan.4 It suggests that poverty is more prevalent in Japan.5 But It is also worthy of noting that the poverty rate in Taiwan increased 2.9 percent between 1999 and 2009. Poverty became also much more prevalent in Taiwan.
4 But if we look at non-monetary indicators, we will find that the percentage of households that declared having experienced different forms of material deprivation is below the OECD average (Mira d’Ecrole). It reflects that Japan’s high per capital income.
5 Again, compared with the OECD average (10 percent), Japan’s poverty rate is around 50 percent higher than the OECD average (OECD).
0.27
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Figure 2 Gini Coefficient (1989-2009, Taiwan)
6
Either from Gini coefficient, the Oshima index, or the poverty rate, all indicators show that income inequality is rising in Japan and Taiwan. What explains rising inequality in Japan and Taiwan?
3) What are driving forces for income inequality?
The rise of inequality occurred in most countries during periods of sustained
economic growth. Similarly, between 1980s and present, GDP per capital respectively increased 3.6 percent annually in Japan and 6.1 percent in Taiwan (OECD, DGBAS, 2010). Why did everybody not benefit from growth in the same way? The following factors have been identified as having the most important impact on widening inequality in Japan and Taiwan although the degree of influence may not be the same.
First, income inequality in Japan and Taiwan is partly driven by demographic change, i.e. aging population. Although some senior people are high ranking managers enjoying peak earnings, most elderly people have retired to live on modest pensions or/and personal savings, and have a lower disposable income than people of working age. Accordingly, as the population is aging and there is a higher percentage of people aged 65 or over, the income disparities between the elderly and the working age group increase.6
Second, what makes the situation more difficult is the changed household structure.
Traditionally, families played an important role in providing basic livelihood security in Japan and Taiwan. When people entered old age, the elderly tended to live with their adult children, in the so called three-generation household. However, recently, the number of three generation householdshas declined, and there are more and more elderly-only households. Research shows that income inequality in elderly households is larger than that of other households (Shirahase, 2002). In addition to the increase in elderly households, there also are more single-parent households, in particular single mothers with small kids. Usually, single mothers do not have full-time jobs and experience high rates of poverty.7
Third, rising inequality also results from the closer integration into the global
6 Due to low fertility rate and population aging, Japan’s population started to fall in 2005. Now, about three working people are supporting one elderly person but in 2030, two working persons will have to support one retiree. In the case of Taiwan, in 2026, every 3.3 working persons support one elderly person and in 2051, the ration is expected to be 1.5 to one.
7 According to statistics, half of working single-parent households lives in relative poverty.
7
economy and rapid technological progress. (a common driving forces for most countries experiencing income inequality) In a globalized economy, there are more multiple corporations. Companies move operation to overseas places with cheaper labor to maintain their competitiveness. Many of Japanese and Taiwanese companies also move their operation overseas. Meanwhile, technologic change has shifted the production in favor of innovation. Globalization and technological process together have brought higher rewards for high-skilled labor. In contrast, low-skilled labor’s wage level has been prevented from increasing if they want to keep their jobs and compete with labor in developing countries. As a result, even the economy as a whole grows, the lower-skilled labor’s wages may not increase and thereby maintain stagnant, leading to widening wage disparities.
Fourth, the tax policy is less distributive in Japan and Taiwan. For example, in Taiwan, there is no tax on capital gains and the rich can accumulate fortunate without paying taxes. Moreover, to promote economic growth, the Taiwanese government
consistently provides tax breaks as incentives. As a result, 71 percent of revenue comes from wage earners who only make up 44.5 percent of GDP（DGBAS, 2010）.
The tax system actually increases inequality instead of reducing the gap. Moreover, as the economy growth slows down, more people become unemployed. The rise of unemployment is also another factor contributing to income inequality.
How does the public in Japan and Taiwan perceive income inequality?
III. Public awareness of income inequality
There are several surveys allowing us to have a basic understanding of how the Japanese and the Taiwanese public think about rising income inequality, and to identify changes, if any, in their awareness of inequality.8
According to the poll, in 2006, 53 percent of respondents thought that income inequality was a serious problem (Yomiuri Shimbun, 2006 06 29). In a later survey conducted in the same year, 73 percent of respondents agreed that income inequality had increased (Asahi Shimbun, 2006 08 29). In 2009, 62 percent of respondents thought that income inequality had gone too far (Asahi Shimbun, 2009 09 03). These results are confirmed by the data from International Social Survey Program (ISSP) (figure 3).
8 Exact wording of each survey questions will be updated further.
8
Source: Tiberghien (2010)
In 2009, 73.7 percent of respondents agreed that difference in income in Japan is too large, rising from 63.8 percent in 1999 (Tiberghien, 2010). There is a strong
perception of income inequality.
In the case of Taiwan, according o the Taiwan Social Change Survey data, in 1992 only less than 30 percent of respondents thought income inequality was a serious issue but in 2005 the number already jumped to 70 percent, a marked increase from the 1992 level (Academia Sinica; Common Wealth2008). According to a series of survey conducted by Common Wealth, 83 percent of respondents were worried about that the degree of income inequality would be worsen in the future, and the number rose up to 93 percent in 2009. In 2010, while asked about the potential conflict between rich and poor, 80.5 percent of respondent thought that the tension was serious (Citizenship and Social Justice Survey, Election Study Center) (figure 4).
0 10 20 30 40 50 60 70 80
Agree Disagree No answer
Figure 3 Do you agree that the income gap in Japan is too large? (ISSP)
1999 2009
9
Source: Citizenship and Social Justice Survey
All the above surveys show that the Taiwanese public thinks that the inequality gap is too large and also has a strong perception of rising inequality.
As to whether or not the government is responsible for reducing the income inequality gap, in Japan, 53.4 percent of respondents agreed that it is the
government’s responsibility to close the income gap, up from 47.4 percent in 1999 (Tiberghien, 2010). In the case of Taiwan, in 2003, 64.4 percent of respondents agreed that the government should reduce the gap (TEDS, 2003) and up to 77.4 percent in 2010 (Citizenship and Social Justice Survey). It is clear that in both Japan and Taiwan, the public expected the government to take an active role in reducing the income gap.
The public’s strong expectation toward the government’s responsibility seems to allow parties to appeal to the public by proposing redistributive polices and claiming themselves as a more responsible government if elected. This leads to another question as to how governments and major parties in Japan and in Taiwan
respectively have responded to the public expectation? More importantly, did rising inequality/welfare issues become a salient issue during elections?
IV. Public reaction to Income inequality (What happened during elections?) 1. Japan
As shown in Figure 1, income inequality started to increase more significantly in the send half of 1990s. At that time, however, income inequality issue did not receive much attention. It could be because people were not sure if it was just one-time
0
Very serious somewhat serious
not serious not very serious
no answer
Figure 4 What Do you consider the tension between rich and poor? (2010)
10
event resulting from the bursting of the bubble economy or presented an on-going change in the Japanese economy. Moreover, at that time, much of the rise in income inequality was attributed to the increase in the number of elderly households, many of which are relatively worse off than young people as mentioned.9 Since it is mainly the aged group that got affected, the rest of the public did not give priority to rising inequality although the public was aware of rising inequality. Even though the DPJ already campaigned on welfare social issues such as child allowance in 2003, voters did not respond to the issue much. In contrast, in the election of 2009, the DPJ again proposed to increase social spending and expand social services. Before the election, the poll showed that pension and health care were the top issue voters consider most (Table 1). After the election, the poll also confirmed that people cast their votes to the DPJ partly because of social welfare related issues such as pension and health care. What explains this change?
Table 1 Which do you consider to be the most important issue? (2009.8)
all male female
Source: Mainichi Shimbun, 2009 08 22
One of reasons attracting people’s attention to rising inequality was the increase in non-regular workers and their low job security.10 Non-regular workers’ wages are much lower than regular workers, about 60 percent of that of regular workers even if they perform the same work (Cargill and Sakamoto, 2008). On hourly basis, part-time workers even only earn 40 percent as much as full-time workers (OECD, 2006). After having experienced the economic difficulties in 1990s, Japanese companies found that they could save money and increase flexibility by hiring non-regular workers.
Now, about one third of the labor force is made up of non-regular workers (Ministry of Health, Labor and Welfare, 2010), up from 19 percent in 1994 (OECD, 2006).
9 Shirahase (2002) points out that the overall extent of economic well-being remains stable, i.e.,the degree of inequality is stable, but within social groups, income inequality among elderly-only households is much higher.
10 Non-regular workers have different kinds of workers, including part-time workers and dispatched/temporary workers.
11
Moreover, their job security is low. When companies encounters trouble, these non-regular workers, especially dispatched workers, are the first to be let go.11 Low wages and low job security together make non-regular workers and their family financially insecure. Having a job no longer means being free of poverty.
Unfortunately, only few of them are able to find full-time jobs afterwards. It is because Japanese companies favor hiring recent graduates. That is, if new graduates are not able to find regular jobs immediately on graduation, it becomes very difficult for them to find stable jobs later. A vicious circle is created. This situation has been particularly harsh on young workers. A rise in income disparity has been sharp among young workers. In other words, the extension of income inequality is spreading toward population of working age.
Under these circumstances, there were more and more media reports on the dilemma of non-regular workers and more relative poverty stories. Plus, the OECD report (2006) clearly attributes the rise of inequality to the rise of non-regular workers. As reports and analyses coming out one after another, the commonly shared view of Japan as solidly middle class and egalitarian was much challenged.12 More importantly, an increase in non-regular workers and the subsequent rise in income inequality made people recognize that Japan’s welfare system was not able to scope with consequences of increasing labor dualism.
Japan’s welfare system provides low public social protection, spending little on unemployment benefits, housing subsidies, family allowance and social services.
They are more insurance based, providing health care, old-age pension and
They are more insurance based, providing health care, old-age pension and