In DINA, we can simply compare the difference between the pre-tax and post-tax income to evaluate the effect of government redistribution in improving income inequality. However, the choice of tax disposable income or post-tax national income is crucial. Post-post-tax disposable income is pre-post-tax national income minus all forms of tax and adding all forms of government trans-fers, without considering collective expenditure. Post-tax national income is disposable income plus collective expenditure. Since collective expenditure a wowowowoworsrsrsrsrsenenenenininingggg
(public goods consumption) is not targeted at any specific population group, but at all citizens, we take the following approach. When we evaluate the effect of all forms of social security programs in assisting people at the bot-tom of the income distribution, we use the disposable income. We call this transfer assistance from social security as the social safety net here. When we evaluate the total effect of the government redistribution policy, we use the post-tax national income. Note that for people at the bottom, the ma-jority of them are non-filers, and the difference between pre-tax income and post-tax disposable income is from individualized transfers (monetary and in-kind transfers). Figure 18 depicts three types of income distribution for the bottom 30%: pre-tax national income (blue), post-tax disposable income (green), post-tax national income (red). For example, on average, the bot-tom 1% receives the annual individualized transfer of about NT$37,000 and imputed collective expenditure of about NT$35,000. Overall, imputed collec-tive expenditure becomes larger as we move upward the income distribution, while the individualized transfer becomes smaller.
Our first finding is that, from 2001 to 2015, the social safety net becomes thinner. That is, social security transfers targeted at the people at the bottom are less in 2015 than in 2001. Figure 19 depicts the pre-tax income and post-tax disposable income of 2001 (lighter color) and 2015 (deeper color) for the bottom 55%. There are two remarks. First, post-tax disposable income for the bottom 30% is larger in 2001 than that in 2015, while pre-tax income is similar. Since the majority of the bottom 30% are non-filers, this means individualized transfers received by the bottom 30% at every income percentile are decreasing from 2001 to 2015. Second, the crossing point of pre-tax income and post-tax disposable income is decreasing, in terms of both income percentiles and real income. In 2001, pre-tax income and
post-at
NT$0
2015 Bottom 30%
Post-tax DI
圖 18: Income Distribution of the Bottom 30% in 2015: Pre-Tax National Income, Post-Tax Disposable Income, and Post-Tax National Income
tax disposable income cross at the 52nd percentile of about NT$350,000; in 2015, they cross at the 41st percentile of about NT$240,000. So not only the amount of individualized transfers become less, the number of people who have higher disposable income than pre-tax income is also decreasing.
Overall, we call this phenomenon a thinner social safety net.
Our second finding is that, despite decreasing social transfers, collective expenditure is increasing from 2001 to 2015, and this makes the post-tax national income in 2015 larger than it in 2001 at every percentile over the income distribution. Figure 20 depicts the pre-tax and post-tax national income of 2001 (lighter color) and 2015 (deeper color) for the bottom 55%.
This finding is important, as it points out that the positive and strong post-tax income growth rate for the bottom 50% shown in Figure 17 is due to the increased collective expenditure, not due to improved social security trans-fers. In fact, if we calculate the growth rate of post-tax disposable income at
I
NT$0
2001 and 2015 Bottom 55%
Post-tax DI
Pre-tax NI 2015 cross at $240,000, 41%
2001 cross at
$350,000, 52%
圖 19: Pre-Tax National Income and Post-Tax Disposable Income, 2001 &
2015
each income percentiles, the bottom 20% of people have a negative growth rate.
Our third finding is that the social security system has little effect in im-proving the income distribution in Taiwan. We divide social security transfers into two categories: social insurance transfer (monetary) and health insur-ance transfer (in-kind). The top panel of Figure 21 draws the average social insurance transfer rate—received social insurance transfers divided by the average national income—of various post-tax national income groups at the period 2001-2015. In 2015, the social insurance transfer is about 3% of na-tional income. We can see that social insurance transfers do differ between the top 10% and the bottom 90% income earners. This is because social insurance has more subsidy programs that are targeted at the low-income households. On the other hand, health insurance transfer drew in the bot-tom panel of Figure 21 does not show much difference across income groups.
NT$0
2001 and 2015 Bottom 55%
Pre-tax NI Post-tax NI
圖20: Pre-Tax and Post-Tax National Income, 2001 & 2015
Taiwan’s National Health Insurance system covers all citizens—in this sense, health insurance is like public goods—and its transfer criterion depends on the frequency of treatments and the degree of illness, not income. Over-all, the social security transfer occupies about 10% of the national income.
Among it, 7% is health insurance transfer, not targeted at income, and only 3% aims at assisting the bottom 90% income earners.
Finally, we calculate the average tax rate across different income groups.
The tax rates we compute take into account all forms of taxes. The average tax rate is computed as all taxes paid divided by total pre-tax national income of a specific income group. Figure 22 shows how effective tax rate varies across income groups. Overall, the average tax rate of the entire population is about 13%. The middle 40% (P50-P90) paid 10% of their pre-tax income, close to the average income tax rate of this group. The top 1% paid about 23% of their pre-tax income, while their average income tax rate is 35%.
1.0%
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
% of average national income
Bottom 50%
Middle 40%
Top 10%
All
Average social insurance transfer rate by post-tax income group
(a)
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
% of average national income Bottom 50%
Middle 40%
Top 10%
All
Average health insurance transfer rate by post-tax income group
(b)
圖21: Average Social Security Transfer and Health Insurance Transfer Rate
9%
11%
13%
15%
17%
19%
21%
23%
25%
27%
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
% of pre-tax national income
Average tax rate by pre-tax income group
Middle 40%
Top 10%
All Top 1%
圖 22: The Average DINA Tax Rate