Status of the Income Tax Burden in Japan Focusing on High-Income Earners *

Transcription

1 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Status of the Income Tax Burden in Japan Focusing on High-Income Earners * Naoki Oka Senior advisor, National Tax Agency Abstract Among tax items, income tax has brought the largest amount of tax revenue to the government and the greatest number of tax returns have been filed. Therefore, we may say that income tax is the most familiar tax among the people. This article attempts to discuss empirically the current status of the income tax base in Japan by using statistical data and by focusing on super high-income individuals (individual taxpayers who file tax returns with a total amount of income exceeding 50 Million yen). I n the United States, it has been reported that both income and the amount of tax paid are concentrated in a small group of ultra-high-income individuals, i.e., the top 1% of taxpayers. In Japan, too, a similar trend is observed at a lesser degree of concentration than in the United States. With regard to the types of income of high income individuals in Japan, it has been observed that employment income, capital gain which is taxed separately (such as immovable property), and capital gain of stocks account for large portions of their income. A comparison of the top 400 individual tax returns in Japan and in the US showed that, in contrast to the behavior of the top 400 taxpayers in the United States, fewer Japanese top 400 individual tax returns reported interest income and dividend income, and claimed deduction for donation and foreign tax credit. The fact that a small group of high-income individuals bring a large portion of income tax revenue to the government, and that the composition of items of income and deductions is different between such people and others suggests that it is useful to take the behavior of high income individuals into account when analyzing the tax base and discussing income taxation. Keywords: income tax, high-income earners, distribution, tax returns, tax rate structure, wealthy people JEL Classification: D31, H24, K34 * Views expressed in this article do not necessarily represent those of the NTA.

2 416 N Oka / Public Policy Review I. Introduction (the aim of this article) Important features of individual income tax (IIT) Individual income tax brings the largest amount of tax revenue to the government. It also involves the largest number of taxpayers and, therefore, can be thought of as a tax that is familiar to each citizen. As for the size of tax revenue, the estimated amount of IIT revenue for FY2013 (national budget) amounts to 13.9 trillion yen (139 billion USD). This is far larger than the estimated amount of consumption tax (CT 1 ), which is 10.6 trillion yen (106 billion USD), and corporation income tax (CIT), which is 8.7 trillion yen (87 billion USD). As of 2013, IIT has continuously been the largest source of tax revenue since 1971 (on a national-budget basis). As for the proportion of taxpayers, the numbers of taxpayers for CIT, CT, and IIT are 2.7 million, 3.23 million, and million, respectively. The proportion of IIT taxpayers is 18.6 times as large as that of CIT and 15.5 times as large as that of CT. The number of IIT taxpayers is far bigger than that of other taxes. After the CT rate hikes scheduled for 2014 and 2015, the share of IIT revenue may not be the largest among the items of tax revenues, as the size of revenue from CT will increase. The characteristics of IIT as a tax item that is familiar to citizens and one of the main sources of tax revenue for the government will, however, be unchanged. Reasons for focusing on high-income earners This article aims to illustrate the behavior of a small group of taxpayers, i.e., high-income earners (taxpayers with high reported taxable income) based on data from actual tax returns 2. There is no agreed definition of high-income earners. However, in 2010, there were 309,000 taxpayers (0.61% of all IIT taxpayers 3 ) with total amount of income (TAI) 4 exceeding 20 million yen, whereas there were only 50,000 (0.09%) with total amount of income exceeding 50 million yen. The usefulness of focusing on a small number of high-income taxpayers is illustrated below. 1 Consumption Tax in Japan is a tax on goods and services, i.e., VAT. 2 In this article, statistics and surveys published by the NTA and statistical data obtained from tax returns and compiled specially by the research department of the National Tax College have been used for the analysis. 3 The total number of IIT taxpayers, including those paying IIT by withholding at source and not filing tax returns, is million. 4 The total amount of income includes capital gains sucject to separate taxation and excludes income subject to withholding tax as final tax burden (e.g., interest income and dividends from listed stocks (by election). See aupplmentary annex 1 at the end of article for technical information regarding income taxation in Japan. See also supplementary annexes 2 and 3 for the translation of Japanese terms into English.

3 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October The amount of income and tax paid are concentrated on a small number of highincome earners. This can be confirmed by statistics published by the NTA. According to the Survey of Actual Status for Salary in the Private Sector ( Minkyu survey ), million employment income earners who worked throughout the year 2011 received a total amount of salary of 186 trillion yen and paid 7.5 trillion yen IIT. Only 1% or 444,000 of such employment income earners with a salary exceeding 15 million yen or more received 10.3 trillion yen (5.5%) of employment income and paid 2 trillion yen as IIT (27.7%). 2. The composition of income items pertaining to high-income earners is different from that pertaining to other groups of taxpayers. This can be inferred from a diagram of the ratio of the income tax burden of taxpayers filing tax returns, which was submitted to the government tax commission 6. According to the diagram (which looks similar to Figure 9 below), the ratio of the individual income tax burden (i.e., the ratio of the amount of tax to the total amount of income (TAI)) peaks at 28.3% for taxpayers with TAI of 50 to 100 million yen, and the ratio of the tax burden for taxpayers with TAI exceeding 100 million yen becomes lower. The following explanation is given: the level of tax becomes lower for very-high-income earners because they own stock and their capital income 7 is taxed by separate taxation at lower rates 8. Items of capital income, such as capital gain from the sale of stocks, are mobile and also the timing of realization of income (and taxation) can be controlled. This is possible for only the small number of taxpayers with high-income. 3. Many high-income earners are active in the community and are in high positions in society (they are earning a high income because of these facts) and their behavior has a large impact on the society we are living in. Based on the above, and by examining the income and tax returns of high-income earners, we expect to obtain useful and practical suggestions on policymaking and examination of the bases of IIT, which is an important source of revenue for the government. What this article discusses This article intends to: (1) examine the distribution of income and tax that generates IIT revenue; (2) study the main features of high-income earners and their tax returns (i.e., age, See diagram at page 21 icsfiles/afieldfile/2012/12/26/24zen5kai3.pdf 7 The term Capital income used in this article generally refers to interest income, dividend income, capital gain from shares, and capital gain from sale of real property. 8 10% or 20%. Top regular rates are 50% and 55% (From 2015) including both national and local taxes

4 418 N Oka / Public Policy Review place of residence and the composition of income); (3) examine the impact of distribution of the amount of income and tax of high-income earners on the overall IIT base; (4) examine the elements that lower the effective (average) tax rates of high-income earners; and (5) compare tax returns filed by the top 400 taxpayers in the US and Japan, while attempting to illustrate the differences between super-high-income earners in the two largest democratic economies. This article will also discuss, where possible, the difference between empirical data in 2007 and II. Individual Income Tax Rates and Shares (who is paying IIT) II-1 Shares of income and tax in the US (by percentile) According to an empirical survey carried out by staff members of the IRS in the US, individual income tax rates and shares of 84.5 million taxable returns (tax returns with payment of tax) in the US (2010) were as follows: (1) The total amount of adjusted gross income was 8.04 trillion USD and the total amount of income tax was billion USD; (2) The top 1% of taxpayers (with an adjusted gross income of 369,000 USD and above) received 18.8% of income and paid 37.3% of tax. On the other hand, the bottom half of taxpayers received 11.7% of income and paid only 2.3% of tax.

5 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 1 Shares of income and tax by income group (US) % 90% 80% 70% 60% Share of tax Share of income (AGI) 70.62% 59.07% 87.11% 67.55% 97.64% 88.26% 50% 40% 30% 20% 10% 0% 45.17% 37.38% 33.78% 18.87% 11.74% 2.36% Top 1% Top 5% Top 10% Top 25% Top 50% Bottom 50% Source: Adopted from data in SOI (Winter 2013) published by the IRS Million tax returns of which 84.5Million taxable returns II-2 Estimation of shares of income and tax in Japan (by quasi percentile) Inspired by the US empirical analysis, the shares of income and tax of IIT (by percentile) are estimated by using two surveys published by the NTA. i.e., Survey for Self-assessment Income Tax ( Shinsho survey) 10 and Survey of Actual Status for Salary in the Private Sector 9 Table 5 of Individual Income Tax Rates and Shares, 2010 by Adrian Dungan and Michael Parisi, IRS SOI, Winter Table 1 of Sample survey of self-assessment income tax 2011 ( Shinsho survey) Page 173 of NTA Statistics Survey The Shinsho survey is compiled from information contained in tax returns filed by taxable taxpayers who filed returns by the 31st of March of the year following the taxable year, i.e., calendar year and with the payment of income tax as of the 30th of June. Therefore, information in tax returns with the amount of income but without payment of tax (e.g., tax returns for refund of tax withheld) are not included. (The due date for filing tax returns and payment of income tax is in principle the 15th of March).

6 420 N Oka / Public Policy Review ( Minkyu survey) 11 (Note) There are several restrictions in the estimation as follows (main items). (1) In Japan, percentile data is not available and, therefore, the data that is closest to descending cumulative percentiles (such as 1% and 5% of taxpayers) is used. (2) The US statistics are based on tax return data of all taxpayers. In contrast, there are no such comprehensive statistics in Japan. Therefore, the abovementioned two surveys were used to represent all taxpayers. (3) Since the amount of tax paid by taxpayers represented by the Survey of Actual Status for Salary in the Private Sector ( Minkyu survey) is not necessarily final (as they may file tax returns as self-assessment income taxpayers), the same taxpayers may be counted in the two surveys. Estimation based on survey of employment income earners In 2011, 45,657,000 employment income earners who worked through the year received a total amount of salary income of trillion yen and paid 7.5 trillion yen as income tax. The top 0.97%, i.e., 444,000 employment income earners with a total amount of salary exceeding 15 million yen paid 2 trillion yen as income tax, which is almost one third of all income tax revenue from salaries. The top 5.5%, i.e., 2.5 million employment income earners, paid 4 trillion yen as tax, which is almost half of all income tax revenue from salaries. 11 Table 16 of Statistical survey of actual states for salary in the private sector 2011 ( Minkyu survey) published the NTA Page 202 of NTA Statistics report 2011 The Minkyu survey is generated by the the following process. First the NTA selects business enterprises (withholding tax agent) as objects of a sample survey. Then the sampled enterprises fill in schedules about their employment income earners. Finally, the NTA totalizes the schedules. For more information abluot the Minkyu survey, please see page 200 of the 138th NTA annual statistics repoft (FY2012).

7 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 2 Shares of income and tax of salary earners (Japan) % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 27.79% 5.53% Share of tax Share of salaries 52.13% 17.66% 58.88% 22.89% 79.56% 51.20% 87.23% 66.54% 33.46% 12.77% Top 0.97% Top 5.5% Top 8.03% Top 26.75% Top 40.78% Bottom 59.22% (Source) Adopted from data in Minkyu survey (2011) published by the NTA 45.6Million employment income earners who worked throughout year. Estimation based on survey of self-assessment income tax In 2011, 6.07 million taxable returns reported a total income of trillion yen and paid 4.71 trillion yen as income tax. The top 0.7%, i.e., 44,000 taxpayers with a total income exceeding 50 million yen paid 1.3 trillion yen as income tax, which is 28.1% of all income tax paid by self-assessment. The top 6.3%, i.e., 380,000 taxpayers with total amount of income exceeding 15 million yen paid 3.2 trillion yen as income tax, which is 68% of all income tax paid by self-assessment.

8 422 N Oka / Public Policy Review Figure 3 Shares of income and tax of taxable returns (Japan) % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 28.12% Share of tax Share of income 75.03% 68.02% 14.50% 38.46% 44.57% 92.90% 69.76% 97.07% 83.28% 16.72% 2.93% Top 0.7% Top 6.3% Top 8.8% Top 27.6% Top 47.1% Bottom 52.9% (Source) Adapted from data in Shinsho survey (2011) published by the NTA 6.0Million taxable returns II-3 Concentration of income and tax on high-income earners in the US In the US, based on empirical data and statistics, it has been reported that after-tax income for the highest-income households grew more than it did for any other group. For example, according to the findings of a report published by the US Congressional Budget Office (CBO), between 1979 and 2007, income grew by 275% for the top 1% of households and by 65% for the next 19% of households. In contrast, income grew only 18% for the bottom 20 percent of households. As a result, the CBO report states that the share of income going to higher-income households rose, while the share going to lower income households fell. On the other hand, the report states that government transfers and federal taxes became less redistributive.

9 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 4 Trends in adjusted gross income share (percentage) by income group (US) Top 1% Top 5% Bottom 50% (Source) Adopted from data published by the Congressional Budget Office 12 Figure 5 Trends in the share of payment of tax by income group (US) Top 1% Top 5% Bottom 50% (Source) Adopted from data published by the Congressional Budget Office (see footnote 12) 12 Congressional Budget Office, Trends in the distribution of household income between 1979 and

10 424 N Oka / Public Policy Review II-4 Concentration of income and tax on high-income earners in Japan Historical data from the survey for Self-assessment Income Tax ( Shinsho survey) shows similar trends in Japan. The proportion of income and tax on very-high-income earners (such as those with a total amount of income exceeding 50 million yen) has increased. The effective (average) tax rates on very-high-income earners have also risen in recent years (see Fig. 9 below). Between 1999 and 2011, while the proportion of very-high-income earners (exceeding 50 million yen) increased by 0.19%, the share of income rose 5 percentage points and the share of tax increased 9 percentage points. (See Figures 14, 15 and 16 below) Table 1 Proportion of high and very high income earners in the population of taxpayers * The total number of taxpayers in 2010 was million. Figure 6 Number of taxpayers and amount of tax paid by high-income earners by income group (2010) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Number of taxpayers (thousand) Amount of tax (billion yen) ,459 (Total calculated tax amount:3.3 trillion yen) million yen million yen million yen million yen million yen million yen 50 million yen and over

11 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October III. Tax returns filed by high-income earners This section discusses the features of tax returns filed by high-income earners by using statistical data from tax returns filed by taxpayers in A comparison between 2007 and 2010 is also made where possible. III-1 Definition of high-income earners in this article Table 2 Definition of high-income earners in this article III-2 Age groups Table 3 shows the proportion of high-income earners by age and income group, and Figure 7 shows a comparison between 2007 and The proportion of elderly high-income earners of 65 years of age and above (people who receive pensions) among all high-income earners was 32.48% in 2007 and 29.04% in Between 2007 and 2010, the proportion of elderly high-income earners decreased by 3.44 percentage points, and the proportion of those between the age of 45 and 54 rose by 2.27 percentage points. Table 3 Share (percentage) of high-income earners by age and income group

12 426 N Oka / Public Policy Review Figure 7 Proportion of high-income earners by age group in 2007 and III-3 Composition of income of very-high-income earners (50 million yen) Among the various types of income of very-high-income earners, the proportion of income from employment income, capital gain from the sale of land, etc., and capital gain from the sale of stocks (i.e., capital income) are relatively high. On the other hand, the proportions of business income and income from real estate are relatively low. The share of capital income (e.g., income derived from the sale of shares and dividend income) to the total amount of income rose for higher income groups. Among capital income of very-high-income earners (excluding income derived from the sale of shares), the proportion of dividend income and miscellaneous income (including distribution from investment funds) is relatively high. However, it is noteworthy that the income from interest is almost non-existing In Japan, income from interest is generally taxed by withholding at source as a final tax burden.

13 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Table 4 Composition of items of income of very-high-income earners by income group (2010) Table 5 Composition of items of income of all very-high-income earners (in 2007 and 2010) Figure 8 Main items of income of very-high-income earners by income group (2010) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Real estate income Business income Employment income Separate capital gains CG from the sale of Million yen stocks, etc.

14 428 N Oka / Public Policy Review Table 6 Capital income of very-high-income earners (2010) III-4. The use of itemized deductions (tax allowances and tax credits) The use of deductions (tax allowances and credits) reduces the amount of taxable income of taxpayers. The use of the following items of deductions/credits rose for higher-income groups: (i) deduction for donations, (ii) credit for dividends and (iii) foreign tax credit. Some tax returns show a very high amount of deductions and credits. Table 7 Use of deductions and tax credits by income group (2010)

15 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Table 8 Average and maximum amount of deductions and tax credits reported in tax returns (2010) III-5 Effective (average) tax rates The effective (average) tax rates of taxpayers by income group show a peak for the income groups earning million yen and a reduction for income groups with a total amount of income higher than million yen. The effective (average) tax rate here is defined by the ratio of the amount of the calculated amount of tax to the total amount of income. This is because the proportion of items of income that are subject to separate taxation at a lower rate rose for those with income exceeding 100 million yen 14. Between 2006 and 2011, on the one hand, effective (average) tax rates for taxpayers with a lower level of income became lower; on the other hand, the effective (average) tax rates for taxpayers with a high level of income became higher. This was probably due to the changes made in nominal tax rates in the FY2007 tax reform, i.e., the tax rate for the lowest income bracket was reduced from 10 to 5% and the tax rate for the highest income bracket was raised 14 Tax rates applicable to total amount of income of high-income earners (2014)

16 430 N Oka / Public Policy Review from 37 to 40%. Figure 9 Effective (average) tax rates of taxpayers filing returns by income group (2006 and 2011) 35% 30% 25% 20% 2006 (H18) 2011 (H23) 15% 10% 5% 0% up to Million yen ,000 Over 1,000 (Source) Adopted from table 1 of each year of Shinsho survey.

17 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 10 Effective (average) tax rates of taxpayers with and without items of income subject to separate taxation 45% Having income subject to separate taxation Having no income subject to separate taxation Average tax rates (tax divided by the TAI) 40% 35% 30% 25% 33.9% 32.8% 32.2% 34.6% 35.2% 35.6% 36.3% 33.0% 33.5% 33.5% 33.5% 33.4% 37.2% 37.9% 37.8% 33.2% 31.6% 29.9% 38.8% 38.7% 39.0% 28.1% 26.3% 20% 17.9% 15% Over Million yen ,000 Over 1,000 III-6 Tax returns of regular very-high-income earners The number of tax returns reporting very-high-income (exceeding 50 million yen) in recent years increased between 2005 and 2007 and decreased between 2007 and It increased a little again in This fluctuation was caused mainly by changes in the number of one-time very-high-income earners. In the economic environment after the financial crisis of , it may have become difficult to reach an income level exceeding 50 million yen. On the other hand, there is a sizable number of taxpayers reporting a very high level of income. They may be categorized as so called super wealthy individuals (generally defined as households with an amount of net financial assets exceeding 500 million yen 15 ). 15 [Japanese] Nomura Research Institute Private Banking Sengyaku (2013)

18 432 N Oka / Public Policy Review Figure 11 Frequency of appearing in the group of very-high-income earners over six years (2005 and 2010) 70,000 68,143 70,531 60,000 62,884 6 times 61,222 50,816 50,000 2 to 5 times 40,000 30,000 1 time The number of taxpayers with TAI over 50M yen 49,025 20,000 10, III-7 One-time rich and regular rich: composition of incomes The regular rich (whose total amount of income exceeds the platform of 50 million yen for every year between 2005 and 2010) and one time rich (whose total amount of income exceeds the platform for one year in the period of six years) have a significantly different mix of incomes. The regular rich tend to have a mix of employment income, income from real estate, and income from dividends. The one-time rich tend to have a mix of income from real estate and income derived from the sale of land, buildings, etc. (capital gains). In both of these groups, the proportion of taxpayers with employment income is relatively high.

19 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Table 9 Composition of income of regular rich (who appear six times) Table 10 Composition of income of one-time rich Table 11 Composition of income of regular rich after the financial crisis of (who appear 3 times)

20 434 N Oka / Public Policy Review Table 12 Use of selected items reported in tax returns (in 2007 and 2010) (Source) Shinsho survey and Author

21 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October III-8 Tax minimizing behaviors As possible items that can be used for active tax minimizing (or avoiding) activities, the following items were examined: (1) aggregation of profits (notably business income and employment income) and losses (notably loss from real estate income)(2) reporting of foreign capital income (i.e., interest income from abroad). (3) the use of foreign tax credits. The ratio of tax returns reporting loss of real estate income to all tax returns was 1.71% in However, the ratio for very-high-income earners was 4.15% and for the top 400 taxpayers was 7.25% in This might imply that the aggregation of profits and loss between income from real estate and other items of income is one of the popular methodologies for tax minimizing among very-high-income earners. If resident taxpayers have deposits and financial assets in foreign countries and receive interest, it is necessary for such taxpayers to file a tax return and pay tax. Among 50,710 tax returns filed by very-high-income earners, only 1,243 tax returns reported interest income and 1,445 tax returns reported foreign tax credit (implying that the taxpayer has foreign source income) in The proportion of use of foreign tax credit among Japanese taxpayers is extremely low if compared with that in the US. For example, only 38 tax returns filed by the top 400 taxpayers in Japan reported foreign tax credit (2010) whereas 325 US top 400 tax returns did so (2009). IV. Comparison of the top 400 tax returns in Japan and in the US In the US, statistical reports on the top 400 individual income tax returns in each tax year from 1992 have been published. 16 The data in the reports are based on individual returns reporting the largest adjusted gross income (AGI) in each specific year. Inspired by the US analysis, a comparison of the top 400 taxpayers in Japan and in the US has been made by analyzing the top 400 tax returns reporting the largest total amount of income (TAI) in each year. In this comparison, for Japan, data for the years between 2003 and 2010 were used 17 and for the US, data for tax years 2003 to 2009 were used. IV-1 Cutoff income for the top 400 The amount of income (TAI in Japan and AGI in the US) required to be in the group of top 400 taxpayers varies from year to year. In Japan, about 600 million to 1 billion yen (6 to 10 million USD) of TAI is required. In the US, about 80 to 140 million USD of AGI is required. The cutoff income for the top 400 in Japan is about one-tenth of that in the US Adjusted-Gross-Incomes 17 See footnote 12 for the source of data used for analysis in this Section.

22 436 N Oka / Public Policy Review (In this article, foreign exchange is calculated by applying a rate of 1USD = 100 yen). Japan Cutoff income Cutoff income Average Average (TAI) (AGI) income US (Thousands (Thousands (Million yen) (Million yen) USD) USD) , , , , , , ,015 2, , , ,062 2, , , ,038 2, , , , , , , , , ,477 IV-2 Income tax and effective (average) tax rates Japan Total amount Average Total amount Average Average tax Average tax US rate (%) (Thousands (Thousands rate (%) (Million yen) (Million yen) USD) USD) , % ,250,277 25, % , % ,550,332 31, % , % ,599,966 39, % , % ,086,563 45, % , % ,924,540 57, % , % ,593,085 48, % , % ,116,865 40, % , % The effective (average) tax rates (see Section IV-6 below) of very-high-income earners, specifically the top 400 taxpayers, are considerably lower than the statutory top marginal rates in Japan (e.g., 40% in 2010) and in the US (e.g., 35% in 2010). This is due to the facts that preferential tax rates (lower tax rates) are applied to capital income (e.g., income from interest, dividends, and capital gain from sale of shares) and that the proportion of capital income is large for wealthy individuals.

23 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October IV-3 Composition of income NB. Net business losses (average) in the US were 5,047 (2006), 2,590 (2007), and 2,400 (2008). Money amounts are in thousands of USD. IV-4 Large amount of miscellaneous income (Japan) Miscellaneous income includes items of income such as distribution from funds and income from trading of foreign currencies. The annual numbers of top 400 tax returns with the largest amount of miscellaneous income in Japan are as follows:

24 438 N Oka / Public Policy Review IV-5 Deductions and credits The number of tax returns reporting a deduction for donations and foreign tax credits in Japan is much smaller than that in the US. IV-6 Distribution of effective (Average) tax rates Effective (average) tax rates are computed as total income tax divided by the total amount of income (TAI) for Japan and by adjusted gross income (AGI) for the US. In Japan, the largest number of tax returns in the top 400 (2010) was in the 15 20% range Figure 12 Effective (Average) tax rates and the number of tax returns in the top 400 returns (Japan) % 10 15% 15 20% 20 25% 25 30% 30 35% 35%+

25 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 13 Effective (Average) tax rates and the number of tax returns in the top 400 returns (US) % 10 15% 15 20% 20 25% 25 30% 30 35% 35%+ 0 (152 returns). The second largest was in the range exceeding 35% (125 returns). In the US, tax returns in the top 400 were concentrated in the 25 30% range in 1992 and However, in 2009, tax returns were not concentrated in a particular range of rates but spread evenly among the 10 15%, 15 20, 25 30%, and 30 35% ranges.

26 440 N Oka / Public Policy Review IV-7 Tax residence (Japan) The geographical tax area of the majority (more than 60%) of the top 400 taxpayers in Japan is the area of the Tokyo regional taxation bureau. The area of the Osaka regional taxation bureau has the second largest number of the top 400 taxpayers in Japan. Table 13 Tax residence of the top 400 (Japan) (The number of tax returns in the top 400) IV-8 Elements reducing the amount of gross income (Japan) In each year, more than 20 tax returns in the top 400 (i.e., having the total amount of income at the level of 600 million 1 billion yen) reported no taxable gross income. The top marginal tax rate (40% for national in 2010) is not applicable to such tax returns. The amount of gross income is computed by (i) adding the amounts of various items of income (excluding items of income subject to separate taxation), (ii) aggregating profits and losses, and (iii) deducting or/and carrying forward net losses and casualty losses, if any. Various exemptions and deductions will be made before applying regular rates (i.e., top rate of 50% or 55%). See supplementary annexes 1 and 2 for more information.

27 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Table 14 Overlook of tax returns in the top 400 reporting no taxable gross income (Japan) ( ) (Note) The money amounts are simple aggregations of 246 tax returns reporting no taxable gross income between 2003 and Table 14 illustrates that: Major items of income of such taxpayers that can be taxed as gross income (and thus may be subject to the top marginal tax rate) include dividend income and employment income.

28 442 N Oka / Public Policy Review Items of income that are taxed separately at preferential tax rates include, notably, capital gains from the sale of unlisted stocks, long term separate capital gains and capital gains from the sale of listed stocks. Aggregation of profits and losses, i.e., net losses of business income and real estate income, reduces the amount of gross income. On the deduction side, deductions for donations have a sizable impact. The amount of foreign tax credit is also considerable. V. End remarks This article illustrated the tax behavior of high-income earners empirically. Its findings and observations (and reconfirmations) include the following. Concentration of tax among high-income earners Concentration of tax payments among groups of high-income earners. As a tax base (source of revenue) in the real economy, groups of high-income earners are very important. This tendency has intensified in recent years.(see Figures 2, 3, 14, 15 and 16) A large amount of tax is paid by a small number of high-income earners, but this is not necessarily due to a high statutory marginal tax rate or progressive statutory income tax rates. In fact, a graph of effective (average) tax rates by income groups shows a regressive curve for the top income groups (lower for top income groups). From a fiscal policy point of view, IIT is expected to play an important role in achieving goals such as revenue raising and income redistribution. However, the proportion of the aggregated (money amount) contribution being made by the vast majority of taxpayers (i.e., non-high-income earners) is small, and therefore, the ability of IIT to carry out such functions has natural limitations in reality. In order to achieve these goals, a mix of approaches (in addition to IIT) is needed. Difference in the composition of incomes of high-income earners and others High-income earners receive a large proportion of their income from capital income (notably capital gain from sale of stocks, dividend income) and capital gain from real estate, etc., taxed separately from gross income. The majority of taxpayers (specifically those who belong to the lower income groups) do not have such items of income at all, or the amount of such items of income is small, if any. Capital income The fact that the proportion of capital income is large for high-income earners, who bring a large amount of tax revenue to the government as a group might imply that global viewpoints are important when making policy in this area. The tax base of capital income is footloose (mobile) in comparison to income from labor (i.e.,

29 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October employment income, salary and wages). Various approaches for optimal taxation have been discussed, notably discussions on dual income tax. Tax minimizing and avoiding The timing of taxation and attribution of capital income can be decided (manipulated) by the taxpayer as the owner of such income. For example, capital gain from sale of stocks, etc. is taxed when it is realized (i.e. sold). If capital income is realized (stocks are sold) after becoming a non-resident of Japan, there will be no taxation in Japan because the capital gain is attributed to the non-resident taxpayer. Such an aggressive form of tax minimizing or avoiding strategies by changing tax residence has monetary meaning to only a small number of very-high-income earners. In order to achieve optimal taxation of IIT, both from the viewpoints of policy making and tax administration, it would be useful to pay attention to: (1) the behavior of groups of high-income earners, as they bring a large amount of tax revenue to the government; (2) tax policy and tax administration concerning capital income, as the proportion of capital income is large for high-income earners; and (3) the tax minimizing activities of high-income earners (i.e., aggregation of profits and losses 18 and cross-border transactions 19 ). 18 For example, marketable tax shelters using aggregation of loss from offshore partnerships and funds with domestic income are known and currently pending at the Supreme Court. 19 For example, mandatory offshore reporting obligations were introduced from CY2013. This obliges residents who have offshore assets (all types of assets, e.g., financial assets and real property) exceeding 50 million yen as of the end of a year to file an information return by the 15th of March of the next year.

30 444 N Oka / Public Policy Review Annex A Share (percentage) of taxpayers, income and tax by income groups of taxable taxpayers calculated from data in the NTA Shinsho survey The NTA has published the Shinsho income tax survey each year since This annex illustrates the share (percentage) of taxpayers, income and tax by income groups in 1999 and This annex also discusses effective (average) tax rates in each year and their variation between 2006 and Figures and tables in this annex were compiled from the Table 1 (Overview) of Shinsho income tax survey for each year. Table 1 (overview) is also found in the National Tax Agency Annual Statistics Report (2-4 Results of Sample Survey for Self-assessment income tax (excerpt)) Year Link Page a/01.pdf Below are some of the key findings from tables and figures in this annex. 1. The amount of total amount of income (TAI) is concentrated among taxable taxpayers with income exceeding 50 million yen. See Figure The concentration has intensified. For example, in 2011, only 0.7% of taxable taxpayers earned 14.5% of the total amount of income and paid 28.1% of income tax of all taxable taxpayers. See Figures 14, 15 and The effective (average) tax rates of very high income earners have increased in recent years. See section (g) of Table Effective (average) tax rates are the ratios of a computed amount of tax to the amount of total income of taxable taxpayers in the Table 1 Overview of the Shinsho survey. Between 2006 and 2011, effective tax rates decreased for all taxpayers with a total amount of income below 10 million yen. The largest decrease of the effective tax rate (-2,31%) was reported by taxable taxpayers with an amount of TAI between 2.5 to 3 million yen. The largest increase of effective tax rate (+4.49%) was reported by taxable taxpayers with an amount of TAI exceeding 1 billion yen. See Tables 16 and 17.

31 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 14 Share (percentage) of taxpayers by income groups (Source) Adopted from table 1 (overview) of Shinsho income survey by the NTA Table 15 Share (percentage) of taxable taxpayers by income group Money amounts are in million of yen (Source) Adopted from table 1 (overview) of Shinsho income survey by the NTA

32 446 N Oka / Public Policy Review Figure 15 Share (percentage) of income by income group (Source) Adopted from table 1 (overview) of Shinsho income survey by the NTA

33 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 16 Share (percentage) of tax by income group

34 448 N Oka / Public Policy Review Table 16 Effective (average) tax rate by income group NB. The lowest tax rate (applicable to an amount of taxable income less than 1,950,000 yen) was reduced from 10% to 5% in 2007.

35 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Table 17 Change of effective (average) tax rate

36 450 N Oka / Public Policy Review Annex B History of the statutory top tax rate for regular income and capital gains in Japan and the US 70 Figure 17 Top statutory tax rate (Japan, National Government) (%) Figure 18 Top statutory tax rate (Japan, local governments) (%)

37 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October Figure 19 Top statutory tax rate since (US) (%) Table 18. Overview of capital income taxation (high income earners) Note: Regular top statutory rates are 50% (2014) and 55% (after 2015) for aggregate income exceeding 18 million yen. Dividends from listed stocks were taxed at 10% until the end of 2013 (by election of taxpayers). Capital gains from the sale of listed stocks were taxed at 10%. Tax rates include national and local income taxes.

38 Supplementary Annex 1: Simplified overview of calculation of income tax in Japan (as of CY 2014) %( ) ( For more information about technical information regarding calculation of income tax in Japan, please see a guidance ( income tax and special income tax for reconstruction guide ) published by the NTA N Oka / Public Policy Review

39 Supplementary Annex 2: Method of taxation, Aggregation of profits and losses, and tax rates (1) Loss from the sale of dowelling house (2) Loss from the sale of listed stocks may be aggregated with dividend income from listed stocks (3) Loss from the sale of listed stocks may be aggregated with dividend income from listed stocks, if deposited at a special account of financial institutions in Japan. NB. This table summarizes tax treatment of items of income that are common to high income earners. Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October

40 454 N Oka / Public Policy Review Supplementary Annex 3 : Glossary 総所得金額 合計所得金額 ( 所 () 法 ) 課税標準 所法 損益通算 申告納税額 事業所得 不動産所得 利子所得 配当所得 給与所得 雑所得 譲渡所得 ( 総合 ) 一時所得 譲渡所得 ( 分離 ) 商品先物取引所 得 株式等譲渡所得 山林所得 退職所得 有価証券の譲渡による所得の課税の特例等 雑損控除 医療費控除 社会保険料控除 小規模企業共済等控除 生命保険料控除 損害保険料控除 地震保険料控除 寄付金控除 勤労者学生障害者控除 配偶者控除 配偶者特別控除 扶養控除 基礎控除 配当控除 外国税額控除

41 Policy Research Institute, Ministry of Finance, Japan, Public Policy Review, Vol.10, No.3, October References NTA National Tax Agency Annual Statistics Report NTA Income tax and special income tax for reconstruction guide (for foreigners) Shigeki Morinobu Capital Income Taxation and the Dual Income Tax Policy Research Institute, Ministry of Finance of Japan (2004) Adrian Dungan and Michael Parisi, Individual Income Tax Rates and Shares, 2010 Statistics of Income Bulletin, Winter 2013, pp The President s Advisory Panel on Federal Tax Reform Simple, Fair, and Pro-Growth: Proposals to Fix America s Tax Nov Growth-Proposals-to-Fix-Americas-Tax-System pdf The Congressional Budget Office Trends in the Distribution of Household Income Between 1979 and The IRS The 400 Individual Income Tax Returns Reporting the Largest Adjusted Gross Incomes Each Year, The Congressional Budget Office Historical Effective Federal Tax Rates:1979 to historicaltaxrates.pdf Yankelovich, Skelly and White, Inc. Taxpayer attitudes study : final report / prepared for Internal Revenue Service December 1984 Isaac Shapiro and Joel Friedman New, unnoticed CBO data show capital income has become much more concentrated at the top January

42