High-income tax returns for 2013.
Two income concepts are used in this article to classify tax returns as high income: the statutory concept of adjusted gross income (AGI) and the "expanded income" concept. (2) The expanded income concept uses items reported on the tax return to obtain a more comprehensive measure of income than AGI. Specifically, expanded income is AGI plus tax-exempt interest, nontaxable Social Security benefits, the foreign-earned income exclusion, and items of "tax preference" for alternative minimum tax (AMT) purposes less unreimbursed employee business expenses, moving expenses, investment interest expense to the extent it does not exceed investment income, and miscellaneous itemized deductions not subject to the 2-percent-of-AGI floor. (3, 4, 5)
AGI and expanded income differed for 43.9 million (29.8 percent) of the 147.4 million individual income tax returns filed for 2013 (Figure A). Expanded income exceeded AGI in two-thirds of these returns. The average difference between expanded income and AGI was $6,545, and the median difference was $3,392. Although expanded income is a more comprehensive measure of income than AGI, for some taxpayers the subtractions from AGI to arrive at expanded income exceed the additions, resulting in expanded income that is less than AGI. Returns where expanded income exceeded AGI were concentrated among returns with $50,000 or less of AGI. Returns where AGI exceeded expanded income were concentrated among returns with more than $50,000 of AGI.
There are also two tax concepts in this article used to classify returns as taxable or nontaxable: "U.S. income tax" and "worldwide income tax." U.S. income tax is total Federal income tax liability, which includes the AMT, less all credits against income tax, and does not include payroll or self-employment taxes. To be considered taxable, a return had to have a positive income tax liability after accounting for all credits (including refundable credits). A nontaxable return, on the other hand, could either have a zero or negative income tax liability after accounting for all credits (including refundable credits). Since the Federal income tax applies to worldwide income and allows a credit (subject to certain limits) for income taxes paid to foreign governments, a return could be classified as nontaxable under the U.S. income tax concept even though income taxes had been paid to a foreign government. Worldwide income tax addresses this circumstance by adding back the allowable foreign tax credit and foreign taxes paid on excluded foreign-earned income to U.S. income tax. (6, 7) The sum of these two items is believed to be a reasonable proxy for foreign taxes actually paid.
Figure B shows a cross-tabulation of tax returns based on whether a return has positive, zero, or negative U.S. income tax liability and whether it also has positive, zero, or negative worldwide income tax. By definition, returns with positive U.S. income tax liability have positive worldwide income tax liability; returns with zero U.S. tax liability have either positive or zero worldwide income tax liability; and returns with negative U.S. income tax liability have either positive or negative worldwide income tax liability. Almost 36 percent of the 147.4 million tax returns filed for 2013 had zero or negative U.S. income tax liability, and 97.0 percent of these returns had AGI of $50,000 or less--unsurprising given the progressive nature of the U.S. income tax. Only 3 out of every 1,000 returns with zero or negative U.S. income tax liability had positive worldwide income tax liability, and 32.2 percent of these returns had AGI of $50,000 or more.
Number of High-Income Returns
For 2013, there were nearly 5.6 million individual income tax returns with an AGI of at least $200,000, and slightly more than 5.6 million returns with an expanded income of $200,000 or more (Figure C). The number of returns with AGI above the $200,000 threshold increased 5.8 percent from 2012 and accounted for 3.8 percent of all returns for 2013. Similarly, the returns with expanded incomes above the $200,000 threshold also increased 5.8 percent from 2012 and also accounted for 3.8 percent of all returns for 2013.
The $200,000 threshold for high-income returns is measured in current-year (nominal) dollars. Because it is fixed in nominal terms, as time passes, more and more tax returns have had incomes exceeding this threshold simply due to inflation. Therefore, to facilitate a comparison of tax returns across time, Figure C shows what the $200,000 threshold in 1976 would be each year after adjusting for inflation and the number and share of returns above this threshold. (8) For Tax Year 2013, this inflation-adjusted threshold was $818,830.
Adjusting for inflation, the number of returns above the AGI threshold was 470,906 for Tax Year 2013--0.32 percent of all returns--a 12.0-percent decrease from the previous year. Since 1977, the number of returns above this inflation-adjusted threshold has increased at a rate of approximately 6.7 percent per year. In comparison, the total number of returns has increased 1.5 percent per year since 1977. Similarly, the number of returns above the inflation-adjusted threshold using the expanded income concept was 480,987 for Tax Year 2013--0.33 percent of all returns--an 11.8-percent decrease from the previous year. Since 1977, the number of returns above this threshold has increased 6.0 percent per year.
From 1977 to 2013, the number of returns reporting incomes of $200,000 or more generally increased annually. The exceptions to this reflect the economic downturns in Tax Years 2001, 2002, 2008, and 2009. The general trend of a rise in the number of returns is similar when using the inflation-adjusted income threshold. However, the increases are much smaller, and the number of returns above the inflation-adjusted threshold decreased in additional years, e.g., Tax Years 1980, 1981, 1989, 1990, 1991, 1993 and 2013. Many of these years also coincided with economic downturns. (9)
The difference in the number of high-income returns between the two income concepts significantly decreased beginning with 1987, when AGI began to include 100 percent of long-term capital gains. This change in the definition of AGI made it more comparable to the expanded-income concept. In addition, due to the inclusion of tax-exempt interest in expanded income in 1987, expanded income for years after 1986 is not strictly comparable to expanded income for years before 1987.
Nontaxable High-Income Returns
Of the 5.6 million income tax returns with an AGI of $200,000 or more, 12,517 showed no U.S. income tax liability for Tax Year 2013 (top panel of Figure D). Also, of the 5.6 million tax returns with expanded income of $200,000 or more, 25,926 had no U.S. income tax liability. This represents a 38.4-percent decrease for high-AGI returns and 19.8-percent decrease for high expanded-income returns with no U.S. income tax from the number of returns for 2012.
Of the returns with an AGI of $200,000 or more, 4,266 returns showed no worldwide income tax liability for Tax Year 2013. For returns with an expanded income of $200,000 or more, 12,794 had no worldwide income tax liability. This represents a 42.4-percent decrease for high-AGI returns and 4.9-percent decrease for high expanded-income returns with no worldwide income tax, compared to the number of returns for 2012. The decrease for the high-AGI returns without worldwide income tax represents the largest decrease since 1979. One possible explanation for this decrease was the inclusion of the net investment income tax (NIIT), which added a 3.8-percent tax on investment income for high-income taxpayers after total tax credits were subtracted from income tax. (10) Tax-exempt interest was not subject to the NIIT and may explain why the decrease in high expanded-income returns without worldwide income tax was much smaller.
Tables 1 and 2 also show the number of all returns, taxable returns, and nontaxable returns cross classified by broad AGI and expanded-income size classes. Most returns fall in the same broad income-size class under both income concepts, but the number of nontaxable returns under the expanded-income concept is generally greater than the AGI concept in each income class over $50,000. They also show that there were 5.5 million returns with incomes of $200,000 or more as measured by both AGI and expanded income, of which 10,656 returns had no U.S. income tax liability and 2,494 returns had no worldwide income tax.
Of the roughly 471,000 returns with an AGI over the inflation-adjusted high-income threshold, 901 returns had no U.S. income tax liability in 2013, down 60.8 percent from the previous year (lower panel of Figure D). Additionally, 217 returns showed no worldwide income tax liability, down 59.4 percent from 2012. Measured using expanded income, of the almost 481,000 returns above the inflation-adjusted threshold, 1,158 returns had no U.S. income tax liability, a 52.7-percent decrease from the number of returns for 2012, as well as 398 returns that showed no worldwide income tax liability, down 27.2 percent from the previous year.
[FIGURE E OMITTED]
Moving from the nominal $200,000 threshold to an inflation-adjusted threshold does not generally change substantially the share of high-income returns that are nontaxable. Using AGI, 0.2 percent of high-income returns had no U.S. income tax, using both a nominal $200,000 threshold and an inflation-adjusted threshold, while 0.08 percent of high-income returns over the nominal threshold and 0.05 percent of high-income returns above the inflation-adjusted threshold had no worldwide income tax for 2013. When using expanded income with a nominal $200,000 threshold, 0.5 percent of high-income returns had no U.S. income tax, and 0.2 percent had no worldwide income tax for 2013, while an inflation-adjusted threshold showed that 0.2 percent had no U.S. income tax and 0.1 percent had no worldwide income tax.
Figure E shows the evolution of the number and share of returns with an expanded income of $200,000 or more with no worldwide income tax. The number of returns above the threshold is read off the left vertical axis, and the share of returns above the threshold is read off the right vertical axis. Two series are shown for each measure, one using a nominal $200,000 threshold and one using an inflation-adjusted threshold.
In this figure, the spread between the two shares was small for the late 1970s, showed an increase for the early 1980s, and then narrowed before widening again after 1988. The spread generally narrowed after 1993 but increased or stayed fairly consistent from 2002 to 2007. From 2008 to 2012, the spread widened considerably. The gap (almost 0.15 percentage points) for 2013 was still high compared with most of the years in this study. (11)
Reasons for Nontaxability
Taxpayers may have their tax liability reduced--possibly to zero or beyond--in a number of ways. These include claiming tax credits, which directly reduces an individual's tax liability; claiming various deductions, which reduces taxable income; and receiving income that is excluded from the calculation of taxable income.
It is possible for certain itemized deductions and certain exclusions from income to lead to nontaxability by themselves, but high-income returns are more often nontaxable for a combination of reasons, none of which alone would result in nontaxability. Moreover, some items, either singly or in combination, may eliminate "regular tax" liability (i.e., income tax excluding the AMT), but cannot eliminate an AMT liability, since these items give rise to adjustments or preferences for AMT purposes.
Because they do not generate AMT adjustments or preferences, tax-exempt bond interest (not including private activity bonds), itemized deductions for interest expenses, miscellaneous itemized deductions not subject to the 2-percent-of-AGI floor, casualty or theft losses, and medical expenses (exceeding 10 percent of AGI) could, by themselves, produce nontaxability.
Due to the AMT exemption of $80,800 for joint returns ($51,900 for single and head-of-household returns and $40,400 for returns of married taxpayers filing separately), a return could have been nontaxable even though it included some items that produced AMT adjustments or preferences. (12) Further, since the starting point for "alternative minimum taxable income" was taxable income for regular tax purposes, a taxpayer could have adjustments and preferences exceeding the AMT exclusion without incurring AMT liability. This situation could occur if taxable income for regular tax purposes was sufficiently negative due to itemized deductions and personal exemptions exceeding AGI, such that the taxpayer's AMT adjustments and preferences are less than the sum of the AMT exclusion and the amount by which regular taxable income is below zero. Note that, because of the AMT, taxpayers may have found it beneficial to report additional deduction items on their tax returns, even if the items did not produce a benefit for regular tax purposes.
The most important item in eliminating tax on the 4,266 returns without any worldwide income tax and AGI of $200,000 or more was total miscellaneous deductions. This was the case in 34.2 percent (1,458 returns) of those returns (Figure F). The next three categories that most frequently had the largest effect in eliminating taxes were: 1) charitable contributions deduction (16.0 percent or 684 returns); 2) medical and dental expense deduction (15.8 percent or 672 returns); and 3) investment interest expense deduction (8.4 percent or 360 returns). The items that most frequently had the second largest effect in reducing regular tax liability for high-AGI returns with no worldwide income tax were the deduction for taxes paid (37.6 percent or 1,606 returns) and the interest paid deduction (15.7 percent or 671 returns).
Of the 12,794 returns without any worldwide income tax and expanded incomes of $200,000 or more, the most important item in eliminating tax, on 56.0 percent of returns, was the exclusion for interest income on State and local Government bonds ("tax-exempt interest") (Figure G). The next three categories that most frequently had the largest effect in eliminating taxes were: 1) medical and dental expense deduction (13.1 percent or 1,676 returns); 2) charitable contributions deduction (9.6 percent or 1,231 returns); and 3) taxes paid deduction (6.5 percent or 830 returns). The items that most frequently had the second largest effect in reducing regular tax liability on high expanded-income returns with no worldwide income tax were the deduction for taxes paid (27.9 percent or 3,574 returns) and medical and dental expense deduction (12.9 percent or 1,655 returns).
Tables 7 and 8 in this article also classify tax returns by items having the largest and second largest effects in reducing or eliminating income tax. For returns on which each of the largest effects was identified, the tables show each of the second largest effects as well. (13) For example, Table 7 shows that on taxable returns with some U.S. income tax and expanded incomes of $200,000 or more, the taxes paid deduction was the most important item 62.9 percent of the time. Where this was the primary item, the interest paid deduction was the second most important item 59.3 percent of the time, and the charitable contributions deduction was the second most important item 26.3 percent of the time. The foreign tax credit was the largest reason for nontaxability for both AGI (62.5 percent of the time) and expanded-income returns with no income tax (38.9 percent of the time). However, in determining worldwide tax, this no longer has an effect (by definition).
Figure H presents another way of illustrating the importance of selected tax provisions in reducing or eliminating income tax liability. It shows the number of high expanded-income tax returns with no worldwide income tax utilizing selected tax provisions by the percentage of income reduced. For example, the itemized deduction for medical and dental expenses reduced income by greater than 60 percent of expanded income on 1,226 returns (or 9.6 percent) of the 12,794 returns, but there were no medical and dental expense deductions on 5,574 returns or 43.6 percent. Conversely, the taxes paid deduction reduced income between 0 and 30 percent of expanded income on 10,093 returns (or 78.9 percent) of the 12,794 returns, but reduced income by greater than 60 percent on only 547 returns. Tables 9 and 10 report the frequencies of returns in finer detail by the percentage of income reduced.
Distribution of Tax Levels
Figure I shows the distribution of high expanded-income returns by the ratio of "adjusted" worldwide taxable income to expanded income. Taxable income was adjusted for this figure by subtracting the deduction equivalents of tax credits and other items from taxable income. (14) Thus, the figure shows the extent to which expanded income is reduced before taxes are imposed on the remaining income. The figure illustrates three important facts about high-income tax returns:
* Only a small portion of all high-income taxpayers did not pay any worldwide income taxes (0.2 percent).
* Another group of high-income taxpayers--small, but larger than the nontaxable group--was able to offset a very substantial fraction of income before being subject to tax. This type of high-income taxpayer pays income tax equal to only a small share of his or her income. Such taxpayers may be called "nearly nontaxable." About 0.6 percent of high expanded-income taxpayers who reported at least some worldwide tax liability were able to reduce their adjusted taxable income to less than 25 percent of their expanded income.
* Overall, most high-income taxpayers were subject to tax on a large share of their income (68.7 percent of high expanded-income taxpayers had adjusted taxable income equal to 80 percent or more of expanded income; and 96.9 percent had adjusted taxable income equal to 50 percent or more of expanded income).
Figure J shows the distribution of all expanded-income returns by worldwide income tax burden. It classifies tax returns by size of expanded income and effective tax rate, i.e., income tax as a percentage of expanded income. This figure illustrates the wide dispersion of effective tax rates for all returns. For example, while 2.9 percent of returns with expanded incomes of $200,000 or more had worldwide income tax of less than 10 percent of income, 21.3 percent had effective tax rates of 25 percent or more. Also, 28.5 percent had effective tax rates between 20 and 25 percent. In contrast, only 2.5 percent of taxpayers with an expanded income between $100,000 and $200,000 had effective tax rates of 20 percent or more, including only 0.1 percent with effective tax rates of 25 percent or more.
Characteristics of Tax Returns
By comparing nontaxable returns with taxable returns, some of the different characteristics of nontaxable returns can be deduced. For example, under the expanded-income concept, returns without worldwide income tax were much more likely (87.0 percent) than taxable returns (27.3 percent) to have tax-exempt interest, and when they did, the average amount for nontaxable returns ($234,576) was much higher than the average amount for all taxable returns ($28,794) (Figure K). Similarly, nontaxable returns (37.4 percent) were much less likely than high-income taxable returns (87.5 percent) to have income from salaries and wages. Tables 5 and 6 show the aggregate frequencies and amounts of all the types of income, the items of tax preference, and the various deductions, credits, and income taxes on high-income returns.
More Detailed Data for 2013
Tables 1 through 12 present data based on income tax returns for 2013, mainly those with income of $200,000 or more (measured in current-year dollars) of AGI or expanded income. Most of the data are shown for taxable and nontaxable returns, both separately and combined. In summary, the tables show:
* The numbers of returns under the two tax concepts, cross-classified by broad AGI and expanded-income size classes (Tables 1 and 2);
* The distributions of taxable income as a percentage of AGI and expanded income (Tables 3 and 4);
* The frequencies and amounts of various sources of income, exclusions, deductions, taxes, and tax credits, as well as the relationship between the two income concepts (Tables 5 and 6);
* The frequencies with which various deductions and tax credits are the most important and second most important items in reducing (or eliminating) income tax (Tables 7 and 8);
* The frequencies with which various itemized deductions, tax credits, and tax preference items occur as certain percentages of income (Tables 9 and 10); and
* The distributions of effective tax rates, i.e., income tax under each definition as a percentage of income as well as the percentage of income that is subject to preferential tax rates, by broad income-size classes (Tables 11 and 12).
The odd-numbered tables use the U.S. income tax concept to classify returns as taxable or nontaxable, whereas the even numbered tables use the worldwide income tax concept.
Lerman, Allen H., "High-Income Tax Returns: 1974 and 1975, A Report on High-Income Taxpayers Emphasizing Tax Returns with Little or No Tax Liability," U.S. Department of Treasury, Office of Tax Analysis, March 1977, and "High-Income Tax Returns: 1975 and 1976, A Report Emphasizing Nontaxable and Nearly Nontaxable Income Tax Returns," U.S. Department of Treasury, Office of Tax Analysis, August 1978.
U.S. Department of Treasury, Internal Revenue Service, Statistics of Income--Individual Income Tax Returns for 1977 through 1982 and 1985 through 1988. (For 1977 and 1978, only the number of nontaxable, high-AGI returns was published.)
Lerman, Allen H., "High-Income Tax Returns, 1983," Statistics of Income Bulletin, Spring 1986, Volume 5, Number 4, pp. 31-61; "High-Income Tax Returns, 1984," Statistics of Income Bulletin, Spring 1987, Volume 6, Number 4, pp. 1-29; "High-Income Tax Returns for 1989," Statistics of Income Bulletin, Spring 1993, Volume 12, Number 4, pp. 23-50; "High-Income Tax Returns for 1990," Statistics of Income Bulletin, Winter 1993-1994, Volume 13, Number 3, pp. 104-132; "High-Income Tax Returns for 1991," Statistics of Income Bulletin, Winter 1994-1995, Volume 14, Number 3, pp. 96-130; and "High-Income Tax Returns for 1992," Statistics of Income Bulletin, Winter 1995-1996, Volume 15, Number 3, pp. 46-82.
Latzy, John, "High-Income Tax Returns for 1993," Statistics of Income Bulletin, Winter 1996-1997, Volume 16, Number 3, pp. 64-101; and "High-Income Tax Returns, 1994," Statistics of Income Bulletin, Winter 1997-1998, Volume 17, Number 3, pp. 31-69.
Cruciano, Therese, "High-Income Tax Returns for 1995," Statistics of Income Bulletin, Summer 1998, Volume 18, Number 1, pp. 69-108; "High-Income Tax Returns for 1996," Statistics of Income Bulletin, Winter 1998-1999, Volume 18, Number 3, pp. 7-59.
Parisi, Michael, "High-Income Tax Returns for 1997," Statistics of Income Bulletin, Winter 1999-2000, Volume 19, Number 3, pp. 6-58.
Balkovic, Brian, "High-Income Tax Returns for 1998," Statistics of Income Bulletin, Winter 2000-2001, Volume 20, Number 3, pp. 5-57; "High-Income Tax Returns for 1999, " Statistics of Income Bulletin, Spring 2002, Volume 21, Number 4, pp. 7-58; "High-Income Tax Returns for 2000, " Statistics of Income Bulletin, Spring 2003, Volume 22, Number 4, pp. 10-62; "High-Income Tax Returns for 2001," Statistics of Income Bulletin, Summer 2004, Volume 24, Number 1, pp. 65-117; "High-Income Tax Returns for 2002," Statistics of Income Bulletin, Spring 2005, Volume 24, Number 4, pp. 6-58; "High-Income Tax Returns for 2003," Statistics of Income Bulletin, Spring 2006, Volume 25, Number 4, pp. 8-57; "High-Income Tax Returns for 2004," Statistics of Income Bulletin, Spring 2007, Volume 26, Number 4, pp. 7-57; "High-Income Tax Returns for 2005," Statistics of Income Bulletin, Spring 2008, Volume 27, Number 4, pp. 16-67.
Bryan, Justin, "High-Income Tax Returns for 2006," Statistics of Income Bulletin, Spring 2009, Volume 28, Number 4, pp. 5-53; "High-Income Tax Returns for 2007," Statistics of Income Bulletin, Spring 2010, Volume 29, Number 4, pp. 3-51; "High-Income Tax Returns for 2008," Statistics of Income Bulletin, Spring 2011, Volume 30, Number 4, pp. 5-54; "High-Income Tax Returns for 2009," Statistics of Income Bulletin, Spring 2012, Volume 31, Number 4, pp. 6-61; "High-Income Tax Returns for 2010," Statistics of Income Bulletin, Spring 2013, Volume 32, Number 4, pp. 4-58; "High-Income Tax Returns for 2011," Statistics of Income Bulletin, Spring 2014, Volume 33, Number 4, pp. 51-110; "High-Income Tax Returns for 2012," Statistics of Income Bulletin, Summer 2015, Publication 1136, online version only (https://www.irs.gov/pub/irs-soi/soi-a-inhint-id1510.pdf).
Appendix A: Income Concepts
Congress wanted data on high-income taxpayers classified by an income concept that was more comprehensive than adjusted gross income (AGI), but that was based entirely on items already reported on income tax returns. To derive such an income concept, it was necessary to begin with a broad, inclusive concept of income. AGI must then be compared to this broad income concept, and the differences identified (both additions and subtractions) from items reported on tax returns.
This appendix begins by defining "Haig-Simons income," a very broad concept of income used by economists and others as a standard. AGI is then compared to Haig-Simons income, and the major differences between the two income concepts are listed. The next section defines "expanded income," a more comprehensive income measure than AGI, based entirely on tax return data, and the final section explains in some detail the adjustment to income for investment expenses.
The broadest measure of annual income generally used by economists and others is defined as the value of a household's consumption plus the change, if any, in its net worth. This income concept is referred to as Haig-Simons income, or H-S income, after the two economists who wrote extensively about it [A1]. The H-S income of a household that consumed $25,000 and saved $2,000 in a year would be $27,000. Alternatively, the H-S income of a household that consumed $25,000 and had no additions to savings, but had assets that declined in value by $1,000 in a year, would be $24,000.
H-S income consists of three broad components: labor income, capital income (income from assets), and income from transfer payments. The major elements of each of these three components are as follows:
Labor income--This includes all forms of employee compensation (including wages and salaries), employee fringe benefits (such as employer-provided health insurance and accrued pension benefits or contributions), and the employer share of payroll taxes (such as Social Security taxes). Labor income also includes the labor share of self-employment income. Expenses of earning labor income would be deducted in arriving at H-S income. Deferred labor income (such as pension benefits) would be counted in the year it was earned, rather than in the year it was received.
Capital income--This includes all income from assets, including interest, dividends, rents, royalties, accrued capital gains (whether or not realized), the capital income share of self-employment income, and the rental value of consumer durables (most importantly, the rental value of owner-occupied housing). Capital income is measured in real (inflation-adjusted) terms and is net of real, economic depreciation and all other expenses (which could exceed capital income).
Transfer payments--These include payments in cash such as Social Security benefits, workers' compensation, unemployment benefits, Aid to Families with Dependent Children (AFDC), and noncash benefits (such as Medicare, Medicaid, and the Supplemental Nutrition Assistance Program (SNAP)).
For purposes of tax analysis, H-S income should be measured on a pretax basis, the amount that would be earned if there were no Federal income tax in place. Most items of income are unaffected, or little affected, by the income tax and so these are reported on a pretax basis. However, certain income items from tax-preferred sources may be reduced because of their preferential treatment. An example is interest from tax-exempt State and local Government bonds. The interest rate on tax-exempt bonds is generally lower than the interest rate on taxable bonds of the same maturity and risk, with the difference approximately equal to the tax rate of the typical investor in tax-exempt bonds. Thus, investors in tax-exempt bonds are effectively paying a tax, referred to as an "implicit tax," and tax-exempt interest as reported is measured on an after tax, rather than a pretax, basis. Income from all tax-preferred sources should be "grossed up" by implicit taxes to measure H-S income properly.
Adjusted Gross Income
AGI is the statutory definition of income for Federal income tax purposes. AGI differs from H-S income by excluding some components of H-S income and by allowing accelerated business deductions and deductions unrelated to income, but also by disallowing or limiting certain expenses of earning income and certain losses. In addition, AGI is not "grossed up" for implicit taxes.
The components of H-S income excluded from AGI include most employee fringe benefits, the employer share of payroll taxes, accrued but deferred employee compensation, accrued but unrealized real capital gains, the rental value of consumer durables, nontaxable Social Security benefits, most other cash transfers, all noncash transfers, and the real income of borrowers due to inflation [A2].
Depreciation and certain other expenses allowed in determining AGI may be accelerated (relative to economic depreciation and other costs) in the early years of an investment, thus understating investment income. In later years, however, investment income in AGI will be overstated because depreciation and other accelerated expenses will then be understated. AGI also excludes some expenses not related to earning income, such as contributions to self-employed retirement (Keogh) plans, deductible contributions to Individual Retirement Arrangements (IRAs), the portion of Social Security contributions for self-employed workers that is analogous to the employer share of such contributions for employees, and contributions to medical savings accounts.
AGI generally exceeds H-S income to the extent that expenses of earning income and losses are limited or disallowed. Most of the expenses of earning income are deductible from AGI in calculating taxable income, but only if the taxpayer "itemizes" deductions and then, in some cases, only to the extent that the sum of all such items exceeds 2 percent of AGI. Expenses incurred in the production of income that are itemized deductions include certain expenses of employees (such as union dues; expenditures for items used on the job but not reimbursed by the employer; and the employees' travel, meal, and entertainment expenses); and expenses attributable to a taxpayer's (passive) investments (as opposed to active participation in a trade or business, for example), including, but not limited to, interest expense incurred in connection with investments in securities [A3]. Note that there are limits on certain types of deductible expenses. In particular, deductible meal and entertainment expenses are limited to 50 percent of total meal and entertainment expenses.
Although net capital losses reduce economic income, only the first $3,000 of net realized capital losses may be deducted in computing AGI. Any additional realized losses must be carried forward to future years. In a somewhat similar manner, passive losses (from investments in a trade or business in which the taxpayer does not materially participate) can also reduce economic income, but, in computing AGI, they can only be deducted from passive income from other, similar investments (although a larger amount may be deducted when the losses are from rental real estate activities).
AGI can also exceed H-S income because of differences in the timing of income between the two concepts. For example, a taxpayer may realize more capital gains in a year than he or she accrues in capital gains. Since AGI includes only realizations of capital gains, whereas H-S income includes only accruals, AGI in this circumstance would exceed H-S income.
Finally, just as AGI understates the income of borrowers due to inflation, it overstates the income of lenders, who include bond owners and owners of bank deposits.
Expanded income is meant to be a measure of income that is conceptually closer to H-S income than AGI, but which is derived entirely from items already reported on income tax returns. Figure L shows the adjustments made to AGI to arrive at expanded income. Since the definition of AGI was changed by legislation several times since 1977, and certain reporting requirements also changed, the adjustments differ over the years, as indicated for each item [A4]. Most of these adjustments are relatively straightforward, but the adjustment for investment requires some explanation.
In measuring H-S income, it generally would be appropriate to deduct all expenses incurred in the production of income, including those related to any income-producing investments, without limit. Investment expenses in excess of investment income would then represent net economic losses. However, such a liberal deduction for investment-related expenses is not necessarily correct when not all income items have been included currently. (Investment income includes interest, dividends, and capital gains.)
If all income has not been included currently, full deduction of investment expenses might represent a mismatching of receipts and expenses and might result in understating income. For example, if a taxpayer borrowed funds to purchase securities, net income would be understated if the taxpayer deducted all interest payments on the loan, but did not include as income any accrued gains on the securities. A similar mismatching of income and expenses would occur if investment expenses that should properly be capitalized were deducted when paid. In these instances, a more accurate measure of income might be obtained by postponing the deduction of the expense until the income was recognized for tax purposes.
Additional problems are created when a person with a loan has both income-producing assets, such as securities, and non-income-producing assets, such as a vacation home or yacht. It is not possible to determine what portion of the interest expense should be attributed to taxable income-producing assets, and therefore, ought to be deductible against the gross receipts from such taxable assets. As a result of these problems, it has been necessary to set arbitrary limits on the amount of investment expenses that are deductible in calculating expanded income.
Investment expenses that have not been deducted in determining AGI generally can appear on a Federal individual income tax return in two places. Investment interest expense is taken into account in the calculation of the itemized deduction for interest paid. Deductible investment interest expense is a separate part of the total interest deduction. Other investment expenses, such as management fees, are included in the miscellaneous category of itemized deductions [A5]. Beginning with 1987, most types of income-producing expenses included as miscellaneous itemized deductions are only deductible to the extent that their total exceeds 2 percent of AGI. To determine expenses that should be deductible in calculating an approximation of H-S income, investment expenses have been defined as deductible investment interest expense. Other investment expenses could not be separated from the remainder of miscellaneous deductions. Hence, they have not been used in the adjustment for investment expenses.
To the extent that interest expenses do not exceed investment income, they are generally allowed as a deduction in the computation of deductible investment interest expense and thus expanded income. Investment interest expenses that exceed investment income are not deductible in calculating expanded income. One consequence of this definition is that investment expenses can never turn positive investment income into investment losses. Generally, allowing investment expenses to offset all investment income is generous and tends to understate broadly measured income. However, in some instances, limiting investment expenses to investment income may overstate income by disallowing genuine investment losses.
Notes to Appendix A
[A1] Haig, Robert M. (ed.), The Federal Income Tax, Columbia University Press, 1921, and Simons, Henry C., Personal Income Taxation, University of Chicago Press, 1938.
[A2] Borrowers receive income due to inflation because the real value of debt is reduced by inflation. Even though inflation may be anticipated and reflected in interest rates, tax deductions for nominal interest payments overstate interest costs because part of these payments represent a return of principal to the lender, rather than interest.
[A3] See references and footnote A4.
[A4] For 1977, about 50 percent of net long-term capital gains were included in AGI. During 1978, the inclusion ratio was changed to 40 percent. This inclusion ratio remained unchanged through 1986. Beginning with 1987, there was no exclusion allowed for capital gains in computing AGI, and, thus, this adjustment was not made in computing expanded income for returns for years after 1986.
Beginning in 1987, taxpayers were required to report on their Federal income tax returns the amount of their tax-exempt interest income from State and local Government bonds. Since 1987, tax-exempt interest has been included in expanded income.
Taxpayers are also required to report Social Security benefits. Since 1988, nontaxable Social Security benefits have been included in expanded income. However, if none of a particular taxpayer's Social Security benefits are taxable, then gross Social Security benefits are not required to be shown on the income tax return. In such instances, which generally only affect lower- and middle-income taxpayers, Social Security benefits are not included in expanded income.
The subtraction of unreimbursed employee business expense and the moving expense deduction is to make the concept of expanded income comparable to years prior to 1987. All current-year moving expenses beginning with Tax Year 1994 were deducted in the calculation of AGI as a statutory adjustment. Due to subtracting non-limited miscellaneous deductions and not subtracting the nondeductible rental loss for 1989, the expanded income concept for 1989 is not strictly comparable to expanded income for 1988. Nor is the expanded income concept for 1990 strictly comparable to expanded income for 1989 because of the addition of the foreign-earned income exclusion. Specific details on the definition of expanded income for any given year are available in the reports and publications found under the References section.
[A5] Some income deferrals and accelerated expense deductions may also be involved in income or losses from rental property, from royalties, from partnerships, and from S Corporations, only the net amounts of which are included in adjusted gross income.
Appendix B: Tax Concepts
This appendix discusses in more detail two tax concepts used in this article. The first section provides a brief summary of the U.S. taxation of worldwide income and the foreign tax credit. The two tax concepts used in this article are then defined in the next section. That section is followed by an explanation of deduction equivalent of credits and other items. A final section discusses the possible implications of the use of unaudited tax return data for this article.
U.S. Taxation of Worldwide Income and the Foreign Tax Credit
Citizens and residents of the United States, regardless of where they physically reside, must generally include in income for Federal income tax purposes income from all geographic sources. Thus, for example, dividends and interest received from a foreign corporation or income earned working abroad is subject to Federal income tax in the same manner as income received from sources inside the United States [B1]. Income from sources outside the United States may also be subject to tax by foreign governments.
To reduce, if not eliminate, the possibility of double taxation of the foreign-source income of U.S. citizens and residents, the Federal income tax allows a credit for income taxes paid to foreign governments. This foreign tax credit is generally limited to the amount of (pre-credit) U.S. tax liability attributable to foreign-source income. This limit prevents the foreign tax credit from offsetting the U.S. tax on U.S.-source income.
As a result of taxing citizens and residents on a worldwide basis but allowing a foreign tax credit, some Federal income tax returns may report substantial income but little or no U.S. tax liability after credits. This may occur, for example, if a taxpayer has income only from foreign sources, e.g., the taxpayer may live abroad the entire year and have no income-producing assets in the United States. It may also occur if a taxpayer has foreign-source income that exceeds a net loss from U.S. sources and pays income taxes comparable to the U.S. tax to a foreign government [B2].
For taxpayers with income from foreign sources, these procedures understate the taxpayers' true worldwide income tax liabilities and effective income tax rates. For such taxpayers, it does not seem appropriate to classify U.S. income tax credits for foreign tax payments as reducing tax liabilities. This is particularly true for tax filers who appear to be nontaxable because they do not have any U.S. tax liability, but who have paid foreign income taxes. A more accurate measure of overall income tax burden, as well as the numbers of nontaxable returns, can be obtained by considering all income taxes--U.S. as well as foreign. Thus, a second tax concept, worldwide income tax, has been used in addition to the traditional U.S. income tax.
Two Tax Concepts
Two tax concepts are used in this article to classify tax returns as taxable (i.e., returns showing an income tax liability) or nontaxable (i.e., returns showing no income tax liability) and to measure the tax burdens on taxable returns: U.S. income tax and worldwide income tax. Worldwide income tax is defined for purposes of this article as U.S. income tax, plus the foreign tax credits reported on the U.S. income tax return, and foreign taxes paid on excluded foreign-earned income (obtained from Form 1116, Foreign Tax Credit). The amount of the foreign tax credits and foreign taxes paid on excluded foreign-earned income is used as a proxy for foreign tax liabilities [B3]. The relationship of U.S. income tax to tax items reported on individual income tax returns, and to worldwide income tax, is shown in Figure M.
Comparing Exclusions, Deductions, Tax Credits, and Special Tax Computations
To compare the importance of various exclusions, deductions, tax credits, and special tax computations (such as the alternative minimum tax on tax preferences), the different types of items must be placed on the same basis. One way of doing so is to calculate the size of the deduction that would reduce (or increase) income tax by the same amount as a tax credit or special computation. This amount is called the "deduction equivalent" of the tax credit or special computation.
The deduction equivalent of a tax credit or a special tax computation is the difference between the taxable income that, using the ordinary tax rate schedules, would yield the actual tax before the provision in question is considered and the actual tax after the provision. For example, the "deduction equivalent of all tax credits" is equal to the difference between "taxable income that would yield income tax before credits" and "taxable income that would yield income tax after credits."
Using this method of equating the value of deductions, exclusions, credits, and special tax computations, the order in which the various credits and special tax computations are calculated may affect the value of their deduction equivalents. Because the tax rate schedules are progressive, with successive increments to income taxed at successively higher tax rates, the deduction equivalent of the credit converted last to a deduction equivalent will be larger (for the same amount of a credit) than the item converted first, unless all relevant taxable income amounts are within a single tax-rate bracket.
The deduction equivalents of tax credits shown in Tables 9 and 10 were computed by assuming that deductions and exclusions reduce taxes before credits. As a result, the deduction equivalent of tax credits may be overstated.
Tax return data used for Statistics of Income have been tabulated as they were reported on tax returns filed with the Internal Revenue Service (IRS). Certain obvious arithmetic errors have been corrected and certain adjustments have been made to achieve consistent statistical definitions. Otherwise, the data have not been altered. In particular, the data do not reflect any changes that either may have or could have been made because of IRS audits. While this is true of data throughout the entire Statistics of Income program, it is particularly relevant for high-income tax returns. Because of the greater complexity of these returns, there is a higher probability of error and a broader scope for disagreement about the proper interpretation of tax laws.
The fact that the data have been drawn from unaudited returns is of even greater importance for those high-income returns that are nontaxable. Almost any audit changes would make such returns taxable. Even where the tax consequences are minor, such returns could be reclassified from nontaxable to taxable, thereby changing the counts of nontaxable returns.
Notes to Appendix B
[B1] An exception is that certain income earned abroad may be excluded from AGI. Any foreign taxes paid on such income are not creditable against U.S. income tax. The tables in this article include such excluded income in expanded income. Foreign taxes paid on such income are reflected in worldwide income tax, as discussed later.
[B2] Although the foreign tax credit is an item of tax preference for AMT purposes, taxpayers below the AMT exclusion thresholds, or with preferences or deductions not subject to AMT, could completely offset pre-credit U.S. income tax liability with foreign tax credits.
[B3] Where foreign tax rates exceed U.S. rates, foreign tax credits will be less than foreign tax liabilities. In such cases, using foreign tax credits as a proxy for foreign tax liabilities understates worldwide income tax liability. In other cases, when foreign tax credits are for taxes paid on income from previous years, use of foreign tax credits as a proxy may overstate or understate worldwide taxes on current-year income.
Figure L Derivation of Expanded Income From Adjusted Gross Income, Tax Years 1977-2013 Adjusted gross income (AGI) PLUS: * Excluded capital gains (tax years prior to 1987) * Tax-exempt interest (1987 and later tax years) * Nontaxable Social Security benefits (1987 and later tax years) * Tax preferences for alternative minimum tax purposes [A5] * Foreign-earned income exclusion (1990 and later tax years) MINUS: * Unreimbursed employee business expenses [A4] * Nondeductible rental losses (Tax Year 1987) * Moving expense deduction (Tax Years 1987 through 1993) [A4] * Investment interest expense deduction to the extent it does not exceed investment income * Miscellaneous itemized deductions not subject to the 2-percent-of-AGI floor (1989 and later tax years) EQUALS: * Expanded income NOTE: Footnotes to this figure are included with the footnotes to Appendix A: Income Concepts, which also includes an explanation of adjusted gross income and expanded income. SOURCE: IRS, Statistics of Income Division, Individual High-Income Tax Returns, May 2016. Figure M Derivation of "U.S. Income Tax" and "Worldwide Income Tax," Tax Year 2013 Tax at regular rates (tax generated) PLUS: Additional taxes (such as tax on accumulation distributions from qualified retirement plans, Form 4972) PLUS: Alternative minimum tax (Form 6251) EQUALS: Income tax before credits MINUS: Tax credits (nonrefundable) PLUS: Net investment tax MINUS: Refundable tax credits  EQUALS: U.S. income tax PLUS: Foreign tax credit PLUS: Foreign taxes paid on excluded foreign-earned income (Form 1116) EQUALS: Worldwide income tax  Includes the earned income credit, additional child tax credit, refundable American opportunity credit, regulated investment company credit, and the health insurance credit. NOTE: See Appendix B: Tax Concepts for further discussion of the information in this figure. SOURCE: IRS, Statistics of Income Division, Individual High-Income Tax Returns, May 2016.
Justin Bryan is an economist with the Individual Returns Analysis Section. This article was prepared under the direction of Michael Strudler, Chief, Individual Research Section.
(1) The statutory requirement is contained in section 2123 of the Tax Reform Act of 1976 (90 Stat. at 1915).
(2) The 1976 Act specified four income concepts for classifying tax returns: adjusted gross income (AGI), expanded income, AGI plus excluded tax preference items, and AGI less investment interest expense not in excess of investment income. Section 441 of the Deficit Reduction Act of 1984 (98 Stat. at 815) eliminated the requirement to use the last two income concepts.
(3) The definition of adjustments to AGI to obtain the expanded income given in the text is for the current year. See Appendix A for a discussion of AGI and expanded income and a list of adjustments covering all years since 1977.
(4) See Notes to Appendix A, note A4.
(5) Tax-exempt interest had to be reported on the individual income tax return starting with Tax Year 1987 and is included in expanded income starting with that year. Beginning with Tax Year 1991, tax-exempt interest was incorporated into the criteria used for sampling returns for Statistics of Income, thus increasing the reliability of the estimates of expanded income.
(6) See Appendix B for a discussion of the tax concepts. In data published for years prior to 1989, either in articles presented in the Statistics of Income Bulletin or in chapters in Statistics of Income--Individual Income Tax Returns (see Reference section), the "U.S. income tax" concept was described as "total income tax," and the "worldwide income tax" concept was described as "modified total income tax."
(7) The inclusion of foreign taxes paid on excluded foreign-earned income, beginning with Tax Year 1990, represents an improvement in the worldwide income tax concept. However, it also represents a slight break in the year-to-year comparability of data for worldwide income tax. Nevertheless, the number of returns with foreign taxes paid on excluded foreign-earned income is extremely small compared to the number of returns with the foreign tax credit.
(8) Inflation-adjusted dollars are based on the Consumer Price Index (CPI-U) as published by the U.S. Department of Labor, Bureau of Labor Statistics. The consumer price index approximates buying patterns of typical urban consumers. The annual index is the average of the monthly indices.
(9) See National Bureau of Economic Research list of business cycles at http://www.nber.org/cycles.html.
(10) Tax Year 2013 was the first year for which the NIIT was in effect. (See section 1411 of the Health Care and Education Reconciliation Act of 2010 [P.L. 111-152].)
(11) Note that before 1991, the number of nontaxable returns with expanded income of $200,000 or more was based on samples. Year-to-year differences in the number and percentages of nontaxable returns with expanded income of $200,000 or more may have represented sampling variability, in addition to actual changes in the number of such returns. Beginning with Tax Year 1991, nontaxable returns with expanded income of $200,000 or more were sampled at higher rates, which reduced the sampling variability of these returns and, therefore, provided improved estimates. Thus, the data for returns prior to 1991 are not entirely comparable with data for more recent years.
(12) The AMT exclusion phases out above certain levels of "alternative minimum taxable income," based on filing status. However, since taxpayers will have some AMT liability in the phase-out range, the phase-out income is not relevant for nontaxable, high-income returns.
(13) Tax-exempt interest and the foreign-earned income exclusion were not included in Tables 7 and 8 as possible tax effects before Tax Year 1994. Thus, caution should be exercised in making comparisons between data prior to 1994 and after 1993.
(14) See Appendix B for a description of how the deduction equivalent of credits was computed.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||p. 1-17|
|Publication:||Statistics of Income. SOI Bulletin|
|Date:||Jun 22, 2016|
|Previous Article:||Individual income tax returns, preliminary data, 2014.|
|Next Article:||Individual income tax shares, 2013.|