Tax legislation enacted in 2001, 2002, and 2003 is scheduled to phase in, phase out, and "sunset" entirely after 2010. As a result, federal tax rules will differ in every year from 2001 through 2011 and, consequently, so will effective tax rates--which are the total federal taxes that people bear measured as a percentage of their income. Because provisions have different impacts on people with different income and because those provisions change from year to year, effective tax rates fall and rise in patterns that vary over both time and income quintiles (or fifths of the distribution). This analysis of effective federal tax rates from 2002 through 2014 uses data on incomes in 2001, the most recent year for which information is available.

To test the effect of that choice of a starting point--that is, the sensitivity of effective tax rates to the income data on which they are based--the Congressional Budget Office (CBO) also analyzed effective tax rates beginning with the higher incomes of 2000. The choice of a beginning year causes only a small difference in the rates, much of which stems from the fact that realized capital gains were nearly twice as large a fraction of gross domestic product (GDP) in 2000 as in 2001--6.6 percent versus 3.5 percent. Realized gains in 2001 were much closer to their historical average, so tax rates based on incomes that year may more accurately represent the results of current tax law over the coming decade.⁽¹⁾

The analysis reflects the expected changes in tax burdens as measured by applying the tax law in effect in each year to the underlying incomes. It does not reflect any change in revenues that would result from changes in taxpayers' behavior. For example, the analysis would not capture any change in tax payments that could result if a lower tax rate on capital gains induced taxpayers to realize more capital gains. Because people who realized additional gains could end up paying more in taxes, including that response by taxpayers could incorrectly suggest that they were worse off as a result of the rate reduction.
 

Three Tax Laws

Lawmakers enacted three major tax bills between 2001 and 2003. The Economic Growth and Taxpayer Relief Reconciliation Act of 2001 (EGTRRA) lowered rates, increased credits, and offered relief from marriage penalties and from the alternative minimum tax (AMT). The Job Creation and Worker Assistance Act of 2002 (JCWAA) increased depreciation allowances for some property and altered certain provisions concerning operating losses. The Jobs and Growth Tax Relief Reconciliation Act of 2003 (JGTRRA) accelerated some of the provisions in EGTRRA and temporarily raised exemption levels for the AMT. None of the tax provisions in the three laws is permanent, and all of the provisions will expire by 2011. Furthermore, because many provisions phase in and phase out between 2001 and 2010, taxes change in every year through 2011.

The Economic Growth and Tax Relief Reconciliation Act of 2001

EGTRRA lowered individual income taxes for all taxpayers by restructuring tax rates and brackets, increasing the child credit and dependent care credit, providing relief from marriage penalties and the AMT, and increasing the earned income credit (EIC) for married couples (see Table 1). The law created a 10 percent tax bracket and lowered the rates for the top four brackets in four steps between 2001 and 2006. It raised the child credit from $500 to $1,000, also in four steps between 2001 and 2010. EGTRRA increased the maximum expenditure eligible for the dependent care credit from $2,400 to $3,000 per child, beginning in 2002, and raised the maximum credit from 30 percent to 35 percent of eligible expenditures. To ease marriage penalties, the law widened the 15 percent tax bracket for joint filers from 167 percent of the bracket for single filers to twice that bracket in four annual steps beginning in 2005 and increased the standard deduction for joint filers from 167 percent to 200 percent of the standard deduction for single filers in five annual steps between 2005 and 2009. Relief from the AMT resulted from increasing the income exemption by $4,000 for joint filers and $2,000 for single filers for 2001 through 2004. EGTRRA removed the limitation on itemized deductions and on personal exemptions in three steps between 2006 and 2010. Finally, the law increased the EIC for married couples. Specifically, in each of 2002, 2005, and 2008, the income range over which benefits phase out is shifted up by $1,000; in 2009 and 2010, the final value is indexed for inflation. All of those provisions sunset in 2011, and the provisions of individual income tax law return to those in effect before 2001.⁽²⁾

The Job Creation and Worker Assistance Act of 2002

JCWAA provided for a depreciation deduction of 30 percent of the adjusted basis for certain property for the tax year in which it was placed in service. That deduction is on top of any depreciation deduction for which that property already qualified. The additional deduction applies only to property purchased (or on which construction was begun) after September 10, 2001, and before September 11, 2004, and put into use before January 1, 2005. The law also raised from two years to five years the period over which taxpayers can claim net operating losses incurred in 2001 or 2002 to recalculate taxes owed for previous years.⁽³⁾

The Jobs and Growth Tax Relief Reconciliation Act of 2003

JGTRRA accelerated the pace at which provisions of EGTRRA phase in, reduced taxes on capital gains and qualified dividends, raised the AMT exemption, and increased first-year depreciation deductions (see the shaded portions of Table 1). The provisions of JGTRRA all sunset by 2009. Specifically, the law broadened the 10 percent tax bracket for 2003 and 2004 and, starting in 2003, lowered the tax rates for the top four brackets to the levels that EGTRRA had set to begin in 2006. It also raised the child credit to $1,000 for 2003 and 2004 (when it would have been $600 under EGTRRA). JGTRRA widened the 15 percent tax bracket for joint filers in 2003 and 2004 to twice that for single filers and set the standard deduction for joint filers equal to double that for single filers for those years. The legislation raised the AMT exemption for 2003 and 2004 to $58,000 for joint filers and $40,250 for single filers, higher than the levels set in EGTRRA by $9,000 and $4,500, respectively. For 2003 through 2008, the law lowered the tax rate for capital gains and for qualified dividends from 20 percent to 15 percent for taxpayers above the 15 percent bracket and from 10 percent to 5 percent (and to zero in 2008) for taxpayers in lower brackets. Finally, JGTRRA increased the first-year depreciation created by JCWAA to 50 percent for property acquired between May 5, 2003, and the end of 2004 and raised to $100,000 the maximum deduction under section 179 of the Internal Revenue Code for property placed into service over the 2003-2005 period.
 

Measuring Effective Tax Rates

This analysis uses the same methodology that CBO employed for its earlier reports on effective tax rates.⁽⁴⁾ CBO combines data from the Internal Revenue Service's Statistics of Income and the Census Bureau's Current Population Survey, classifies and ranks households into percentiles based on adjusted comprehensive household income, and simulates individual income tax liabilities. The income measure used includes pretax cash income plus income from other sources. Pretax cash income includes wages, salaries, self-employment income, rents, taxable and nontaxable interest, dividends, realized capital gains, cash transfer payments, and retirement benefits; taxes paid by businesses (corporate income taxes and employers' shares of Social Security, Medicare, and federal unemployment insurance payroll taxes); and employees' contributions to 401(k) retirement plans. The comprehensive income measure also includes in-kind benefits (Medicare, Medicaid, employer-paid health insurance premiums, food stamps, school lunches and breakfasts, housing assistance, and energy assistance).

Defining Income Categories

Income categories are defined by ranking all people by their comprehensive household income adjusted for the size of the household--that is, divided by the square root of the household's size. A household consists of the people who share a housing unit, regardless of their relationships. Quintiles--or fifths of the distribution--are created by dividing the entire population into five parts, each containing the same number of people. Because households vary in size, quintiles generally contain unequal numbers of households. Effective tax rates differ markedly among households in the top quintile, so that group is further divided into the top 10 percent, top 5 percent, and top 1 percent for some analyses.⁽⁵⁾ Households with negative income--from business losses, for example--are excluded from the lowest income category but included in the totals (in the tables in this paper). Those households generally differ from other households in the bottom quintile because they typically have significant assets; including their negative income and tax payments would have significantly affected the dollar-weighted results for the bottom quintile.

Calculating Effective Tax Rates

Effective tax rates equal the taxes paid by or imputed to households divided by their pretax income.⁽⁶⁾ For a given segment of the income distribution, the analysis calculates a dollar-weighted average rate as the sum of taxes falling on households in that segment divided by the sum of the pretax incomes of those households. Effective tax rates for individual households may differ substantially from the group average, particularly given the fact that the adjustment for the size of households results in various kinds of households with substantially different unadjusted incomes falling in the same income category. As a result, effective tax rates represent the situation not for particular taxpayers but that for groups of taxpayers with similar amounts of adjusted household income.

CBO estimated the effective rates for four federal taxes--individual and corporate income taxes, payroll taxes, and excise taxes--under current law for each year from 2001 through 2014. The estimates for 2001 match CBO's historical values.⁽⁷⁾

The analysis assigns taxes to households on the basis of who bears the burden of the taxes. In particular, it assumes the following:

• Households bear the burden of all taxes that they pay directly, specifically, individual income taxes and the employee's share of payroll taxes.



• Households pay excise taxes according to their consumption of taxed goods, such as tobacco and alcohol. In the case of excise taxes on intermediate goods, such as components of consumer goods, households bear the taxes in proportion to their overall consumption.



• The burden of taxes levied on businesses actually falls on households. In line with most economists, CBO assumes that the employer's share of payroll taxes falls on employees and thus assigns those payments to employees both as income and taxes paid. The analysis assumes that corporate income taxes fall on the owners of capital and allocates those liabilities--again, both as income and as taxes--to households in proportion to their income from interest, dividends, rents, and capital gains.⁽⁸⁾

Because of uncertainty about the incidence of other taxes and some data limitations, the analysis excludes estate and gift taxes, tariffs, and other miscellaneous sources of revenue. It also excludes some of the lesser provisions of EGTRRA, such as those providing education incentives and tax benefits for retirement saving, because data are not available to estimate their effects among the various income quintiles.

Incorporating Assumptions About the Growth of Incomes

The analysis, based on reported 2001 incomes, assumes that those incomes grow at a constant rate of 4.5 percent per year and that inflation is steady at a 2.2 percent annual rate--assumptions that are consistent with the average growth of GDP and the consumer price index reported in CBO's January 2004 Budget and Economic Outlook. ⁽⁹⁾ The analysis assumes that income growth is the same for all sources of income and for households throughout the income distribution. Thus, the share of income going to each quintile does not change over the period examined.⁽¹⁰⁾
 

Estimating Changes in Tax Law

CBO generally estimated individual income taxes by simulating the applicable provisions of law, adjusted for income growth, on tax returns filed in 2001. The analysis thus does not account for incomes changing in response to the tax cuts. CBO assumed that the reduction in the tax rate on dividends from corporate stock would accrue to the owners of capital in proportion to their income from interest, dividends, rents, and capital gains.⁽¹¹⁾

As described, JCWAA and JGTRRA both include more generous depreciation allowances for businesses, and JCWAA changed the rules for use of net operating losses. CBO assumed that those provisions would cause a drop in revenues equal to the amount estimated by the Joint Committee on Taxation at the time of their enactment.⁽¹²⁾ CBO adjusted that change in revenues to align with the underlying 2001 level of income, using the expected growth in income in the absence of those legislative changes. CBO then divided the resulting estimated tax benefits between corporate income taxes and individual income taxes--allocating the corporate portion to taxpayers in proportion to their income from capital (as done with corporate income taxes) and allocating the individual income tax portion among taxpayers in proportion to their shares of total income taxes paid on noncorporate business income.

The effects of the changes to partial expensing occur across years. The provisions shift the deductibility of depreciation of business assets in time: businesses can claim more depreciation during the 2002-2005 period and consequently cannot claim as much depreciation in later years. That shifting decreases taxes in the near term but increases them in subsequent years. The provision extending the period for claiming net operating losses also shifts some deductions to earlier years, offsetting effects in later years. Because tax changes for a given year reflect only that year's impact and ignore any offsetting effects that occur in other years, no single year conveys completely the full effects of the provisions.⁽¹³⁾
 

Effective Tax Rates in Future Years Under Current Tax Law

Under current law--and the assumption that incomes grow at a constant rate between 2001 and 2014--the total effective federal tax rate for all taxpayers drops from 21.5 percent in 2001 to 19.6 percent in 2004 before reversing course and climbing over the next decade (see the top panel in Table 2). From the 2004 low of 19.6 percent, the overall effective tax rate jumps to 21.4 percent in 2005 as most features of JGTRRA and JCWAA disappear--decreasing the child credit, lessening the relief from marriage penalties, and raising the AMT. The effective rate climbs slowly over the next five years, to 22.1 percent in 2010, primarily because the unindexed AMT affects more and more people and the growth of real incomes pushes taxpayers into higher tax brackets. The effective tax rate takes another jump to 23.6 percent in 2011 after EGTRRA sunsets and thereafter resumes its slow climb driven by continued real income growth and the widening reach of the AMT (by 2014, nearly 22 million taxpayers will be subject to the alternative tax). Overall, then, under current law, the effective federal tax rate will increase from 21.5 percent in 2001 to 24.1 percent in 2014. Because tax legislation enacted since 2001 focuses primarily on the individual income tax, the pattern of changes for the total effective tax rate over the 2001-2014 period derives almost entirely from changes in the effective individual income tax rate (see the second panel in Table 2).

The pattern of changes in the effective federal tax rate for all taxpayers is repeated for each income quintile and the top income percentiles, albeit with slightly different turning points and different degrees of change between 2001 and 2014 (see Table 2). For the bottom four quintiles, the effective individual income tax rate turns upward in 2004, a year ahead of the rise for the top quintile. All quintiles experience a jump in their overall effective tax rate in 2011 following the expiration of EGTRRA. They also all have a higher effective rate in 2014 than in 2001. For example, the effective rate for the lowest quintile increases from 5.4 percent in 2001 to 8.3 percent in 2014; in contrast, the rate for the top quintile climbs from 26.8 percent to 28.8 percent and that for the top 1 percent of taxpayers rises from 33.0 percent to 33.6 percent over the same period.

The differential increase in effective tax rates among quintiles is reflected in a shift down the income distribution in shares of taxes paid (see the third and fourth panels of Table 2). The share of taxes paid by the top quintile falls from 65.3 percent in 2001 to 62.8 percent in 2014, even though that group's share of income does not change. Four-fifths of that decline occurs for the top 1 percent of taxpayers, whose share falls by 2 percentage points, to 20.7 percent of federal taxes in 2014. The share of taxes paid by each of the middle three quintiles climbs by about 0.7 percentage points.

Changes in tax law explain much of the expected changes in effective tax rates, but as previously noted, effective tax rates would change over time even in the absence of changes in the law. Real income growth increases effective tax rates over time because the individual income tax is indexed not for the rise in real income but, rather, only for inflation or not at all. Separating the effects of changes in the law and income growth requires calculating the changes in rates that would result from income growth alone. This analysis accomplishes that by calculating effective tax rates for each year under the provisions of 2000 tax law as a point of comparison. For example, under that scenario, the overall effective tax rate would have been expected to rise by 1.2 percentage points, from 22.2 percent in 2001 to 23.4 percent in 2010 (see Table 3). In contrast, using the tax provisions that will apply in each year under current law, which yields the combined effect of changing law and income, the effective tax rate increases by 0.6 percentage points, from 21.5 percent to 22.1 percent, over the same period (see Table 3). Thus, changes in tax law reduce the expected increase in the overall effective rate between 2001 and 2010 by about half.

Relative to the situation in 2000, the three major tax laws enacted between 2001 and 2003--EGTRRA, JCWAA, and JGTRRA--reduce effective federal tax rates for each quintile in every year from 2001 through 2010. For all but the lowest quintile, the reduction is greatest in 2004, when all three tax laws are in effect. Effective tax rates in 2004 are lower than those under 2000 law by 1.5 percentage points for the lowest quintile, 3.9 percentage points for the highest quintile, and 6.8 percentage points for the top 1 percent of taxpayers (see Table 4). With the expiration of most provisions of JCWAA and JGTRRA in 2005, the reduction lessens but then is partially restored in 2008 for the top quintile and in 2010 for the bottom four quintiles. With the sunset of EGTRRA in 2010, federal tax law reverts in most respects to that in effect in 2000, and by 2014, effective tax rates for all income quintiles return to the levels obtained under 2000 law.

The AMT grows in importance over the next decade, but its impact differs among income quintiles and under different tax law.⁽¹⁴⁾ The AMT raises effective rates above what they would be in its absence, with the largest effects in the fourth and fifth quintiles. Over time, the AMT reaches down the income distribution, slightly increasing effective rates as far down as the second quintile by 2010. The AMT interacts with regular income taxes, having a greater impact when they are lower. The differences between tax rates under current law and tax rates under 2000 law are therefore smaller than they would be without the AMT: for all taxpayers, the AMT lowers those differences by 0.2 percentage points in 2005 and 0.6 percentage points in 2010 (see Table 5). The effect is greatest in the highest quintile, where the AMT reduces the difference between the effective tax rate under current tax law and that under 2000 law by 1 percentage point in 2010. The impact on the fourth quintile is roughly half as large, and the bottom three quintiles are virtually unaffected. Within the top quintile, the AMT affects the top percentile of taxpayers less than other households because those taxpayers face high regular tax rates that exceed AMT rates.

Table 2.

Table 3.

Table 4.

Table 5.