A Tale of Three Taxpayers
A Tale of Three Taxpayers
Cecil E. Bohanon
MERCATUS 
RESEARCH
Bridging the gap between academic ideas and real-world problems
Copyright  2014 by Cecil E. Bohanon 
and the Mercatus Center at George Mason University
Mercatus Center at George Mason University
3434 Washington Boulevard, 4th Floor
Arlington, VA 22201
703-993-4930
mercatus.org
Release date: July 1, 2014
ABOUT THE MERCATUS CENTER AT GEORGE MASON UNIVERSITY
The  Mercatus  Center  at  George  Mason  University  is  the  worlds  premier 
 university source for market-oriented ideasbridging the gap between academic 
ideas and real-world problems.
A university-based research center, Mercatus advances knowledge about how 
markets work to improve peoples lives by training graduate students, conduct-
ing research, and applying economics to offer solutions to societys most pressing 
 problems.
Our mission is to generate knowledge and understanding of the institutions that 
affect the freedom to prosper and to find sustainable solutions that overcome the 
barriers preventing individuals from living free, prosperous, and peaceful lives.
Founded in 1980, the Mercatus Center is located on George Mason Universitys 
Arlington campus.
www.mercatus.org
ABOUT THE AUTHOR
Cecil E. Bohanon is a professor of economics at Ball State University. He obtained 
his BA from Wilmington College (Ohio) and his PhD from Virginia Tech, where 
he studied under Nobel Prizewinning economist James Buchanan. Bohanon has 
published more than 30 refereed professional articles, notes, and comments and 
more than 100 popular articles, policy monographs, and newspaper editorials. His 
research interests include public choice, applied microeconomics, and economic 
education. He has won numerous teaching awards, and he writes regularly for the 
Indiana Policy Review and the Indianapolis Business Journal.
ACKNOWLEDGMENTS
The author wishes to thank his undergraduate research assistant, Colin Philips, 
for assistance and insight. All errors remain the authors.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
4
ABSTRACT
How  have  federal  personal  income  tax  obligations  evolved  over  the  past  60 
years? A common perception is that the federal income tax burden on the poor has 
increased while the tax burden on the rich has declined. This study focuses on three 
archetypical households. The first consists of two children and a single mother earn-
ing 125 percent of the current poverty line. The second consists of two children and 
a married couple whose income is close to the current median household income. 
The third consists of two children and a married couple whose income is just below 
that of the top 1 percent of current earners. All three households maintain constant 
inflation-adjusted income over time. The changes in relative and absolute federal 
tax liabilities measured in 1953, 1973, 1993, and 2013 are informative and somewhat 
surprisingand cannot be summarized in any simple slogan. In absolute dollars the 
rich and poor seem to have fared best, while in relative terms the poor have benefit-
ted disproportionately.
JEL codes: A1, H2, I3
Keywords: federal income tax, federal tax policy, federal tax burden, federal tax 
cuts, earned income tax credit, child tax credit, progressive tax system, marginal 
tax rates, tax burden on poor, tax burden on rich
5
W
hich of the following claims do you believe? Changes in the federal 
income tax over the past 60 years have benefited the rich at the expense 
of the poor and the middle class. Changes in the federal income tax over 
the past 60 years have benefited the poor at the expense of the rich and the middle 
class. Changes in the federal income tax over the past 60 years have benefited the 
rich and the poor at the expense of the middle class. The abovementioned claims 
contradict each other, yet each is likely believed by significant portions of the popu-
lation. So which is true? This essay proposes a simple method to sort this out and 
concludes that, in a way, they all are.
The nonpartisan Congressional Budget Office projects that the federal individual 
income tax will collect about $8.3 trillion in tax revenue or 8.8 percent of GDP over 
the fiscal years 20142018. The individual income tax is the largest revenue source 
for the federal government.
1
 The federal income tax finances federal expenditures. 
It also directs resources to particular uses through so-called tax expenditures in 
the code. Tax deductibility of charitable contributions, for example, is designed to 
increase giving to charitable institutions by taxpayers. Moreover, the progressive 
nature of the federal income tax impacts posttax income distribution. If a household 
with an annual income of $1 million pays a higher percentage of its income to the 
federal government than the $10,000-a-year household pays, the posttax income 
ratio between the two households falls. Whether by design or political happen-
stance, it is clear the federal income tax code has impacts beyond its immediate 
fiscal implication for federal spending.
Over the past six decades, the marginal tax rates assessed on top income earners 
have declined dramatically. As recently as 1963, the highest rate was 91 percent; 
since 1987, the top rate has varied between 28 percent and the current 39.6 percent. 
It should also be noted that these reductions in top marginal tax rates were accom-
panied by significant tax reductions in rates assessed on lower income brackets. In 
1. See Individual Income Tax Receipts and the Individual Tax BaseFebruary 2013 Baseline, 
Congressional Budget Office, 2013, http://www.cbo.gov/publication/43901. The weighted average for 
the entire five-year time frame is 8.8 percent. The Congressional Budget Offices projection is based on 
an assumption of economic growth, which tends to elevate the revenue percentage over time.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
6
1953, for example, the marginal tax rate associated with the lowest tax bracket was 
22.2 percent, whereas in 2013 it stood at 10 percent. An examination of the Federal 
Income Tax Rates History available at Tax Foundations website shows that since 
1950 tax rates have generally been cut across all brackets and that brackets have 
been consolidated in the major tax reforms of the 60s, the 80s, and the first decade 
of the 21st century.
2
 Table 1 shows how the top rate, bottom rate, and number of 
brackets for joint returns have evolved over 20-year intervals since 1953. These 
developments, along with indexation of tax brackets to inflation in the early 1980s, 
imply that the comparative burden of taxation cannot be encapsulated in a single 
statistic such as the highest marginal tax rate. This study focuses exclusively on 
income taxes for a consistent examination of how the burden of this tax has changed 
over time. Payroll and excise taxes disproportionately affect lower- and middle-
income families, but these taxes are not factored into the following analysis.
TABLE 1. TOP MARGINAL TAX RATES, LOWEST MARGINAL TAX RATE, REAL AND INFLATION-
ADJUSTED INCOME THRESHOLDS FOR BRACKETS, AND NUMBER OF BRACKETS1953, 
1973, 1993, AND 2013
Year Top marginal rate Lowest marginal rate Number of brackets
1953
Rate 92% 22.2%
26 Income threshold in nominal dollars $300,000 $2,000
Income threshold in 2013 dollars $2,610,997 $17,406
1973
Rate 70% 14%
25 Income threshold in nominal dollars $200,000 $1,000
Income threshold in 2013 dollars $1,046,751 $5,265
1993
Rate 39.6% 15%
5 Income threshold in nominal dollars $250,000 $36,900
Income threshold in 2013 dollars $402,039 $59,341
2013
Rate 39.6% 10%
7
Income threshold in nominal 
dollars and 2013 dollars
$450,000 $17,850
Source: Tax Foundation, Bureau of Labor Statistics, consumer price index  calculator; and authors calculations.
2. See Federal Income Tax Rates History, Tax Foundation, 2014, http://taxfoundation.org/article/us 
-federal-individual-income-tax-rates-history-1913-2013-nominal-and-inflation-adjusted-brackets. 
Adjustment in 2013 dollars based on authors calculations from  the Bureau of Labor Statistics consumer 
price index website at http://data.bls.gov/cgi-bin/cpicalc.pl?cost1=1000&year1=2011&year2=2013.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
7
Much of the popular rhetoric about taxation implies that there has been a sys-
tematic shifting of the tax burden from the wealthy to the middle and lower classes 
over the past 60 years. The wealthy in this scenario have obtained the lions share 
of benefits from tax cuts. Cries to increase taxes on the top 1 percent are a recent 
reflection of this view. A 2011 article captures the flavor of this populist strain: 
the effective tax burdens on the wealthy are lower now than at almost any time 
in the past fifty years, thanks to past rate cuts and a proliferation of tax exemptions 
which . . . shower most of their benefits on the affluent.
3
There are many ways to parse tax data and changes in tax codes. The IRS reports 
annual summary data of the number of tax returns by various income categories. 
It also reports the total dollar amount of income and federal income taxes paid in 
each category. For example, it is a straightforward exercise to calculate the average 
income of taxpayers earning between $50,000 and $70,000 and the average federal 
tax liability of the households. One way of considering the issue of how comparative 
federal income tax liabilities have evolved over the past 60 years is to delve in detail 
into that data.
4
This essay takes a different approach. It will not offer a host of summary charts 
on tax burdens, nor will it provide a nuanced calculation of progressivity mea-
sures. Rather, it will examine the taxes paid by three hypothetical but archetypical 
American households. Each households income, deduction status, and tax liabil-
ity are estimated for 2013. Each households pretax earnings and tax deductions 
are then held constant in inflation-adjusted terms for 1953, 1973, and 1993. What 
impact did the federal income tax have on their command over goods and services in 
each year? How has the evolution of the tax code affected each households dispos-
able income? The first archetypical household consists of two children and a single 
mother earning just enough to keep the family above the poverty line; the second 
consists of two children and a married couple earning close to the median household 
income;
5
 the last consists of two children and a married couple earning just below 
the threshold of the top 1 percent of all earners.
6
3. See Paul Glastris, Playing Chicken with History, Washington Monthly, July/August 2011, http://
www.washingtonmonthly.com/magazine/julyaugust_2011/editors_note/playing_chicken_with_history 
030506.php.
4. For example, the IRS routinely calculates what percentage of federal income tax revenues are col-
lected from the top 1 percent, 5 percent, 10 percent, and 50 percent of all households. These data can be 
found at http://www.irs.gov/file_source/pub/irs-soi/09in03etr.xls. Other approaches may calculate pre-
tax and posttax Gini coefficients for income and compare how they have changed over time.
5. Note that because real income has risen over the six-decade time frame, the income calculated for the 
households in earlier decades tends to upgrade their relative income status in the income distribution. 
When the median-income household of 2013s income is adjusted into 1953 dollars using the standard 
consumer price index, its income is well above the median household income of 1953.
6. Why these three archetypes? Three households allow for tractable comparisons. Much media atten-
tion is placed on middle-income households, lower-income single-parent households, and aspiring well-
off households. Of course, other households could be considered. However, the three chosen households 
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
8
DIFFERENT FEDERAL TAX STORY1953, 1973, 1993, AND 2013
Over the period from 1946 to 2018, tax data and projections indicate that the fed-
eral individual income tax raised an amount of revenue on average equal to 8 per-
cent of GDP. The percentage has ranged from a low of 5.7 percent in 1949 to a high 
of 10.2 percent in 2000. The percentage for three-quarters of the years falls between 
7.2 percent and 8.8 percent.
7
Splitting the time frame between periods is revealing. For the two decades of 
the 50s and 60s, the income tax as a percentage of GDP averaged 7.6 percent. It 
subsequently increased to 8.2 percent in both 20-year periods of the 70s and 80s 
and the 90s and the decade following 2000. The absence of variation from decade 
to decade is striking, especially given the general trend toward lower marginal tax 
rates for all classes.
The year-in-year-out variation, however, is as much a by-product of general eco-
nomic conditions as the specific marginal tax rates and brackets. After major tax 
cuts were enacted, tax revenues as a percentage of GDP tended to decline. However, 
during periods of robust economic growth, household incomes tended to increase 
and those rising incomes became subject to higher marginal tax rates. This increases 
the average tax of the tax code. Just the opposite occurs when the economy declines. 
It is as if policy and economic events gravitate toward the federal income tax rev-
enue equal to one-twelfth of GDP. So how have the typical earners fared over time?
LOW-INCOME SINGLE MOTHER WITH TWO CHILDREN
The first household consists of a single mother and her two children. The moth-
ers income is 125 percent of the 2013 poverty line.
8
 Her children are eligible for a 
break on school lunches and a number of other benefits not part of this calculation.
9
 
are of widespread interest. Also, absent from this discussion are payroll or social security taxes. However, 
many of the qualitative results of the analysis are not appreciably affected by their inclusion. But at a 
more fundamental level, the increasing importance of social security taxes and their evolution over time 
is a story that deserves treatment in its own right. Suffice it to say that as both the social security tax rate 
and the amount of income subject to social security taxes have risen over time, the tax has become more 
important and less regressive.
7. The data used in this paragraph are available in table format to accompany this paper; see Receipts by 
Source as Percentages of Gross Domestic Product, 19342017 at http://mercatus.org/publication/tale 
-three-taxpayers. Data on Federal Income Tax Revenue and GDP are from Tax Facts: Historical Source 
of Revenue as Share of GDP, Urban Institute and Brookings Institution, 2014, http://www.taxpolicy 
center.org/taxfacts/displayafact.cfm?Docid=205. Data are supplemented by the authors calculations.
8. Poverty-line data are taken from the US Department of Health and Human Services 2013 Poverty 
Guidelines at http://aspe.hhs.gov/poverty/13poverty.cfm.
9. The low-income household is eligible for a number of federal, state, and local benefit programs such as 
food stamps and access to low-cost housing that are not included in this calculation. For a detailed analy-
sis of these benefits, see Michael Tanner and Charles Hughes, The Work versus Welfare Trade-Off: An 
Analysis of the Total Welfare Benefits by State (Washington, DC: Cato Institute, 2013), http://www.cato 
.org/publications/white-paper/work-versus-welfare-trade.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
9
Her 2013 gross adjusted income is $24,412. Using the federal consumer price index 
from June of each of the reference years makes her equivalent nominal income in 
the previous years
1953 $2,718
1973 $5,604
1993 $15,139
2013 $24,412
In each year, she files as a single earner. She receives three exemptions, the standard 
deduction, and all available tax credits.
10
Her final tax liability in nominal terms and as a percentage of her earned income 
is as follows:
1953 $144 3%
1973 $299 5.3%
1993 ($444) 2.9%
2013 ($4,594) 18.8%
A low-income single parents federal tax liability has never been large. Part of the 
design of the federal tax system has always been to assess a relatively low tax liability 
on low-income earners. This liability was essentially unchanged between 1953 and 
1973 at just above 5 percent of this hypothetical mothers earnings. What is striking 
is that the low-income single mother became a beneficiary of the income tax in 1993, 
when she not only paid no federal income tax but also got refundable tax credits of 
$444, about 3 percent of her income. This benefit expanded over the next 20 years 
as her refundable tax credit rose to over $4,500, almost 19 percent of her income. If 
the low-income single mother were paying in 2013 what she was paying in 1953 as 
a percentage of her income, her tax liability would be $1,294, an amount that would 
reduce her disposable income by $5,888.
The claim to negative taxes is primarily a result of two tax policies: the child tax 
credit (CTC) and the earned income tax credit (EITC); the single-parent household 
qualifies for $2,000 in child tax credits, which are also refundable. These credits 
began in 1997 under the Clinton administration and were expanded after 2000 
under both the Bush and the Obama administrations.
Originally established in 1975 under the Ford administration, the EITC has been 
expanded by various administrations with bipartisan support. The EITC currently 
supplements low-income household earnings by matching every dollar earned with 
a 40-cent refundable tax credit up to $13,430 for a single parent with two children. 
10. The sources for these and all subsequent calculations are listed in the data appendix.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
10
No additional credits are awarded for additional income up to $17,350. For every 
dollar earned beyond $17,530, the earner loses 21.06 cents in tax credits.
This makes the EITC, like most government benefits designed to help the work-
ing poor, a means-tested program. It implies that if low-income earners take risks 
or bear costs in attempting to increase their income, success will be punished by 
the loss of government benefits. The EITC alone imposes a 21.06 percent marginal 
tax rate on earnings above $17,530.
11
MIDDLE-INCOME COUPLE WITH TWO CHILDREN
The second hypothetical household, consisting of a married couple and two 
children, has close to the median household income for 2013 with a gross adjusted 
income of $50,000. Using the federal consumer price index from June of each of the 
reference years, the equivalent income in the previous years is listed below. Note 
that these figures are not the median household income of the years listed. I calcu-
late that the median income in 1953 was only $3,700, reflecting the fact that median 
incomes have risen in real terms over the past 60 years:
12
1953 $5,567
.12
1973 $11,477
1993 $31,006
2013 $50,000
In each year, the couple files a joint return. They receive four exemptions and take 
the standard deduction and all available tax credits.
The final tax liability in nominal terms and as a percentage of their earned income 
are as follows:
1953 $579.10 4%
1973 $1,224 10.7%
1993 $2,686 8.7%
2013 $1,352 2.7%
The middle-income couples federal tax liability rose slightly between 1953 and 
1973, from 10.4 percent of their earnings to 10.7 percent. The tax liability declined 
to 8.7 percent of income by 1993 and shrank to 2.7 percent by 2013. If the couple 
11. Taxation and the Family: What Is the Earned Income Tax Credit?, Tax Policy Institute (Urban 
Institute and Brookings Institution joint venture), last modified February 12, 2014, http://www.taxpolicy 
center.org/briefing-book/key-elements/family/eitc.cfm.
12. Taxpayers who earned the medium income in each year would see less of a decline in their tax liabil-
ity as a percentage of their income because their 1953 tax liability would have been less.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
11
were paying in 2013 what they paid in 1953 as a percentage of their income, the 
tax liability would be $5,200, reducing disposable income by $3,848. The 2013 
federal tax liability is 25 percent of what it would be had the 1953 code never been 
modified.
The tax relief received is a result of not only the general decline in marginal tax 
rates but also the child tax credit that saves the household $2,000. Interestingly, the 
households income is just above the 2013 threshold of $48,378 for qualifying for the 
earned income tax credit.
UPPER-INCOME COUPLE WITH TWO CHILDREN
The third hypothetical household also consists of two children and a married 
couple, but their earnings put this household close to the top 1 percent of all house-
holds filing returns. The couples 2013 gross adjusted income is $350,000.
13
 Using 
the federal consumer price index from June of each of the reference years shows 
the equivalent income in the previous years as listed below:
1953 $38,972
1973 $80,341
1993 $217,041
2013 $350,000
The couple files a joint return and receives four exemptions. They also itemize 
deductions, which are 16.7 percent of earnings. This figure is based on 2010 data on 
average deductions for earners in the earnings range. The final tax liability in nomi-
nal terms and as a percentage of the couples earned income is as follows:
14
1953 $10,611 27.2%
14
1973 $16,269 20.2%
1993 $47,255 21.8%
2013 $67,412 19.3%
Between 1953 and 1973, the upper-income households federal tax liability fell 
from 27.2 percent of the couples earnings to 20.2 percent. The tax liability actu-
ally increased to 21.8 percent of their income by 1993 but declined to a low of 19.3 
percent by 2013. If they paid in 2013 what they were paying in 1953 as a percentage 
13. It is assumed that the household income is ordinary labor income and not capital gains or other forms 
of income subject to preferential treatment.
14. The 1953 income of $10,621 would have placed the taxpayer in the top one-fourth of 1 percent of earn-
ers in that year. A household that earned just enough to qualify for the top 1 percent would have seen less 
of a decline in its tax liability as a percentage of income over the past 60 years because its 1953 tax liabil-
ity would have been less.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
12
of income, their tax liability would be $95,200, reducing their disposable income by 
$27,788. Their federal tax liability is 29 percent less in 2013 than it would be had the 
1953 code never been modified.
Note that the upper-income household receives neither the earned income tax 
credit nor the child tax credit. The decline in its tax liability is exclusively attribut-
able to reductions in marginal tax rates. Interestingly, most of the gain came from 
the lowering of marginal tax rates and rebracketing that occurred before the Reagan 
tax cuts.
WHO GOT THE BIGGEST TAX BREAK?
The tax arithmetic presented indicates that all three taxpayers pay less in taxes 
in 2013 than they would have in previous years. If the tax code had been unchanged 
since 1953, the disposable posttax income of all three would be thousands of dollars 
less. But who got the largest tax cut? Here is where the parsing and spinning of data 
often happens. Imagine the following three headlines:
Rich households tax breaks greater than income of earner near poverty line
Tax cuts give low-income worker benefits three times those of  
middle and high earners
Middle-income workers tax breaks small compared to those  
of rich and poor
The first headline is consistent with a view that the 60 years of revising the tax 
code has helped the rich. The second indicates that the poor have received most of 
the benefit of tax cuts. The third indicates that the middle class has been unfairly 
shut out of the tax-cutting bounty. Although the headlines seem contradictory, all 
three are supported by the data.
The upper-income households 2013 tax savings compared to the 1953 code are 
$27,788, which is more than the single-mothers earnings of $24,412the first head-
line is statistically factual.
The lower-income households disposable income in 2013 is 24 percent more 
than it would be had the 1953 code been in place. Comparable percentages for the 
middle-income household and upper-income household are 7.7 percent and 7.9 per-
cent, respectivelythe second headline is statistically factual.
15
15. The low-income, single-parent households $5,888 in tax savings as a percentage of $24,412 in earn-
ings are 24 percent, the middle-income households $3,848 in tax savings as a percentage of $50,000 
in earnings are 7.7 percent; the upper-income households $27,788 in tax savings as a percentage of 
$350,000 in earnings are 7.9 percent.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
13
The middle-income households tax savings in dollar terms are only $3,848, while 
those of the upper-income household are $27,788 and those of the lower-income 
household, which earns less than half of what the middle-income household earns, 
are $5,888. Moreover, the middle-income households tax savings as a percentage of 
income are the lowest of the threethe third headline is statistically factual.
HOW DOES EVERYONE GET A TAX BREAK?
If all three of the taxpayers are paying less today than in the past, then how is it 
possible for federal income tax revenue to be a relatively constant proportion of 
GDP? The answer lies in per capita GDP growth. Measured in 2009 constant dollars, 
per capita GDP was $16,037 in 1953. By the same measure, it was $49,283 in 2012. 
This is more than a threefold increase. The average annual percentage increase in 
real per capita GDP during that time was just over 1.9 percent. Although per capita 
GDP and household income are not identical measures, they clearly correlate. In 
essence, income has risen, driving more earners into higher tax brackets.
Put another way, the exercise presented here implies that all three households 
had stagnant real income. Yet, on average, real income rose for most households. 
There are, in effect, more upper-income households today than in 1953meaning 
that even as such households and, indeed, all households surrender a lower percent-
age of their income to the federal government, what the federal income tax collects 
remains a relatively constant portion of GDP.
It is interesting to speculate whether the economic growth that spawned the 
increase in earnings, and especially earnings in the upper-income household, would 
have occurred had marginal tax rates remained at their 1953 level. In 1953, the rich 
household faced a marginal tax rate of 67 percent. In 2013, that rate was 33 percent. 
It is certainly plausible that this halving of the marginal tax rate at the top end of 
the brackets facilitated at least in part the economic growth that allowed for cuts 
in marginal tax rates at lower brackets and provisions that are designed to benefit 
lower- and middle-income workers.
The analysis suggests a kind of dynamic inevitability of income tax cuts in a grow-
ing economy. As the economy expands, taxpayer income rises, and under a progres-
sive tax system, more households are subject to higher marginal tax rates. This tends 
to increase the federal income tax take as a percentage of GDP. This likely increases 
both the political demand and the fiscal ability to offer tax cutsand explains their 
bipartisan appeal.
CONCLUSION
So what is the conclusion? The federal income tax imposes less of a burden on all 
three archetypical households today than it did 60 years ago. Which household got 
the biggest cut? The answer depends on what measure is used. Because the rich 
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
14
household earned more and pays more in taxes, it is hardly surprising that its tax 
cut is large compared to those of less affluent households when measured in dollars. 
Yet as a percentage of earned income, the biggest gainers from tax changes have 
been the lower-income households. Middle-income households have also obtained 
significant tax relief.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
15
APPENDIX: TAX-INCOME DATA SOURCES AND USES
The selection of the three households income seems arbitraryand it is. The 
2013 poverty line provides a metric for the working-mother, low-income household. 
Setting this single parents income at 125 percent of the poverty line has no particu-
lar justification except that it puts her slightly above the poverty line.
The middle-income, two-parent family is designed to be the median-income 
household. There are understandably no data on household median income for 
2013 yet. The most recent data from the US Census Bureau indicate that median 
household income has declined. In 2009, it was $49,777 in current dollars. Rather 
than modeling or forecasting projected median income for 2013, the $50,000 figure 
was chosen.
The  income  of  the  $350,000  household  is  obviously  seven  times  that  of  the 
middle-income household. The $350,000 income level is also at the median of 
the $200,000$500,000 adjusted gross income (AGI) category, for which the IRS 
reports detailed tax data.
16
In 2010, those with adjusted gross income above $500,000 accounted for the 
top 0.6 percent of all income returns filed. This implies that the threshold for being 
in  the  top  1  percent  was  somewhere  in  the  $200,000$500,000  category.  The 
$350,000 AGI is likely just below the top 1 percent. The 2010 data indicate that for 
the $200,000$500,000 AGI range, itemized deductions for married-filing-jointly 
returns equaled 16.67 percent of these couples total income. This was used to cal-
culate the deductions the upper-income household took in each year.
The actual calculation of tax liability is straightforward. Data for the exemptions 
and standard deductions in nominal dollars for each year were downloaded from 
the IRS, Forbes Magazine, and the Tax Policy Center.
17
Tax tables were found at the Tax Foundation.
18
 Data for the EITC are from 
The Earned Income Tax Credit by V. Joseph Hotz and John Karl Scholz for the 
National Bureau of Economic Research and Taxation and the Family from the 
 
16. SOI Tax StatsIndividual Statistical Tables by Size of Adjusted Gross Income, IRS, last modified 
May 14, 2014, http://www.irs.gov/uac/SOI-Tax-Stats---Individual-Statistical-Tables-by-Size-of 
-Adjusted-Gross-Income.
17. Robert A. Wilson for the IRS, Personal Exemptions and Individual Income Tax Rates, 19132002, 
data release, accessed May 19, 2014, http://www.irs.gov/pub/irs-soi/02inpetr.pdf; Kelly Phillips Erb, 
IRS Announces 2013 Tax Rates, Standard Deduction Amounts and More, Forbes, January 15, 2013, 
http://www.forbes.com/sites/kellyphillipserb/2013/01/15/irs-announces-2013-tax-rates-standard 
-deduction-amounts-and-more/; Historical Standard Deduction, Tax Policy Institute (Urban Institute 
and Brookings Institution joint venture), May 12, 2014, http://www.taxpolicycenter.org/taxfacts 
/displayafact.cfm?Docid=171.
18. Federal Individual Income Tax Rates History, Tax Foundation, accessed May 19, 2014, http://tax 
foundation.org/sites/taxfoundation.org/files/docs/fed_individual_rate_history_nominal_adjusted-2013 
_0523.pdf.
MERCATUS  CENTER  AT  GEORGE  MASON  UNI VERSI TY
16
Tax Policy Center.
19
 The CTC was calculated using the IRS 2012 formsbecause the 
credit is a fixed dollar amount, it will not change in 2013.
20
The AGIs for the three households were subject to sensitivity analysis. The sin-
gle-parent, low-income households AGI was varied in 12.5 percent increments from 
50 percent of the poverty line to 200 percent of the poverty line; the middle-income 
households AGI was varied in $2,500 increments from $40,000 to $60,000; and the 
upper-income households AGI was varied in $50,000 increments from $250,000 
to $500,000. Although the actual tax liabilities vary, the qualitative conclusions are 
largely unaffected. This spreadsheet is available to any interested reader by emailing 
the author at cbohanon@bsu.edu.
19. V. Joseph Hotz and John Karl Scholz, The Earned Income Tax Credit, in Means-Tested Transfer 
Programs in the United States, ed. Robert A. Moffitt for the National Bureau of Economic Research 
(Chicago: University of Chicago Press, 2003), http://www.nber.org/chapters/c10256.pdf; Taxation 
and the Family: What Is the Earned Income Tax Credit?, Tax Policy Institute (Urban Institute and 
Brookings Institution joint venture), last modified February 12, 2014, http://www.taxpolicycenter.org 
/briefing-book/key-elements/family/eitc.cfm.
20. See http://www.irs.gov/pub/irs-pdf/p972.pdf and http://www.irs.gov/pub/irs-pdf/f1040s8.pdf.