The Distribution of Household Income and
the Middle Class
Linda Levine
Specialist in Labor Economics
November 13, 2012
Congressional Research Service
7-5700
www.crs.gov
RS20811
CRS Report for Congress
Pr
epared for Members and Committees of Congress

---

The Distribution of Household Income and the Middle Class

Summary
Although not itself a subject of legislation, the shape of the income distribution enters Congress’s
decision-making process concerning such policy issues as taxes, means-tested benefits, and social
insurance programs. Congress also considers legislation specifically in the name of those in the
middle class, which is variously defined as some income level or income range within the
distribution of U.S. households with income. After briefly analyzing the distribution of household
money income in 2011, the report attempts to put the term “middle class” into some perspective.
The first key point of the report is that, although there are a variety of ways to describe the
income distribution, all show that income is concentrated among high-income households.
Relatively few households can be found in the upper end of the income distribution. Of the
121,084,000 households with income in 2011, only 2.3% had incomes of at least $250,000. (The
Census Bureau does not disaggregate income within the $250,000-or-more income class.) In
addition, a disproportionately large share of total money income accrues to those at the upper end
of the distribution. In 2011, the top 5% of U.S. households with income accounted for 22.3% of
total income, and the top 20% of households (which includes the top 5%) had 51.5% of all money
income.
The second major point is that there is no official government definition of who belongs to the
middle class, and the term means different things to different people. The middle class may refer
to a group with a common point of view or to those having similar incomes, for example.
Thirdly, absolute income appears to partly determine who belongs to the middle class. By
combining money income data from the latest Annual Social and Economic Supplement to the
Current Population Survey with results from surveys that asked people to identify their social
class, the middle class may refer to households with income levels in 2011 that ranged from
$38,521 (the bottom of the middle quintile, 20%, of households) and extended into the top
quintile (households with income of $101,583 or more)—perhaps including households with
incomes somewhat over $200,000.
Lastly, relative income may also be a defining characteristic of the middle class. In other words,
the middle class appears to identify itself relative to the income of a reference group (e.g., their
neighbors or coworkers). According to studies of self-reported well-being, those who constitute
the middle class seemingly are of like minds with regard to their economic situation. Specifically,
having incomes far above those at the lower end of the income distribution appears to be a source
of satisfaction to the middle class, but when those at the upper end of the distribution fare much
better than they do, it can be a source of consternation to the middle class. As authors of one
study put it, staying ahead of the Smiths and keeping up with the Joneses is important to the
middle class. This outlook may in part explain the antipathy expressed in some quarters toward
the compensation of senior executives, among others, at some of the nation’s largest corporations
generally and firms in the financial services industry specifically.
Congressional Research Service

---

The Distribution of Household Income and the Middle Class

Contents



The Distribution of Household Income ........................................................................................... 1
The Middle Class ............................................................................................................................. 3
Absolute Income ........................................................................................................................ 3
Relative Income ......................................................................................................................... 5

Tables
Table 1. Distribution of Household Money Income by Selected Income Class, 2011..................... 2
Table 2. Distribution of Household Money Income by Quintile, 2011 ............................................ 3

Contacts
Author Contact Information............................................................................................................. 7

Congressional Research Service

---

The Distribution of Household Income and the Middle Class

he shape of the income distribution is not itself a subject of legislation, but Members of
Congress appear to consider it in their decision-making process concerning a number of
Tpolicy issues such as taxes, means-tested benefits, and social insurance programs.
Congress also takes up legislation specifically in the name of those in the middle (however
defined) of the income distribution who commonly are referred to as the middle class. Some
Members have, for example, proposed bills to improve U.S. competitiveness as a means of
increasing exports manufactured by workers in “good” (middle-class) jobs. Similarly, training
policy generally has been crafted to provide individuals with the skills thought necessary to attain
a middle-class standard of living.
This report first presents a brief analysis of the distribution of income across households in 2011,
the latest year for which annual data are available from the Census Bureau.1 It then attempts to
put the term middle class into some perspective: first, by applying results from public opinion
surveys on social class to the Census Bureau’s data on the income distribution in 2011; second, by
reviewing findings from empirical studies on the contribution of relative (as opposed to absolute)
income to the identity of the middle class.
The Distribution of Household Income
The Census Bureau conducts a survey from which it derives annual estimates of the distribution
of income across households, families, and individuals with income.2 Households, which are
more numerous than families, appear to be most relevant to the discussion here because
household members are assumed to pool whatever income they have and share a common
standard of living.3
The Bureau’s official measure of income is “money income.” It includes earnings, interest,
dividends, pensions, child support, and government nonmeans-tested income (e.g., Social
Security benefits, unemployment compensation, and veterans’ payments). Money income is
calculated on a pre-tax basis; that is, it excludes payments for personal income taxes, Social
Security, Medicare, union dues, etc. It also excludes the value of noncash means-tested benefits
(e.g., food stamps, public housing subsidies), employer-provided fringe benefits, and capital
gains. Money income is the measure on which official estimates of poverty are based.4
Data describing the distribution of income across households can be presented in a number of
ways. The top portion of Table 1 displays the number and share of households with income by

1 For information on trends in the distribution of income, see CRS Report R42400, The U.S. Income Distribution and
Mobility: Trends and International Comparisons, by Linda Levine.
2 The complete Census Bureau report (Income, Poverty, and Health Insurance: Coverage in the United States: 2011) is
available at http://www.census.gov/prod/2012pubs/p60-243.pdf.
3 In the Current Population Survey (CPS), a household is defined as all of the individuals who occupy a housing unit as
their usual place of residence. A family is defined as a group of two or more individuals who reside together and who
are related by birth, marriage, or adoption. A household may be composed of one or more families or no families at all;
that is, a person living alone in a housing unit is counted in the CPS as a household.
4 The Census Bureau periodically calculates alternative measures of income which can be used to assess the effects of
federal (cash and noncash) transfers, capital gains, employee health benefits, return on own-home equity, federal and
state income taxes, payroll taxes, and the Earned Income Tax Credit on the shape of the income distribution and on the
number of persons living below the poverty threshold. The latest estimates cover 2009, and are available at
http://www.census.gov/hhes/www/cpstables/032010/rdcall/toc.htm.
Congressional Research Service
1

---

The Distribution of Household Income and the Middle Class

income class. The income class sizes range from about $5,000 toward the lower end of the
distribution to about $50,000 at the upper end of the distribution, in which relatively few
households are found. In 2011, 1.9% of households (2,297,000 out of 121,084,000) had incomes
between $200,000 and $249,999 for example. Another 2.3% of households (2,808,000 out of
121,084,000) had incomes of $250,000 or more. (The Census Bureau does not disaggregate
income within the group of households with incomes of $250,000 or more.)
Table 1. Distribution of Household Money Income by Selected Income Class, 2011
Income Class
# of Households
(in thousands)
% of Households
Al Households
121,084
100.0
Less than $5,000
4,261
3.5
$5,000 to $9,999
4,972
4.1
$10,000 to $14,999
7,127
5.9
$15,000 to $19,999
6,882
5.7
$20,000 to $24,999
7,095
5.9
$25,000 to $29,999
6,591
5.4
$30,000 to $34,999
6,667
5.5
$35,000 to $39,999
6,136
5.1
$40,000 to $44,999
5,795
4.8
$45,000 to $49,999
4,945
4.1
$50,000 to $59,999
9,420
7.8
$60,000 to $69,999
8,268
6.8
$70,000 to $79,999
7,058
5.8
$80,000 to $89,999
5,602
4.6
$90,000 to $99,999
4,845
4.0
$100,000 to $124,999
9,130
7.5
$125,000 to $149,999
5,311
4.4
$150,000 to $199,999
5,875
4.9
$200,000 to $249,999
2,297
1.9
$250,000 and above
2,808
2.3
Median Income
$50,054
Mean Income
$69,677
Source: U.S. Census Bureau, 2011 Annual Social and Economic Supplement to the Current Population Survey.
The concentration of income among the relatively few households in the upper tail of the income
distribution also can be seen by comparing two summary measures of the shape of the income
distribution shown in the last two rows of Table 1. One summary measure, median income, is the
level below and above which one-half of all households lie. It is a better indication of the income
of the “typical” household than is the mean (average), which can be greatly affected by the upper
end of the distribution. One-half of households had income above the median of $50,054 in 2011,
whereas many fewer households (about one-third) had income above the mean of $69,677. The
Congressional Research Service
2

---

The Distribution of Household Income and the Middle Class

mean being greater than the median reflects the concentration of total money income among high-
income households.
Another often-used way to describe the distribution of income is the share of total money income
going to different household groups in order of increasing income. In 2011, there were a total of
121,084,000 households with income, so each quintile (fifth) in the distribution represents
24,217,000 households. The Census Bureau separately calculates the total income share of the top
5% of households because these 6,054,000 households account for a disproportionately large
percentage of aggregate income. As shown in Table 2, the top 5% of households accounted for
22.3% of all income in 2011. By comparison, the lowest quintile (20%) of households had only
3.2% of aggregate income. This pattern, like the relationship between median and mean income,
illustrates the unequal distribution of income across U.S. households.
Table 2. Distribution of Household Money Income by Quintile, 2011
Quintiles

Lowest
Second
Third
Fourth
Highest
Top 5%
Number
of
households 24,217,000 24,217,000 24,217,000 24,217,000 24,217,000 6,054,000
Range of income class
$20,262
$20,263
$38,521
$62,435
$101,583
$186,000
or less
to $38,520
to $62,434
to $101,582
or more
or more
Share of household
3.2 8.4 14.3 23.0 51.5
22.3
income (%)
Source: U.S. Census Bureau, 2011 Annual Social and Economic Supplement to the Current Population Survey.
The Middle Class
Although Congress often considers legislation specifically in the name of the middle class, there
is no official government definition of the group, and it is not the aim of this report to develop
one. What constitutes the middle class is subjective and comparative, meaning different things to
different people. The term may refer to a group with a common point of view or to those having
similar incomes. Most people likely have decided views as to whether they are middle class, and
those who refer to the middle class have a rough idea of whom they have in mind. How closely
these definitions correspond is another matter.
Absolute Income
Any discussion of the middle class necessarily starts with those at the very middle of the income
distribution. As previously discussed, median household income was $50,054 in 2011. How far
the middle class stretches above and below the median is the question.
A narrow view of the middle class might include only those in the middle (third) quintile, that is,
those households with money income between $38,521 and $62,434 in 2011 (see Table 2). It
seems unlikely, however, that so narrow an income range would account completely for those
commonly considered to be middle class. For that reason, the three middle quintiles are often
combined to broaden the definition. In 2011, according to this broader definition, the 72,651,000
households with income above $20,262 and below $101,583 might have composed the middle
Congressional Research Service
3

---

The Distribution of Household Income and the Middle Class

class. These 60% of all households in the distribution accounted for a smaller share (45.7%) of
total income in 2011.
A still broader definition of middle class can be constructed by adding that part of the top quintile
up to the point at which the top 5% of households begins. In 2011, then, 90,814,000 households
with incomes above $20,262 and below $186,000 might have composed the middle class. These
75% of all households in the distribution accounted for a slightly larger share (78%) of total
household income in that year.
The results of opinion surveys offer further insight into the income bounds that may define the
middle class. A few surveys asked people to indicate their social class. In a 2005 New York Times
survey, for example, 35% of respondents identified themselves as working class and 7% said they
were part of the lower class. Another 42% considered themselves to be middle class and 15% said
they were part of the upper middle class, whereas 1% of respondents identified themselves as
members of the upper class.5 In contrast, the top income class ($250,000 or more) in the Census
Bureau’s detailed data on the household income distribution shown in Table 1 accounted for
2.3% of all households. Assuming some correspondence between the survey responses and
Census income data, this discrepancy suggests that the 15% of respondents who said they were
upper-middle class includes some households with incomes over $250,000. Comparing just the
42% of respondents who reported being middle class with the Census data suggests that the upper
bound of the middle class might instead be lower at somewhat under $125,000. The lower bound
of the middle class might be just over $40,000, based on a comparison of the 42% of respondents
who reported they were working or lower class with the Census data.
The National Opinion Research Center (NORC) at the University of Chicago began asking
people in 1972 to identify their social class in the General Social Survey (GSS). The cumulative
results of self-reported class divisions in the GSS can be compared with the Census income
distribution as was done above with the Times survey. According to the 1972-2010 GSS data,6
3.2% of the population consider themselves to be upper class. This would put the dividing line
between middle and upper class at somewhat over $200,000. The 1972-2010 GSS results further
indicate that 5.7% of the population consider themselves to be lower class and 45.7% identified
themselves as working class. This would put the lower income level for the middle class at just
over $50,000.
As part of a larger study of the middle class, the Pew Research Center sampled opinions
regarding social class in 2012. It found that 51% of surveyed adults living in households with
incomes between $30,000 and $49,999 identified themselves as middle class and 65% of
surveyed adults with household incomes between $50,000 and $99,999 identified themselves as
middle class. Although these results suggest that the income boundaries of the middle class might
extend from $30,000 to $99,999, the finding that 46% of surveyed adults with household incomes
of $100,000 or more self-identified as middle class suggests that the upper bound of the middle
class might exceed $100,000. But by how much? The Pew survey also found that 7% of
Americans consider themselves to be lower class; 25%, lower middle class; 49%, middle class;
15%, upper middle class; and 2%, upper class.7 If the 15% of survey respondents who considered

5 The New York Times published a series of articles in May 2005 under the title “Class Matters,” which is available at
http://www.nytimes.com/indexes/2005/05/15/national/class/index.html.
6 NORC, General Social Survey, http://www.norc.uchicago.edu/GSS+Website/.
7 Pew Research Center, The Lost Decade of the Middle Class, August 22, 2012, available at
http://www.pewsocialtrends.org/files/2012/08/pew-social-trends-lost-decade-of-the-middle-class.pdf. The results from
(continued...)
Congressional Research Service
4

---

The Distribution of Household Income and the Middle Class

themselves upper middle class were added to the middle class rather than upper class category as
Pew has done, then the upper bound of the middle class might be somewhat over $200,000.
Taken together, the survey responses suggest that the lower bound of the middle class might start
in the middle (third) quintile of the income distribution shown in Table 2 (households with
income of $38,521 to $62,434). The surveys’ results further suggest that the middle class extends
into the top (fifth) quintile of the income distribution (households with income of $101,583 or
more)—perhaps including households with income somewhat over $200,000.8
Relative Income
The satisfaction (utility) that households derive from their income appears to be associated both
with their absolute level of income and how that income level compares with the incomes of
others.9 While the middle class may be content with absolute income levels that afford them a
much better-than-subsistence standard of living, the satisfaction of the middle class also may be
affected by their income relative to a reference group. If individuals care about how their income
compares with that of their neighbors or coworkers, for example, it lends support to the notion of
like-income groups possessing a shared outlook on economic (and perhaps other) policy.
Since the 1990s, economists have increasingly turned their attention to identifying factors that
contribute to utility (well-being or happiness), including one’s position relative to others.10 One
study, for example, analyzed whether group-specific suicide rates are associated with
interpersonal comparisons of income.11 Daly and Wilson used the suicide rate as an indicator of
current and expected future happiness, and developed three measures of relative income: the ratio
of income at the 90th percentile to median income (90/50), the ratio of median income to income
at the 10th percentile (50/10), and the ratio of income at the 90th percentile to income at the 10th
percentile (90/10). Among their findings was that an increase in the 50/10 ratio was associated
with a decrease in the suicide rate, and that an increase in the 90/50 ratio was associated with an
increase in the suicide rate. That is to say, those in the middle-income group (the 50th percentile)
were (a) happier the larger the gap was between them and those at the low end of the distribution

(...continued)
the 2012 survey are similar to those from a 2008 survey conducted by Pew, Inside the Middle Class: Bad Times Hit the
Good Life, available at http://pewsocialtrends.org/pubs/706/middle-class-poll.
8 If those who consider themselves to be upper class were underrepresented in the surveys, the number of those who
consider themselves to be upper class would be larger. If that is the case, the actual level of income that divides the
middle from the upper class would be lower than is suggested by the surveys.
9 Disagreement in the economics literature exists about the extent to which absolute versus relative income contributes
to well-being and about what is the most appropriate measure of well-being. See, for example, Richard Ball and
Kateryna Chernova, “Absolute Income, Relative Income, and Happiness,” Social Indicators Research, September
2008, pp. 497-529; Daniel Kahneman and Angus Deaton, “High Income Improves Evaluation of Life but Not
Emotional Well-Being,” Proceedings of the National Academy of Sciences, vol. 107, no 38, September 2010, pp.
16489-16493; Richard Layard, Guy Mayraz and Stephen Nickell, Does Relative Income Matter? Are the Critics
Right?, CEP Discussion Paper No. 918, March 2009; and Betsey Stevenson and Justin Wolfers, “Economic Growth
and Subjective Well-Being: Reassessing the Easterlin Paradox,” Brookings Papers on Economic Activity, 2008, pp. 1-
87.
10 Andrew E. Clark, Paul Frijters, and Michael A. Shields, “Relative Income, Happiness, and Utility: An Explanation
for the Easterlin Paradox and Other Puzzles,” Journal of Economic Literature, vol. 46, no. 1 (2008), pp. 95-144.
11 Mary Daly and Dan Wilson, “Keeping Up with the Joneses and Staying Ahead of the Smiths: Evidence from Suicide
Data,” Federal Reserve Bank of San Francisco Working Paper 2006-12, April 2006.
Congressional Research Service
5

---

The Distribution of Household Income and the Middle Class

and (b) unhappier the larger the gap was between them and those at the high end of the
distribution. Daly and Wilson considered the results evidence that relative income influences a
group’s sense of well-being. More specifically, increases in inequality at the upper and lower tails
of the income distribution appear to have opposite effects on the satisfaction of the middle-
income group.
Luttmer also presented evidence that relative income is an important determinant of well-being.12
He matched survey data on self-reported happiness and other measures of well-being (e.g.,
frequency of marital disagreements, score on a depression index) with income data by locality.
Luttmer estimated that an increase in neighbors’ income had a negative effect on one’s own
happiness. Dynan and Ravina built on Luttmer’s study by using the GSS to assess the effect of
differences in relative incomes over recent decades on happiness. Like Luttmer, they found that
relative income is strongly associated with self-reported happiness. Dynan and Ravina further
estimated that relative income most affects well-being when a group’s income is above the
average in its state but not at the top (90th percentile) of the income distribution. Or, as the authors
put it, “relative concerns primarily affect people who have above-average income but are not
extremely rich.”13
Guven and Sorensen concluded that relative income plays a substantial role in explaining self-
reported happiness among the middle-income group.14 Because the GSS asks respondents how
they think their income compares with that of the average family, the economists also were able
to measure the relationship between happiness and perceived (as opposed to actual) relative
income. Guven and Sorensen estimated that well-being is more affected by the perception of
relative income than by actual relative income for the middle-income group. They further found
that perceptions about relative income are more important in explaining the well-being of the
middle-income group than the happiness of the low- and high-income groups.
The conclusions of these studies might not seem robust because they are based on self-
assessments of well-being, but if they are correct, they appear relevant to the notion of like-
minded income groups. Together, the studies suggest that the satisfaction of the middle class
(however defined) depends not only on keeping up with the Joneses but also staying ahead of the
Smiths, as Daly and Wilson put it in their study. In other words, whatever else they may have in
common, those who constitute the middle class seemingly have similar sentiments with regard to
their economic situation. Specifically, having incomes far above those at the lower end of the
income distribution is a source of satisfaction to the middle class, but when those at the upper end
of the distribution fare much better than they do, it can be a source of consternation to the middle
class. This outlook may partly explain the current antipathy expressed in some quarters toward
the compensation of senior executives, among others, at some of the nation’s largest corporations
generally and firms in the financial services industry specifically.


12 Erzo F.P. Luttmer, “Neighbors as Negatives: Relative Earnings and Well-Being,” Quarterly Journal of Economics,
August 2005.
13 Karen E. Dynan and Enrichetta Ravina, “Increasing Income Inequality, External Habits, and Self-Reported
Happiness,” AEA Papers and Proceedings, vol. 97, no. 2 (May 2007), pp. 226-231.
14 Cahit Guven and Bent E. Sorensen, Subjective Well-Being: Keeping Up With the Joneses, Real or Perceived?,
November 2007, available at http://www.uh.edu/~cguven/papers/JonesesCahit_SEP262007.pdf. See also Cahit Guven
and Bent E. Sorensen, “Subjective Well-Being: Keeping Up with the Perception of the Joneses,” Social Indicators
Research, August 2011, pp. 1-31.
Congressional Research Service
6

---

The Distribution of Household Income and the Middle Class

Author Contact Information

Linda Levine

Specialist in Labor Economics
llevine@crs.loc.gov, 7-7756




Congressional Research Service
7

---