Numbers, Facts and Trends Shaping Your World

A Nation of “Haves” and “Have-Nots”?

Far More Americans Now See Their Country as Sharply Divided Along Economic Lines

by Jodie T. Allen, Senior Editor, Pew Research Center and Michael Dimock, Associate Director for Research, Pew Research Center for the People & the Press


Over the past two decades, a growing share of the public has come to the view that American society is divided into two groups, the “haves” and the “have-nots.” Today, Americans are split evenly on the two-class question with as many saying the country is divided along economic lines as say this is not the case (48% each). In sharp contrast, in 1988, 71% rejected this notion, while just 26% saw a divided nation.

Of equal importance, the number of Americans who see themselves among the “have-nots” of society has doubled over the past two decades, from 17% in 1988 to 34% today. In 1988, far more Americans said that, if they had to choose, they probably were among the “haves” (59%) than the “have-nots” (17%). Today, this gap is far narrower (45% “haves” vs. 34% “have-nots”).


These shifting attitudes have occurred gradually over the past two decades, although the perception of personal financial stringency appears to have risen more rapidly in recent years. As recently as 2001, a 52%-majority still viewed themselves as resting on the positive side of the economic balance, compared with 32% who felt they were monetarily in need. Since then the number of self-described “haves” has fallen by seven percentage points, a decline as large as that which occurred over the previous 13 years.

The share of Americans who see the country as divided along economic lines has also continued to tick upward, though at a somewhat slower rate in recent years (Have/have-not perceptions rose by 18 points over the 13 years between 1988 and 2001 compared with a rise of four points over the last six years).

The increased prevalence of both views — that the country is increasingly divided along economic lines and that a given individual is on the wrong side of that divide — finds support in national economic data. As numerous studies have demonstrated in recent years, income gains over the last few decades have been heavily concentrated at the very top of the income distribution. For example, in an update of their earlier study of long-term U.S. income trends,1 economists Piketty and Saez compute that the share of income going to families in the top 1% of the income scale has doubled from 8% in 1980 to 16% in 2004 even excluding capital gains.2 (For a review of other recent studies see an earlier Pew commentary, “Pinched Pocketbooks: Do Average Americans Spot Something That Most Economists Miss?3)

Meanwhile, Congressional Budget Office data4 show that despite the increase in the number of families with two or more earners and widespread income gains in the latter half of the 1990s, families in the middle fifth of the income distribution realized only a modest $6,600 increase in annual income between 1988 and 2004, while the top 1% of families saw their incomes rise from $839,100 to an average $1,259,700. Recently released Census Bureau data show that in 2006, median household income adjusted for inflation was still 2.1% below its 1999 level.5 More sensationally, recently reported on a study showing that “top private-equity and hedge fund managers made more in 10 minutes than average-paid U.S. workers earned all of last year.”6

Factors Driving Perceptions of an Economic Divide

These objective facts of economic life might seem, at first glance, to be the primary source of the rising perception of a have/have-not nation. And the trends are surely reflected in the growing numbers of Americans who view themselves personally as on the wrong side of that divide. But in judgments about the larger state of the country Americans have traditionally turned a deaf ear to commentaries or analyses that might be characterized as evocative of class warfare, whether grounded in objective facts or not. Successive Pew Global Attitudes polls, for example, find that at every income level, Americans are far more likely than Europeans to believe that individuals, not society, are responsible for their own failures, economic and otherwise.7 Also, economists have pointed to other factors contributing to modest economic gains among middle-income households — such as an aging population and smaller families — that do not necessarily suggest a growing socioeconomic divide among wage and salary earners, as well as to higher levels of consumption and expectations across the income board.


Analysis of polling data over the years also strongly suggests that the growing perception of societal divide is driven as much by political factors as by economic ones. Not that the phenomenon is restricted to those of one political persuasion: Republicans as well as Democrats have recorded an increase among party adherents who see a have/have-not divide: Between 1988 and 2001, the number of Republicans viewing the country as so divided increased from 19% to 34%, declining slightly since then to 33% but still resulting in a net increase of 14 points over the period.

But the rise in the number of Democrats perceiving a divided society started from a higher base (32%) and has risen continuously. Now, in 2007, fully 63% of Democrats see a social divide, a share nearly twice as great as that seen among Republicans. As a result, the increased perception of societal division in recent years coincides with a widening partisan gap in views of American society.

Differences in the current prevalence of this view across other demographic groups are far more modest by comparison, as seen in the table. Although upper-income Americans are less prone to see a divide than are those with middle- or lower-incomes, and college grads are less likely than those with lesser education, only the difference between black and white adults (67% and 45% respectively) rivals the partisan gap in size.

Nor are substantial differences seen in the degree to which persons in various income and demographic categories have become more likely to see a “have/have-not” division over the 1988-to-2007 period. One exception is that people who are middle-aged record a somewhat higher percentage-point increase in this perception compared with other age groups. The same is true of people living in the eastern part of the country compared with those in other regions. But these within-category variations are dwarfed by the political affiliation divide.

This tendency to view the national economy through the prism of politics is not a new phenomenon: Over the last two decades, the gap between Republicans and Democrats on opinions about income distribution has consistently been larger than the gap between upper- and lower-income respondents. Nor is this partisan cleavage unique to opinions about the nation’s economic divide. As earlier Pew analysis has shown, a similarly strong political influence is observed in judgments about the current and future state of the national economy.8 As noted there, the causal link between political party affiliation and perceptions about the larger state of the economy and the nation may operate in both directions: Families in the top income-tiers are both heavily Republican and, understandably, more likely to look with favor upon the nation’s larger economy and their own place within it.


Still while many people may view both their choice of political party and the state of the nation through the prism of personal economic circumstance, other factors may dominate. For example, evangelical Christians are among the GOP’s strongest adherents, yet Pew surveys find that their average incomes are somewhat lower than those of other Protestants or of seculars.

To filter out the independent effects of party affiliation, income and other demographic variables including income, race, sex and education, multiple regression analyses were run on the July 2007 data.

As seen in the chart, political party affiliation is far and away the most important independent determinant of views about whether the country is divided between “haves” and “have-nots.” Race is the second most important determinant, though its potency is only half that of party while income and education have virtually no independent effect on such perceptions.

Which Side of the Divide Are You On?


This sharp partisan partition is not observed, however, when the focus shifts from the state of the nation to the state of one’s personal finances. Not surprisingly, views about one’s own position relative to the income divide are heavily influenced by the independent factor of personal income. While party affiliation remains a significant influence, it is dominated by the objective reality of economic status as well, to a lesser degree, by race.

A more detailed look at the increased prevalence of have-not status across various economic and demographic categories shows surprisingly little variation across groups. Although Republicans remain more likely to see themselves as among society’s “haves,” (50% do so now, as compared with 44% of Democrats), both groups record a substantial decline in the share of those who count themselves among the economically favored (13% and 12% respectively).


Declines in perceived personal economic status vary little among those in the top, middle and bottom thirds of the income distribution. However, middle-income families record the largest drop (from a solid 61%-majority in 1988 to a 43%-minority in 2007) in the number of those seeing themselves as among America’s “haves.”

Among age groups, young people (ages 18-29) are the most likely to see themselves as “haves,” although the share taking that view fell by a sizeable 19 percentage points over the years between 1988 and 2007. Those ages 30-49 experienced a similar decline in perceptions of “have-ness” but starting from a lower base so that the middle-aged are now the most likely age group to classify themselves as “have-nots.” Interestingly, the elderly (65 and over), while recording a substantial gain in perceptions of economic status in 2001 (59% haves vs. 45% in 1988) have now reverted to their 1988 reading.

Women are more likely to see themselves as have-nots (37%) than are men (30%) and the gap between the sexes has widened slightly since 1988. Across geographic regions, persons residing in the South experienced the smallest decline in the number classifying themselves as “haves” (minus 10 percentage points compared with declines of 15 points in the East and West and 17 points in the Midwest). As a result, the South now registers, by a small margin, as the economically most satisfied region in the country.

Haves, Have-Nots and Horse Races


What ramifications might these differing perceptions of a socioeconomic divide have for the coming 2008 primary elections? At the moment the likely impact seems slight. A look at voter preferences among the leading candidates in both political parties at the end of July shows little if any difference between those who see the country divided between haves and have-nots and those who see no such schism.

On the Republican side, the only significant difference is a somewhat stronger showing for John McCain among those who see a divide than among those who do not, a preference perhaps accounted for by the higher proportion of Republican-leaning independents among his supporters.


On the Democratic side, only Barack Obama receives significantly stronger support among those seeing America as a nation split along economic lines — nearly a quarter (24%) of such — name him as their preferred Democratic candidate for president compared with 16% among those seeing no such divide – a finding consistent with other analyses showing Obama recording greater popularity among liberal Democrats than among other factions of the party. By contrast, John Edwards, despite his overtly populist campaign message, draws equal support from those on their side of the “divide” divide.

This is not to say that such perceptions will not play some role in the subsequent general election. But given the strong correlation between partisan preference and public opinion on the issue, it seems unlikely that the have/have-not question will play a strong independent role in the final outcome.

About this Survey

Results for this survey are based on telephone interviews conducted under the direction of Schulman, Ronca & Bucuvalas, Inc. among a nationwide sample of 1,503 adults, 18 years of age or older, from July 25-29, 2007. For results based on the total sample, one can say with 95% confidence that the error attributable to sampling is plus or minus 3 percentage points.


1Piketty, Thomas and Emmanuel Saez. “Income Inequality In The United States, 1913-1998,” Quarterly Journal of Economics, 2003, v118 (1,Feb), 1-39.

2See “Thomas Piketty and Emmanuel Saez Respond to Alan Reynolds,” Economist’s View, Jan. 7, 2007.

3Allen, Jodie T. and Andrew Kohut, “Pinched Pocketbooks: Do Average Americans Spot Something That Most Economists Miss?” Pew Research Center, March 28, 2006.

4Congressional Budget Office, Historical Effective Federal Tax Rates: 1979 to 2004, December 2006.

5U.S. Census Bureau, “Income, Poverty and Health Insurance in the United States: 2006,” August 2007 release. The report finds that median income for “white households rose 1.1 percent between 2005 and 2006 — the first real increase in annual household income for this group since 1999. The changes in median household income for the remaining race groups and Hispanics were not statistically significant.” Also, see Center on Budget and Policy Priorities, “Number and Percentage Of Americans Who Are Uninsured Climbs Again,” Aug. 31, 2007.

6Ian Katz, “Top Fund Managers Made 22,300 Times Average Wage, Study Says,”, August 29, 2007.

7See Kohut, Andrew and Bruce Stokes, America Against the World: How We Are Different and Why We Are Disliked, Times Books, 2006, Chapter 3.

8Allen and Kohut, “Pinched Pocketbooks,” op.cit.

Icon for promotion number 1

Sign up for our weekly newsletter

Fresh data delivery Saturday mornings

Icon for promotion number 1

Sign up for The Briefing

Weekly updates on the world of news & information