Exceptionally Unequal: US Income Inequality in Comparative Perspective

8/2/2019 Exceptionally Unequal: US Income Inequality in Comparative Perspective

1/47

Exceptionally Unequal:

US Income Inequality in Comparative Perspective

Joe Hines

Over the past forty years, the US has seen a steep increase in income inequality,

even as many other advanced democracies have seen their income distribution

decline or remain steady. A countrys income distribution results from domestic

political decisions, as countries have similar levels of inequality before taxes and

redistributions but vastly different levels afterwards. Scholars have offered

various explanations for why income inequality has risen in the U.S, including the

unrepresentative nature of American political institutions, Americans' attitudes

toward inequality, declining levels of unionization, and the partisan polarization

of elites. This paper explores the merits of these explanations from a comparative

perspective. First, I find that the although the American political system has the

highest number of anti-majoritarian features among advanced industrial

democracies, those features have not changed significantly over time are not

sufficient to explain why income inequality has increased in the U.S. Second,

although Americans may be somewhat more permissive about economic

inequality than Europeans, Americans' attitudes have remained largely

unchanged over the past 35 years and so cannot during the increase in inequality

during that period. Third, the low percentage of workers covered by collective

bargaining agreements in the United States compared to other democracies

bolsters the argument that the decline in unions contributes to rising income

inequality. Finally, the most striking correlation has been the increase in political

polarization in Congress. This polarization has made it inordinately difficult to

pass legislation that would lessen inequality.

1


2/47

Table of Contents

Introduction 3Its not the economy, stupid 5Culture: Mass Attitudes, Misperceptions, and Voter Turnout 12The Decline of Unions and Collective Bargaining Coverage 23Institutional Explanations: The Number of Veto Players and the

Unrepresentative Senate 29Elite Politics: Polarization

35

Conclusion 39Bibliography 44

2


3/47

Introduction

Originally, it was Europe that was unequal. In 1831, Alexis de Tocqueville visited the United

States to report on the progress of democracy, later published asDemocracy in America. As a

French aristocrat, Tocqueville analyzed American democratic institutions to inform Europes

political institutions. Americas love of equality struck Tocqueville as unique, a factor he

attributed to the countrys democratic nature. In his opening paragraph, Tocqueville noted that

in the United States, nothing struck [him] more forcefully than the general equality of

conditions (Tocqueville 1). Although most developed in the United States, in Tocquevilles view

of the progress of civilization, we shall scarcely meet with a single great event, in the lapse of

seven hundred years, which has not turned to the advantage of equality (Tocqueville 3). The

European aristocrat was diminished as the American citizen ascended. A new era of equality,

stustained through democratic institutions, would prevail. Tocqueville believed, with some

trepidation, that the role of a new political science was to guide the inevitable march towards

equality.

Over the past the last forty years, the only inevitability about equality in the US, at leasteconomically, has been its steady decline.Democracy in America contains many prescient

observations, but Tocquevilles theory about an irresistible progress towards greater equality of

conditions led by the US has not fared well. Today, Tocquevilles old, aristocratic Europe is

significantly more equal than the United States. The income gap narrowed after WWII and

remained equitable until the late 1970s. Since then, the difference between the rich and poor in

the US has steadily increased. This increase has been more pronounced in the United States than

any of its democratic and developed peers.

3


4/47

Were Tocquevilles predictions premature or have the political circumstances changed?Political scientists have struggled to grapple with the rise in income inequality due to the myriad

of contributing factors. Since WWII the United States has undergone many cultural, political and

economic changes. Still, those changes were not all relegated to the United States, but the US has

seen an anomalous increase in income inequality compared to its peers. If US politics are to

blame for its increasing income inequality, understanding the similarities and differences

between countries polities helps to illuminate the cause.

A frequent scapegoat for the increase in income inequality is the rise in globalization. Abasic look at the United States compared with its peers undermines this theory. After-tax income

inequality has been rising in the US disproportionately to its peers from 1979 until the mid

2000s. An American Political Science task force reported that:

the top quintile has enjoyed more than 40 percent of income in the United States

since 1947. What stands out over the past three decades is the sharp and

unmistakable increase in the concentration of income at the top. (APSA 652)

Although its concentration of wealth indicates that the United States was never as equitable as

observers like Tocqueville believed, the recent trend towards greater inequality has been

extreme. In 1915, Willford I. Kings influential The Wealth and Income of the People of the

United States, one of the first systematic surveys of the U.S. economy, noted, with concern, that

the richest 1.6 percent of the population controlled 19 percent of the income annually (King

239). Today, the top 1 percent control 23.5 percent annually, the highest proportion since 1928

(Hacker 15). Figure 1 shows the change in after-tax Gini coefficients since 1968 within the

United States and comparable democratic and developed countries. The Gini coefficient, a

common measure of inequality, represents the ratio between the amount of income earned by

4


5/47

each income bracket compared with the percentage of population.1 Based on their analysis of the

Gini coefficient, comparativists Stepan and Linz observe that by 1968, equality had improved

greatly, with the Gini index of inequality falling to .388, the best Gini ever recorded in the United

States (Stepan 843). Figure 1 demonstrates that income inequality has increased since the 1970s

and this upward trend is relegated to English speaking countries. Before the trend began, the

United States was in the middle of its peers in terms of income inequality. Afterwards, the US

had the highest Gini coefficient of all countries surveyed.

The United States exceptional growth in income disparity calls for a comparativeappraisal. If domestic policy drives income inequality, then certainly comparing the domestic

policies of the US and its peers will provide insights into how and why inequality has risen. Has

political polarization risen in other democracies? Has union density and bargaining strength

declined as well? Are our political institutions, including polarization and the structure of our

legislature, truly unique? Finally, are voters in the US just more accommodating of income

inequality than voters in other democracies? Analyzing the cross-national differences in several

domestic policy trends unravels the puzzle at why our political system has unexpectedly, at least

to Tocqueville, failed in achieving greater income equality.

Its not the economy, stupid

5

1 The World Bank defines the scale of the Gini Coefficient as between 0, which reflects complete equality and 1,

which indicates complete inequality (World Bank 1). The Gini ratio expresses the ratio of difference between the

percentage of income controlled by each income bracket as compared with a line of perfect equality. The line of

perfect equality would reflect every part of society controlling an equal percentage of the income.The ratio of the

difference between these figures is the Gini coefficient. Small changes in the Gini coefficient can indicate large

fluctuations in income inequality.


6/47

One explanation for the rise of inequality, often championed by conservative commentators, is

that it inevitably results from globalization. However, the increase in income inequality has not

occurred to the same extent among similarly globalized countries. Since 1978, the only countries

to see a consistent increase in income inequality have been the United Kingdom and Canada. In

fact, Finland, Sweden, and France have lower income inequality now then they did in 1968.

A change in a countrys Gini coefficient after government steps in does not indicatewhether rising income inequality has been the result of politics or economics. Figure 1 shows the

Gini coefficient after taxes and redistribution. Since 1968, almost every country has seen its

income inequality increase before government taxation and redistribution. Figure 2 illustrates

how this trend has been almost universal among advanced democracies since 1975. The rise in

pre-tax income inequality shows how underlying market trends contribute to income inequality.

Figure 2 demonstrates that, while the market contributes to income inequality, pre-tax pre-

taxincome inequality can be corrected through politics. For example, the Netherlands and France

have both seen distinct increases in income inequality before taxes, but decreasing income

inequality after taxes.

6


7/47

Figure 1

Source: OECD Dataset

Figure 2

Source: OECD Dataset

7


8/47

Tax policy profoundly affects the Gini coefficient. The differences between Figure 1 andFigure 2 show the role that government has in shaping its countrys income distribution. Taxes

and redistribution account for almost all the differences between countries Gini coefficients.

Germany has an even higher income inequality before taxes than the United States, but after

government redistribution it ends up in the middle of the pack. As indicated in Figure 3, the

direct redistribution of income dramatically changes the United States position related to other

nations. The US goes from being in the middle of income inequality among long-standing

democracies to being the most unequal after taxes. Further 3 further indicates that the rise in

income inequality is due to differences in how governments choose to redistribute income.

Figure 3:

Source: OECD Data

8


9/47

The difference between the tax policies of the United States and comparable democracieshas become more pronounced in the past forty years. In the mid 1970s, US tax policies were in

line with that of other advanced democracies. In the 1980s, however, as pre-tax income

inequality began to increase, almost all other advanced countries have met growing income

inequality with greater government redistribution of income through taxes. This change is

particularly evident when one looks at the changing tax rates for the top 1, .1 and .01% of wage

earners over the past forty years. Figure 4 shows that, in the United States, tax rates have almost

always been lowered since 1978. This impacts income inequality, because the economic growth

of the global economy went increasingly to the top 1% of wage-earners during from 1978

onwards.

Figure 4: Changes in Total Federal Tax Rate

9


10/47

Source: Leonhardt 1

Understanding the changes that reduced inequality during the Great Compression helps toilluminate what has happened since. In Conscience of a Liberal, Paul Krugman shows how the

Great Compression, produced by the New Deals economic policy changes, was a defining

event in American history (Krugman 39). Krugman argues that there was no trend towards

income leveling out until the 1930s when, as a result of domestic policy, it happened quite

suddenly. Krugman identifies tax changesas a principal cause of this decline in income

inequality, noting that by the mid 1950s:

The real after-tax incomes of the richest 1 percent of Americans were probably 20

or 30 percent lower than they had been a generation earlier. And the incomes of

the really rich--say those in the top tenth of one percent-- were less than half what

they had been in the twenties. (Krugman 42)

Figure 5 reinforces Krugmans claim, showing the decrease in the amount of income controlled

by the top .1% after the outbreak of WWII.

The diminishing tax rate for the top 1% helps to explain why the change in incomeinequality is even more significant when one looks at what is happening with the top 1% of

wage-earners in addition to the Gini ratio.2 The seminal study in US income inequality of the top

1% in the US is Thomas Piketty and Emmanuel Saezs work exploring after tax income changes

in from 1913 until 2000. Piketty and Saez look at tax returns to construct a clear picture of the

trend in changes in the top 1%, .1% and .01% share of national income and how it changed over

the 20th century. While other income brackets have kept a largely similar share of the annual

10

2 The amount of income given to the top 1% has been demonstrated to correlate to higher rates of economic

inequality in other measurements, including the Gini coefficient (Leigh 1).


11/47

income in the United States,3 the top 1% as a whole has seen its annual income share grow from

10% to nearly 25% in 2007 (Atkinson 6). Even more dramatically, the top .1% has gone from

controlling 2% of the income in the 1960s to controlling nearly 6% in 1998 (Atkinson 1). Figure

5 demonstrates how this trend has been exceptional in the United States by comparing income

controlled by the top .1% in the United States, United Kingdom and France. After closely

following the trajectory of the United Kingdom and France, the .1% in the United States began to

control an increasing share of annual income in the late 1970s. Figure 5 demonstrates that the

top .1% has only dramatically increased its income share in the United States and the United

Kingdom, with its increase being more extreme in the United States.

Figure 5: Top .1% Income Share in the US, UK and France

Source: Atkinson 720

11

333 This trend is reinforced by whats happened in the top 10% as well. Andrew Leigh shows that, in addition to the

ratio getting larger in terms of the Gini coefficient, the top 10% has been earning more and more of the United

States annual income. Leigh also illustrates that, by this metric, income inequality increased markedly in the past

thirty years in the United States and other English speaking countries even as it has diminished among many of its

peers (Leigh F595).


12/47

In the late 1970s and 1980s, the market began creating much more significant pre-taxincome inequality everywhere. In Winner-Take-All Politics, Jacob Hacker and an Paul Pierson

argue this increased inequality by concerted, active resistance in other countries, but in the

United States these trends have preceded with little government interference (Hacker 52).

After a careful review of the evidence, it seems that while the global economy did undergo

profound shifts over the past forty years, those shifts alone do not account for the rise in income

inequality. So, while driven upwards by broader economic trends, the United States exceptional

income inequality is a result of domestic policy. Policymakers in the United States responded to

growing pre-tax income inequality in an aberrant way, lowering taxes on the richest instead of

raising them, and failing to meet increased economic hardship on the poorest Americans with

government redistribution of income. Hacker and Pierson place the blame for growing income

inequality squarely on the shoulders of the US political system, blaming drift. They define drift

as the systematic, prolonged failures of government to respond to the shifting realities of a

dynamic economy (Hacker 43). The United States had an anomalously high after tax income

inequality because it did not adjust its tax policies and social spending. The question then is why

the American political system alone failed to react.

Culture: Mass Attitudes, Misperceptions, and Voter Turnout

Conventional wisdom indicates that rising income inequality results from the American peoples

unique tolerance for income inequality. Differentiating mass attitudes from political institutions

is difficult because the nations political structures reflect and shape attitudes. Still, some, albeit

weak, evidence suggests that culture and income inequality could be related. In addition to the

12


13/47

difference between the English speaking countries and everyone else, there is a difference

between continental Europe, Nordic countries, and Anglo countries. In an article entitled

Inequality and Sociology, Lane Kenworthy examines how differently income inequality has

grown in the three different cultural groups. Although separated by geography, the UK, Canada

and the US all share a common cultural background. The difference between regions that share

cultural similarities, particularly Anglo countries and everyone else, suggests that culture plays a

role in income inequality.

The relationship between Nordic countries and low inequality similarly suggests that acultural correlation with income inequality. Even compared to continental Europe, Nordic

countries have low income inequality. Nordic countries are anomalous in almost every metric

related to income inequality: they have the highest levels of unionization, bargaining coverage,

left-wing parties, voter turnout, and, as a result, the highest income equality. The confluence of

all of these factors among a culturally similar region implies a relationship between culture and

inequality. Figure 6 clearly demonstrates that different regions, which share common cultural

backgrounds, have seen different trends in income inequality.

13


14/47

Figure 64:

Source: Kenworthy 591

One way to assess differences in culture is to assess differences in mass attitudesregarding redistribution. In a analysis based on International Social Survey Program (ISSP) data,

Ursula Dallinger finds that, outside of the United States, inequality tends to increase popular

demand for redistribution (Dallinger 333). Why has the US not seen an increase in support for

redistributionary policies along with an increase in income inequality? Comparing mass attitudes

14

4 The p90/p10 ratio is a ratio comparing the top 10% with the bottom 90%. In his analysis of the correlation between

Gini coefficient and different measurements of inequality, Andrew Leigh finds there to be a strong correlation

between a countrys p90/p10 ratio and Gini coefficient (Leigh F595).


15/47

and behaviors between comparable nations should elucidate how culture influences inequality.

US political culture could be at fault for increasing inequality for three reasons. Firstly,

Americans could be exceptionally opposed to redistribution, perhaps due to an exceptional faith

in economic mobility. Secondly, Americans could be exceptionally ill-informed as to the actual

income distribution, and therefore lack the motivation to demand redistribution or a change in

tax policy. Finally, declining voter turnout could reduce the demand for inequality reducing

measures by distorting the preferences of high-income citizens.

Are Americans particularly opposed to government redistribution? InAmericanExceptionalism:A Double-Edged Sword,published in 1997, Seymour Lipset represents the

conventional wisdom and argues that Americans are uniquely opposed to government

redistribution. Lipset affirms Tocquevilles claim that the United States is culturally exceptional

in affirming the individual over the collective. He differentiates between equality of opportunity

and equality of outcomes, arguing that Americans, when surveyed, choose the former. In Lipsets

analysis, Americans, instead of choosing to help the less affluent by means of state solutions

such as welfare, public housing, public employment, and medical care (Lipset 21), place undue

importance on individual solutions to societal ills. Lipset argues that this preference results from

a culture of anti-statism, the result of American deference to liberty throughout its history.

Similarly, in The American Ethos, Herbert McClosky and John Zaller reinforce Lipsetsdifferentiation between equality of outcomes and equality of opportunity. McClosky and Zaller

argue that American opinion has, from the early days of the Republic, regarded equality of

opportunity as a fundamental principal of the Republic (McClosky 88). Based upon

contemporary surveys of American opinion, however, they find that this principle applies more

to equality of opportunity than outcomes. McClosky and Zaller argue that the American public

15


16/47

changes it support for government pursuit of equality depending on whether they perceive that

pursuit as a remedy to a political or economic problem. Americans favor political intervention to

preserve equality, but oppose egalitarian intervention into the economy.

Some surveys on mass attitudes support the theory that Americans are less likely thencitizens of other countries to favor equality of outcomes. In one such survey, Alberto Alesina,

Rafael Di Tella and Robert MacCullough compared voter surveys on happiness and inequality,

broken down by levels of income.5 Their results indicate a divergence in American and

Europeans perspectives on inequality, at least in terms of how it effects societal wellbeing. They

found that, they find that left-wingers in Europe are more hurt by inequality than left-wingers in

the US. And the poor in Europe are more concerned with inequality than the poor in

America (Alesina 1). This data, measuring happiness, suggests that rising inequality has

affected attitudes of the poor in in Europe and the United differently. However, it does not depict

attitudes on redistribution. It also does not show whether the American public continues to

oppose equality of outcomes during a period of rising inequality.

While interesting, Lipsets conclusions, and Alesinas observations, are undermined bymore recent studies. In her analysis of ISSP data, Leslie McCall undermines Lipsets theory by

analyzing Americans understanding of the degree of income inequality over time. She argues

data from 1987 to 2000 shows that despite the prevailing image that Americans are tolerant of

inequality, well over half of the population thinks that inequality is too high, and more

importantly, this share rose significantly over the period of rising inequality (McCall 1) In their

study of attitudes towards income inequality, Lars Osberg and Timothy Smeeding also examine

16

5 Alesina and his coauthors compare data from the United States Social Survey from 1972-1997, the Euro-barometer

survey series from 1975-1992, and the changes in Gini coefficients over the years surveyed. The survey data from

both the European and American examples can be compared levels of happiness and Gini coefficients.


17/47

survey data collected by the ISSP and compare American and European public opinions.6 They

find little evidence for American exceptionalism in average attitudes (Osberg 451). Osberg and

Smeeding explain that their survey data differs from those of their predecessors because prior

comparativists measured attitudes about income inequality by emphasizing a presumed

American belief in the ideology of mobility and opportunity (Osberg 450). That presumption

led prior authors to conflate different questions on inequality.

Surveying inequality can range from surveying attitudes about the difference betweenmiddle and lower class, redistribution, what people should earn, and the maximum and minimum

people should earn. Osberg and Smeedings solution is to separate these questions out and survey

each facet of inequality. They argue that the differences that authors like Lipset identified

disappear in most measures of inequality. When asked whether income differences are too large,

a majority of respondents in all nations agreed, and the United States actually had a higher

percentage than the average indicating that they strongly agreed. On the basis of this more

refined measurement, they conclude that the the United States is nota clear outlier when mean

responses are compared across nations (Osberg 456).

Osberg and Smeeding also surveyed respondents on their perceptions of what peopleshould earn and what they believe people actually do earn. Osberg and Smeeding translated

ISSP responses to the Gini index. The ratio used in Figure 7 gauges the average respondents

belief in the fairness of their countrys economic system. Figure 7 shows that, in this measure,

the United States is again in the middle with an average level of should learn inequality at

about .35, very close to the European and all nations average of .34 (Osberg 461). Figure 7

shows a strong correlation between what respondents perceive people should and do earn, with

17

6 This data is also more interesting than that of Alesina and his coauthors because it draws from the same dataset and

the same survey questions for Europe and the US.


18/47

an R^2 = .78. This indicates that 78% of the change in the average respondents perception of

what the income distribution should be can be accounted by what they perceive to be the actual

income distribution. Further, Figure 7 indicates that the United States is not more tolerant of

income inequality than its peers. This seriously undermines Lipset and McClosky and Zallers

premise that Americans do not favor equality of outcomes.

Figure 7: Relationship Between Perceptions on Redistribution and Perceptions of the

Degree of Income Inequality

Source: Osberg 462

Osberg and Smeeding show that Americans are not less tolerant of governmentredistribution, but mass attitudes could still be culpable. A better theory implicating mass

18


19/47

behavior is that Americans misperceive the extent of income inequality and therefore do not

demand redistribution. Larry Bartels articulates this theory well, arguing that the lack of

egalitarian sentiment in the US is due to Americans misperceptions of the actual income

distribution in the United States. Bartels finds that when voters accurately perceive the degree of

income inequality, they adjust their attitudes on redistribution accordingly, as evident in the

section above. After examining survey data in the United States throughout the three decades that

income inequality has been increasing, Bartels argues that public perceptions of economic

inequality bear so little relationship to actual trends in inequality, that we must temper any

overly optimistic assessment of the extent to which ordinary people are aware of changes in the

relative fortunes of the rich and poor (Bartels 146).

Bartels demonstrates that, even during periods where income inequality had slowedsubstantially, the public would perceive vastly higher income inequality than the year before.

This leads Bartels to conclude that although Americans may express genuine allegiance to

egalitarian values, they are not sufficiently attuned to the political debate to see how those values

are implicated in major policy choices (Bartels 161). Bartels provides a plausible theory for

why Americans may have normal attitudes towards redistribution, yet still do not demand a

policy response to growing income inequality.

Similarly, the inherent structure of the American political system effects mass attitudes.Larry Bartels explains that, because there are only two parties, the cost of casting a

conservative vote is greater in the United States than systems with more moderate right-wing

alternatives (Bartels 95). Based on cross-national evidence from voting patterns in 20

democracies, Bartels finds that, in the United States, the gap in right-wing electoral support

between rich and poor voters was considerably larger (Bartels 95), than other countries. This

19


20/47

strongly suggests that the two-party system in the United States has exacerbated polarization

among voters, perhaps distorting their preference for redistribution.

The ISSP surveys apply to all Americans, but since the 1970s declining voter turnout haschanged the makeup of those Americans that vote, in a way that may make these prior studies of

mass attitudes irrelevant. That decline has disproportionately been among low-income voters

who would favor redistribution. Figure 8 shows that the United States low turnout is strongly

related to its high income inequality. Low-income groups traditionally turn out less. This is

problematic for supporters of greater redistribution because, as Vincent Mahler notes, if a larger

share of the electorate participates in national elections, the preferences of low-income groups

will be better reflected in public policies affecting earnings (Mahler 1041). This supports Nolan

McCartys case that in the United States it is about who is voting rather than voters attitudes. He

argues that while overall inequality has increased, the pressure for redistributive policies has

been sharply mitigated by the fact that the income inequality of voters did not

increase (McCarty 192). As low-income voters have stopped turning out to vote, support for

redistributionary policies has decreased.

Income inequality has also changed mass attitudes towards political parties and thepolitical process. Pontusson and Rueda find that increased income inequality decreases political

mobilization for left-wing parties and therefore empowers right-wing parties. In their estimation,

based upon the authors statistical analysis of advanced democracies, low-paid workers are stuck

in a vicious cycle in which

Increasing inequality makes their preferences for redistribution stronger, but

decreasing mobilization makes their demands less relevant to left parties, which in

turn makes these parties less redistributive when they get to power and so

20


21/47

inequality continues to grow. Decreasing mobilization, moreover, makes right

parties more likely to respond to inequality in accordance with the preferences of

their core constituencies (that is, by opposing redistribution as inequality rises)

(Pontusson 347).

Income inequality decreases support for redistributive policies among the right and left alike as

mobilization decreases, when poorer voters stop turning out. So, despite increasing demand

among the base of left-wing parties for redistributive policies, declining turnout decreases their

influence. So, increased income inequality has changed the composition of parties and reduced

voter mobilization. This indicates that the conservative American electorate could be explained

by decreasing turnout in elections. Even if American policymaking runs contrary to the economic

interests of all constituents, it should better reflect the interests of those Americans that actually

show up and vote.

Figure 8: Income Inequality by Turnout

Source: Mahler 131

21


22/47

The United States has an aberrantly low level of voter turnout and has seen a decreasesince 1960. Figure 9 shows that countries with higher voter turnout tend to be countries with

lower income inequality, including Sweden, Germany, the Netherlands and Norway. In their

research on the impact of declining voter turnout on leftwing parties, comparativists Jonas

Pontusson and David Rueda found that even a very big increase in income inequality has no

effect on the position of Left parties when low-income mobilization is limited (Pontusson 695).

Further, based on cross-national evidence, they find that declining voter turnout has

disproportionately been among low-income voters who would ostensibly favor redistribution.

Figure 9 demonstrates that the decline in voter turnout has occurred over the same period as the

rise in income inequality. Further, Figure 9 indicates that countries with lower Gini coefficients

have not seen the same decrease in turnout as the United States. Turnout then evidences why

increasing income inequality has not led to an increase in support for redistributionary policies or

the enactment of those policies.

Figure 9: Voter Turnout from 1960-2000

Source: Pontusson 690

22


23/47

Why do low income voters abstain from voting? In their study on Inequality, Ideologyand Legitimacy in Europe, Christopher Anderson and Matthew Singer argue that not only does

income inequality decrease voting turnout, it does so specifically to left-wing voters. Based on

their cross national comparison of opinion polling and turnout data, the authors find that, to

leftwing voters, inequality breeds discontent with the political regime (Anderson 583). This

discontent is muted among those on the right (Anderson 564). Finally, Anderson and Singer

find that discontent, or lack of trust, is highly correlated with declining voter turnout. This

indicates that declining voter turnout is related to income inequality, either as a cause or an

effect.

Turnout figures indicate that, even if mass attitude would support greaterredistributionary policies, voter turnout would mitigate those demands. Over the same period that

income inequality has risen, voter turnout has decreased. Voter turnout has particularly decreased

among poorer Americans. This trend has made government more responsive to the policy

preferences of richer Americans as they make up an increasing percentage of actual voters. In

other words, the lower voter turnout and lack of influence of voters on government is more

important than citizens acquiescence to an unequal income distribution. Reduced support for

more left-wing parties is the result of declining voter interest rather than a change in voters

opinions.

The Decline of Unions and Collective Bargaining Coverage

23


24/47

Declining levels of unionized workers helps explain diminishing voter turnout among low-

income voters. Reinforcing Hacker and Piersons theory that the decline of unions has led to a

loss of a political rival to other special interests, Benjamin Radcliff and Patricia Davis find that

declining unions has limited low-income voter turnout. Based on a study comparing rates of

unionization among advanced democracies, they conclude that unions indirectly encourage

turnout through their role as advocates for the interests of low- and middle-income citizens and

serve as the most important political advocate for working people (Radcliffe 133). Looking at

the change in union density over time, they find that, after examining numerous other potential

causes of declining turnout, that the dramatic post-1960 decline of organized labor may also be

an important cause of demobilization (Radcliffe 137). In summary, the loss of unions over the

past forty years has made the voices of low-income voters even weaker.

Unions represent the interests of working class Americans. Paul Krugman argues that,during the Great Compression, unions were the most important force in creating income equality.

Unlike Hacker and Pierson, who argue that unions serve as an important counterbalance to

business interests in affecting policy, Krugman sees unions as equalizing income between

management and workers as well as raising the wages of blue-collar workers. In the 1950s, 40

percent of American private sector workers were unionized and 5 percent of public sector

workers (Freeman 16). In the 1970s, Krugman argues that the collapse of the U.S. union

movement that took place has no counterpart in any other Western nation (Krugman 151).

Union density has reached a historic low in the United States, driven by political hostility to

collective bargaining. Their decline along with inequalitys rise implies a relationship.

The evidence of the effect of union density on income inequality is more mixed thanKrugman suggests. Figure 10 shows that unionization rates have gone down gradually among

24


25/47

most countries in the sample, with the exception of Nordic countries, where theyve continued to

rise, and the UK, where they collapsed precipitously. After Margaret Thatcher reduced

unionization in the UK, the union density there declined from 50% to 38% almost immediately.

Figure 6 illustrates the wildly divergent levels of union density among long-standing

democracies. The highly different rates of union density make comparison between countries

difficult, though it does indicate that globalization does not necessary lead to collapse of unions.

Nations with much higher unionization like Sweden, Finland, Norway and Belgium have high

per capita incomes and are thoroughly integrated into the global economy.

Several cases also challenge the concept that union density alone significantly contributesto income inequality. Although Canada and the United States both have relatively high Gini

coefficients, Stepan and Linz note that between 1960 and 2005, the union share of wage and

salary workers in the United States went from 30.4% to 12.5%, whereas in Canada, union share

remained virtually the same, shifting only from 32.3% to 32.0% (Stepan 849). While these

examples do not disprove the impact of unions on income inequality, they challenge a direct

causal connection between a changing percentage of unionized workers and income inequality.

25


26/47

Source: Viser

However, if one looks at the number of workers allowed to collectively bargain, asopposed to mere union density, one finds a much stronger relationship with income inequality.

This relationship is evident in Figure 11, showing the percentage of workers covered by a

collective bargaining agreement. Collective bargaining is a better measure of the strength of

unions then union density. For example, economist Richard Freeman demonstrates how France,

which has the lowest union density in Europe, still has strong unions because labor law requires

that most workplaces have designed union representatives with legal responsibilities (Freeman

18). This facet of union strength is conveyed in the percentage of workers covered by a collective

bargaining contract, but not in union density. Lew Daly, a senior fellow at the Demos institute,

observes that collective bargaining declined from a high of 29% in 1950 to a low of 15 percent in

1999 while in most European countries there was little or no change (Daly 1). As a result,

0%

23%

45%

68%

90%

1970 1980 1990 2000

Figure 10: Union Density, 1970-2000

PercentofWorkersBelongingtoaUnion

France United States Germany

United Kingdom Finland Sweden

26


27/47

Freeman observes that today the United States represents the decentralized extreme in wage

setting (Freeman 19).

Collective bargaining coverage allows us to see how unions influence a countrys incomedistribution and politics. The nations in Figure 11 with the highest percentage of workers covered

by collective bargaining contracts corresponds with the countries with the lowest Gini

coefficients. The correlation between the percentage of workers covered by collective bargaining

agreements in a country and a countrys Gini coefficient is .67, indicating that collective

bargaining explains 67% of the increase in a countrys Gini coefficient. This relationship is

robust and statistically significant.7 This provides strong evidence that collective bargaining

coverage and Gini coefficients are correlated.

Lane Kenworthy goes so far as to argue that were the US to allow as many workers tocollectively bargain, there might well have been little or no rise in earnings inequality to

explain (Kenworthy 595). His evidence for the importance of unions contrasts the case of the

United States and the United Kingdom. Kenworthy argues that the uptick in income inequality in

the UK is due to declining unions because: the United Kingdom was the sole country among

these 10 in which collective bargaining coverage changed markedly during the 1980s and 1990s.

It declined from 70% in 1978 to 47% in 1990 (Kenworthy 595). Collective bargaining coverage

accounts for how unions raise wages for the average worker, even among workers not covered

by a union contract. This, in turn, creates a more equal workforce.

A good example of the influence of collective bargaining coverage, rather than uniondensity, is in Germany. While only 30% of German workers are actually part of a union, the

contracts that unions negotiate covers almost 80% of workers. This disparity indicates that

27

7 The p-value, or probability that this relationship is nonexistent, is .016. This is below the threshold for statistical

significance.


28/47

unions can have an outsized political result, influencing the labor market well beyond the actual

workers in the union. The cross-national evidence strongly suggests that their decline in the

United States has contributed to widening income inequality.

Source: OECD Data on Gini Coefficients and Bargaining Coverage

The example of the UK, as well as the relationship between countries that have highpercentages of workers covered by collective bargaining agreements and low Gini coefficients,

supports Krugmans theory that the decline of unions has contributed to income inequality. The

relationship between the percentage of workers covered by collective bargaining agreements and

income equality supports both Hacker and Pierson and Krugmans theories about why unions are

important. Hacker and Pierson see the loss of unions as the loss of the most important lobbying

0.15

0.21

0.28

0.34

0.40

US Japan Canada UK Sweden Finland0.15

0.35

0.55

0.75

0.95

Figure 11: Collective Bargaining Coverage and Income Inequality

GiniCoeff

icient,1995

PercentageCoveredbyCollectiveBargainingAgreement

Gini Coefficient, 90s Bargaining Coverage, 90s

28


29/47

force for the middle-class. Krugman argues that if there was a single reason blue-collar workers

did so much better in the fifties than they had in the twenties, it was the rise of

unions (Krugman 49). Krugman sees their most important influence in equalizing wages, which

is measured by collective bargaining coverage.

The rise of unions accounted for the leveling of incomes after WWII, and the evidencesuggests that their declining influence has contributed to widening income inequality during the

Great Divergence. The low percentage of workers covered by collective bargaining agreements

shows the limited influence that unions have in the United States. The decline of unions helps to

explain, or serve as a proxy for, the diminishing importance of the demands of the working class,

including declining voter turnout. The United States may have low levels of unionization, but the

decline in unions has been widespread across Europe as well, without an accompanied rise in

inequality. The various exceptions and inconclusive statistical evidence prevents them from

satisfactorily explaining the United States income inequality.

Institutional Explanations: The Number of Veto Players and the Unrepresentative Senate

If mass attitudes and unions cannot fully explain the rise in income inequality, perhaps the

system itself has made reducing inequality more difficult. Hacker and Pierson argue that our

political system is prone to stasis and the protection of entrenched interests, which, although

partly a result of our lack of a parliamentary system, has been exacerbated over the past thirty

years by the rise of an astonishingly dynamic capitalism (Hacker 297). In their study of social

expenditures and income inequality between several different countries, Jonathan Schwabish and

his colleagues dispute this theory. They find that the US is not anomalous in terms of mass

attitudes concerning redistribution, but rather in terms of the dissimilarities in institutions that

29


30/47

represent social and economic rights in the political arena which determines redistributive

government spending (Schwabish 35).

Alfred Stepan and Juan Linz test the hypothesis by proposing that income inequality isgreater in the United States because of the number of its veto players. They rely upon the work of

other comparativists in defining each electorally generated veto points as individual or

collective actor whose agreement is required for a policy decision (Stepan and Linz 844). So,

within the United States, the veto players would be the Senate, House of Representatives,

President, and the ability of the public to ratify a constitutional amendment. Finding the

relationship between the intrinsic stasis of the US system and comparable states is difficult.,

Stepan and Linz assess the state of the legislature in the United States, finding that the United

States is politically exceptional in the high number of electorally based veto players who

potentially can block social change, by blocking key bills or amendments (Stepan and Linz

844). Electorally generated veto players are anti-majoritarian features that constrain the ability of

legislatures to enact legislation.

The United States is an outlier in the number of its veto players. Stepan and Linz theorizethat there is a positive correlation between the number of veto players in a political system and

income inequality. This relationship is graphed in Figure 12. Stepan and Linz theory that the

more veto players there are in a political system the more difficult it is to construct a win-set to

alter the political status quo (Stepan and Linz 844). The number of veto players indicate that

there are many places in the United States political system where important legislation could be

blocked by minority interests. This provides an interesting piece of evidence contextualizing

Hacker and Piersons drift theory. With many places in which legislation could be stopped, the

structure of our government makes passing redistributionary policies more difficult than in other

30


31/47

longstanding democracies. Even if there were a relationship between Gini coefficient and veto

players, Stepan and Linzs focus on veto players does not address the puzzle of why income

inequalitys rise has been so pronounced over the past thirty years, because the basic structure of

the United States political system has not changed.

The United States exceptional number of veto players do not explain its exceptionalincome inequality. After running a regression analysis between a countrys Gini index and its

number of veto players, shown in Figure 12, there is a correlation of 62%. This correlation barely

meets the threshold for statistical significance.8 In addition, problematically, when the United

States is removed from the data set, the correlation between Gini coefficient and veto players is

no longer statistically significant. That suggests that the United States is an outlier and that the

relationship between Gini coefficient and veto players is not a satisfactory explanation for the

United States income inequality.

Source: Stepan and Linz 844

0.15

0.21

0.28

0.34

0.40

US Italy Can. Germ. Bel. UK Swe.0

1

2

3

4

Figure 12: Relationship between Veto Players and Gini Coefficient

GiniCoefficient

NumberofVetoPlayers

Gini Coefficient, 2005 Veto Players

31

8 The p-Value, which indicates the probability that the regression is the result of random events, of the regression

between Gini coefficients and veto players is .042. This indicates that the null hypothesis, or the probability of there

being no relationship, is 4.2%. That is not very compelling evidence, but it is below the common threshold of .05.

That indicates that the regression .62, while no robust, is statistically significant because we can reject the null

hypothesis.


32/47

The number of veto players alone cannot explain income inequality, but perhaps thenature of the veto players themselves are to blame. One veto player that has changed since the

Great Compression is the Senate. Examining the United States exceptionally unequal upper

body better shows the substantial institutional barriers to dramatic legislation than the veto

players. In Figure 13, the Senates uniquely unrepresentative structure is contrasted with that of

eight other long-standing federal democracies. Figure 13 shows the difference between the best

represented federal unit with the worst in and its relationship with inequality. Not only is the

Senate the least representative legislature surveyed, higher Gini coefficients are correlated to low

levels of representation in legislative bodies. A linear regression analysis of a nations Gini

coefficient and its ratio of best to worst represented federal unit, suggests that there is a robust

relationship between a lack of representation and income inequality. The r-value for the

correlation between a nations lack of representation in its upper body and Gini coefficient is .72,

indicating that a countrys lack of representation explains 72% of the increase in its Gini

coefficient.9

Although the Senates lack of representation is not entirely new, it has escalated alongwith population growth. In addition, Stepan and Linz argue that the Senate in addition to being

the least representative, has the most power to induce inequality because of its ability to stop

constitutional amendments that would make the political structures more representative and its

ability to make federal appointments.

32

9 Given the low number of nations in this sample, 8, it is surprising that this regression model has a p-value of .007,

well below the .05 needed for statistical significance. This indicates that there is a strong correlation that is not the

result of a random distribution.


33/47

Source: Stepan and Linz 845

The Senate, perhaps the most influential veto player, has gotten more powerful due to therise of the filibuster. Stepan and Linz argue that in the latter half of the 20th century, the United

States has undergone a process of de-democratization. Like Hacker and Pierson, they write that

the rise of the filibuster has given the Senate the ability to stall legislation desired by even the

majority of Senators, by making it difficult to pass legislation without the necessary 60 votes to

break invoke cloture. This change makes the United States unique among its peers as no other

democratic upper house in the world allows such unconstitutionally entrenched, minority-

induced vetos (Stepan 847). Figure 14 documents the number of times that cloture has been

invoked to formally end a filibuster (Klein 1). From Figure 14, it is evident that cloture

invocations have risen precipitously since 1990. Although this is an effective measure of the

increasing usage of the filibuster, Figure 14 also underestimates its effects, because it only

documents cloture votes. In so doing, it only accounts for the times in which a majority attempts

0

0.1

0.2

0.3

0.4

US Switz. Can Aus. Germ. Spain Bel.2

19

36

53

70

Figure 13: Relationship Between Represenation and Inequality

GiniCoefficient

RatioofBestRepresentedtoWorstDistrict

Gini Index Ratio of Representation

33


34/47

to break a filibuster by filing for cloture. The filibuster's normalization has made the lack of

representation in the Senate even more severe over the past forty years. Since the 1980s,

Senators have gained an powerful tool to prevent the passage of legislation.

Still, the United States has had its four electorally generated veto players since itsinception and the Senate has had distinctly undemocratic features throughout the 20th century,

yet during the Great Compression its income inequality was comparable or even better than other

advanced industrial democracies. This indicates that, after a careful examination of various

institutional explanations, American institutions alone do not explain the growth in income

inequality. The increasing usage of the filibuster is a problem, but it is not the primary culprit.

Figure 14

34


35/47

Source: Klein 1

Elite Politics: Polarization

Despite the filibusters impact in making the Senate even less representative, it alone could not

create the growing income disparity in the United. Indeed, mass attitudes, unions, and

institutions do not seem to account for the extremity of the United States rise in income

inequality. The most compelling explanation, the unrepresentative Senate and the filibuster,

points to another cause. The United States unequal political institutions empower the

preferences of elites, whether political interest groups or politicians. In Elites and the Idea of

Equality (1987), Sidney Verba and his coauthors analyze the difference between political elites in

the US, Sweden and Japan. They find that almost all political groups oppose government

redistribution, finding that no American group occupies the position of left groups in Japan or

Sweden (Verba 85). Although Verba and his coauthors find that some American elites favor

35


36/47

equality of opportunity, they find that all American leadership groups-- including those farthest

to the left-- [reject] radical redistribution and any policy directed toward equality of

results (Verba 82). Over the past forty years the influence of political elites have been magnified

by the more frequent use of the filibuster, decreasing voter turnout and confused mass attitudes

on the degree of inequality. Verba and his coauthors research suggests that elite preferences are

to blame for the United States exceptional income inequality.

Although there is limited evidence that polarization has increased in the Americanpopulace, there is much stronger evidence for it being an elite phenomenon. In an article entitled

Inequality as the Source of Political Polarization, Jonas Pontusson and David Rueda analyzed

the ideological makeup of ruling parties from 1980 to 2000 and how that ideology is associated

with wage inequality. Most importantly, they indicate that, even absent other variables like

unionization and turnout more inequality is strongly associated with more partisan

polarization (Pontusson 4) in the ruling party. Further, also they find that although wage

inequality tends to be associated with left-skewed polarization (Pontusson 346) among political

elites, it also diminishes the likelihood that constituents will actually vote. The increase in

political polarization then worsens inequality because even as it should make the demand for

redistribution stronger among the populace, it makes those preferences felt less by political

parties.

In Unequal Democracy, Larry Bartels argues that once elected political elites in theUnited States have considerable ideological latitude. Bartels finds that massive differences in

the preferences of middle- and upper income constituents had less effect on senators policy

choices than their own partisan ideologies (Bartels 287). Bartels finds this influence particularly

strong in areas where voters have difficulty understanding, like tax policy. Even if the electorate

36


37/47

is getting more polarized on its own, political elites have dictated the actual policymaking.

Morris Fiorina, in an article surveying the state of literature on mass polarization, finds that the

most direct evidence-- citizens positions on public policy issues-- show little or no indication of

mass polarization over the past two to three decades (Fiorina 20).

Figure 15 also shows that the rise of income inequality has directly coincided with anincrease in elite polarization, with both increasing dramatically after 1978. Nolan McCarty, Keith

Poole and Howard Rosenthal effectively explain why elites, rather than voters, have been to

blame. McCarty, Poole and Rosenthal find, like Fiorina, that voters have not shown any

demonstrable increase in polarization. Instead, parties in the United States are particularly

undisciplined, they argue, so politicians have had increasing motivation to defect from a party

platform at the national median toward the position of the district median. (McCarty 192). This

produces parties that are increasingly polarized from each other, as they do not reflect a national

consensus. Similarly, politicians respond to moderate voters by asserting their own ideologies.

McCarty and his coauthors summarize that if there is sufficient uncertainty about voter

preferences, the distinct ideological preferences of politicians and the activist basses of each

party will produce divergence in platforms (McCarty 192). Figure 15 shows the correspondence

between the rise in income inequality and an increasingly polarized American political elite.

Figure 15

37


38/47

Source: McCarty 2

Polarization is among Hacker and Piersons chief culprits for drift as well. Over the pastforty years the House and Senates polarization has risen dramatically and for the past thirty that

increase has been accompanied by corresponding increase in income inequality. The increasingly

polarization of political elites effects our system more profoundly then others. They find that an

increase in distortions of the other partys proposals and an increased recalcitrance in

negotiation (McCarty 194). Such recalcitrance is particularly damaging in the American case

due to its particular institutional structure. Compounding the problem, minorities in the Senate

have slowly replaced the need for a simple majority to pass bills to requiring sixty votes by

exploiting the threat of a filibuster. Hacker and Pierson concludes that, as a result, in a highly

polarized political system, the rule of sixty is a recipe for frozen coffee (Hacker and Pierson

270).

38


39/47

Cross-national evidence also suggests that polarization mitigates the the politicalresponse to inequality. Henning Finseraas, analyzing 13 countries, examines whether rising

income inequality increases the demand for redistribution after considering political polarization.

To account influence of unions, Finseraas controls for union density in his statistical analysis. 10

Finseraas uses survey data collected by World Values Survey, he finds that the interaction

between inequality and polarization is negative in all models (Finseraas 288). In other words, he

finds that an increase in political polarization decreases the degree to which voters demand

redistributionary policies. Finseraas finds that, although the electoral system is important in

determining the demand for redistribution, that polarization has an independent and robust effect.

In addition, Finseraas evidence suggests that an increase in voter turnout does not increase the

redistribution (Finseraas 300) when coupled with this level of polarization.11 That is because

party polarization, in and of itself, is so strongly correlated with a weaker relationship between

income and leftism (Finseraas 283). Party politics, in other words, determine a governments

level of redistribution.

Conclusion

39

10 Finseraas cites the superiority of union density over collective bargaining coverage because it varies over time,

so that its effect can be estimated in a model with country-fixed effects (Finseraas 295).

11 Finseraas attributes this difference to his observance, unlike Kenworthy and Mahler, of country-fixed effects. This

means that Finseraas, in his statistical analysis, assumes that there will be a connection between polarization and

inequality, and so only considers the effects variables that have a demonstrable, fixed effect on the trend.


40/47

Tocqueville feared the emergence of a perfectly equal, democratic society in which all men

would remain paralysed by a state of common ignorance and equal servitude (Tocqueville 37).

Tocqueville predicted that the naturally unequal faculties of men would be equalized by

government. The concept of American exceptionalism originated from Tocquevilles description

of the unique American character. Thankfully, he wrote, the United States has the exceptional

belief in liberty to temper this democratic impulse. Today, surveys indicate that American mass

attitudes are unexceptional. As outcomes have become tremendously unequal and income

inequality has risen precipitously, American mass attitude has naturally favored increasing

distribution. Americans have not chosen to be exceptional unequal. It is the result of a myriad

political trends.

What has changed since the late 1970s? Compared with other long-standing democracies,in terms of its political institutions, the United States has a unique structure. It has the most of

electorally generated veto players and the most unrepresentative upper body in its legislature (the

Senate). This alone does not explain why income inequalitys rise. Although American

institutions have become less democratic through the increased use of the filibuster and

decreasingly representative senate, they have not fundamentally changed since Tocqueville

visited in 1831. The filibuster may contribute to other causes, but reforming the filibuster would

be insufficient to make the income distribution more equitable.

The declining strength of unions provides a better explanation. Unions havefundamentally changed since over the course of the 20th century. Since the rise of unions under

President Roosevelt significantly contributed to the narrowing of incomes after WWII, one

would expect their decline to be connected to the widening income distribution. Today, the US

has the lowest union density, other than France, and the lowest percentage of workers covered by

40


41/47

collective bargaining agreements, a good proxy for the unions political strength. The robust

correlation between the amount of workers covered by collective bargaining agreements and the

Gini coefficient strongly suggests that the decline of unions has played a significant role.

Finally, the striking increase in political polarization among the elites and parties in the United

States has made wide-reaching national legislation more difficult to pass.

The progress of equality may not be inevitable, but the rise of pre-tax income inequalityshould have prompted a change in US tax policy. The dramatic rise in income inequality over the

past forty years suggests that Tocquevilles fears about the demise of the aristocracy were

exaggerated. A new aristocracy of decreasingly representative political elites and parties

dominate the American political system. The checks and balances built into the American

Republic, which Tocqueville lauded as ensurers of liberty, have enabled these elites to stymy any

broad redistributionary measures. Among Tocquevilles most profound fears was that democratic

institutions would enable a tyranny of the majority. Today, the tyranny of the majority is far from

the problem in the United States today. Instead, a tyranny of minority, of political elites,

empowered by the anti-majoritarian features of American institutions, have mitigated those

demands. The divergence in political parties and elites has enabled the rise of income inequality

by making redistribution, or indeed any broad policy initiatives, more difficult to accomplish.

Hacker and Pierson aptly title this stasis drift. Their analysis of drift as the culprit for American

inequality is born out by cross-national evidence. In addition to making the passage

redistributionary policies less likely, elite polarization reduces turnout, which disproportionately

effects low-income voters.

American policymakers have begun to recognize the problem. In his most recent State ofthe Union address, President Obama identified economic inequality as the defining issue for our

41


42/47

time, and argued we can either settle for a country where a shrinking number of people do

really well, while a growing number of Americans barely get by (Wallsten 1). To reintroduce

fairness into the American economic system, Obama called for an increase of taxes on the very

wealthy.

This is a start, as mass attitudes opposition to redistribution hinges on the averageconstituents misapprehension of the extent of income inequality. Any campaign to reintroduce

income inequality into the national conversation is progress, because income inequality is a

political issue and it must be addressed with political reform. There is much we can learn from

the experiences and policies of comparable nations, such as the importance of unions,

unrepresentative institutions and polarization in contributing to inequalitys rise. The United

States peers have avoided a divergence, with most showing little to no increase in income

inequality over the past forty years. Increased income inequality should make the US

government increase its income redistribution, but its uniquely unrepresentative Senate, coupled

with the wide latitude afforded legislators, declining voter turnout, and declining unionization

has mitigated those egalitarian demands.

As long as the United States remains so tremendously unequal, its motto E pluribusunum, will not be fulfilled. Income inequality has profound societal ramifications. In The Spirit

Level: Why equal societies almost always do better, Richard Wilkinson and Kate Pickett provide

evidence that unequal societies have worse outcomes in health, social trust, and quality of life.

They argue that, based on their statistical analysis, reducing inequality is the best way of

improving the quality of the social environment, and so the real quality of life, for all of

us (Wilkinson and Pickett 29). Even if their claim is too extreme, the correlations between

42


43/47

income inequality and a whole host of societal ills is difficult to ignore. A successful society is

predicated upon an equitable division of income.

43


44/47

Bibliography

Atkinson, A B, and Thomas Piketty. Top Incomes: A Global Perspective. Oxford: OUP Oxford, 2010.

Alesina, A, Tella R. Di, and R MacCulloch. "Inequality and Happiness: Are Europeans and Americans Different?"Journal of Public Economics. 88 (2004): 2009-2042.American, Political S. A. T. F. "American Democracy in an Age of Rising Inequality." Perspectives on Politics. 2.4 (2004): 651-666.Anderson, Christopher, and Matthew Singer. 2008. "The Sensitive Left and the Impervious Right. Comparative Political Studies. 41 (4-5): 4-5.Bartels, Larry M. Unequal democracy: the political economy of the new gilded age. New York:Russell Sage Foundation, 2008.Beramendi, Pablo, and Christopher Anderson. Democracy, Inequality, and Representation: A

Comparative

Perspective. New York, N.Y: Russell Sage Foundation, 2008.

Bernasek, Anna. "Income Inequality, and Its Cost ." 25 June 2006. The New York Times. 2 February 2012 .Bradley, David et all. "Distribution and Redistribution in Postindustrial Democracies." World Politics55.2 (2003): 193-228.

Dallinger, Ursula. "Public Support for Redistribution: What Explains Cross-National Differences?" Journal of European Social Policy. 20.4 (2010): 333-349Daly, Lew. The Church of Labor.Democracy: A Journal of Ideas. 22. (2011). http://

www.democracyjournal.org/22/the-church-of-labor.php?page=all (Accessed on 4/30/2012).

Fiorina, M.P, and S.J Abrams. "Political Polarization in the American Public." Annual Review of Political

Science. 11 (2008): 563-588.

Finseraas, Henning. "What If Robin Hood Is a Social Conservative? How the Political Response to Increasing Inequality Depends on Party Polarization." Socio-economic Review. 8.2 (2009):283-306.

Fortin, Nicole M, and Thomas Lemieux. "Institutional Changes and Rising Wage Inequality: Is There a Linkage?" Journal of Economic Perspectives. 11.2 (1997): 75-96. Print.Freeman, Richard B. Working Under Different Rules. New York: Russell Sage Foundation, 1994.

Glazer, N. "Why Americans Don't Care About Income Inequality." Irish Journal of Sociology. 14.1 (2005): 5-12.Hacker, Jacob S. and Paul Pierson. Winner-Take-All Politics: How Washington Made the Rich Richer-- And Turned Its Back on the Middle Class. New York: Simon & Shuster Paperbacks, 2010.

44
http://www.democracyjournal.org/22/the-church-of-labor.php?page=allhttp://www.democracyjournal.org/22/the-church-of-labor.php?page=allhttp://www.nytimes.com/2006/06/25/business/yourmoney/25view.htmlhttp://www.nytimes.com/2006/06/25/business/yourmoney/25view.htmlhttp://www.nytimes.com/2006/06/25/business/yourmoney/25view.htmlhttp://www.nytimes.com/2006/06/25/business/yourmoney/25view.htmlhttp://www.democracyjournal.org/22/the-church-of-labor.php?page=allhttp://www.democracyjournal.org/22/the-church-of-labor.php?page=allhttp://www.democracyjournal.org/22/the-church-of-labor.php?page=allhttp://www.democracyjournal.org/22/the-church-of-labor.php?page=allhttp://www.nytimes.com/2006/06/25/business/yourmoney/25view.htmlhttp://www.nytimes.com/2006/06/25/business/yourmoney/25view.html


45/47

Iversen, Torben and David Soskice. "Electoral Institutions and the Politics of Coalitions: Why Some Democracies Redistribute More Than Others." American Political Science Review. 100.2 (2006): 165-181.Leonhardt, David. Coming Soon: Taxmaggeddon. The New York Times. http://www.nytimes.com/

2012/04/15/sunday-review/coming-soon-taxmageddon.html?pagewanted=all, 2012. (Accessed: 4/30/2012)Kenworthy, Lane. "Inequality and Sociology."American Behavioral Scientist. 50.5 (2007): 584-602.Kim, HeeMin, and Richard C. Fording. "Voter Ideology in Western Democracies: an Update." European Journal of Political Research. 42.1 (2003): 95-105.Klein, Ezra. Breaking the Filibuster in One Graph The Washington Post.http://

voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.html, 2010. (Accessed: 4/30/2012).Krugman, Paul. The Conscience of a Liberal. New York: W.W. Norton & Company, Inc., 2007.

Leigh, Andrew. "How Closely Do Top Income Shares Track Other Measures of Inequality." The Economic Journal 117 (2007): F619-F633.Lijphart, Arend. Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries. New Haven: Yale University Press, 1999.Linz, Juan and Alfred Stepan. Comparative Perspectives on Inequality and the Quality of Democracy in the United States. Perspectives on Politics, 9 (2011)., pp 841-856 Lipset, Seymour M. American Exceptionalism: A Double-Edged Sword. New York: W.W. Norton, 1996.

Mahler, Vincent A. "Economic Globalization, Domestic Politics, and Income Inequality in the Developed Countries: a Cross-National Study." Comparative Political Studies. 37.9 (2004): 1025-1053. -- 2002. "Exploring the Subnational Dimension of Income Inequality: An Analysis of the Relationship between Inequality and Electoral Turnout in the Developed Countries". International Studies Quarterly. 46 (1): 117-142. --2008. "Electoral turnout and income redistribution by the state: A cross-national analysis of the developed democracies.European Journal of Political Research. 47 (2): 161-183.Maddocks, Emily. "The Spirit Level: Why More Equal Societies Almost Always Do Better

(review)." Canadian Public Policy. 36.1 (2010): 133-135.McCall, Leslie. "Increasing Inequality in United States: Trends, Problems, and Prospects."Economic and Political Weekly. 35.22 (2000) -- "Increasing Inequality in United States: Trends, Problems, and Prospects." Economic and Political Weekly. 35.22 (2000).

45
http://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://www.nytimes.com/2012/04/15/sunday-review/coming-soon-taxmageddon.html?pagewanted=allhttp://www.nytimes.com/2012/04/15/sunday-review/coming-soon-taxmageddon.html?pagewanted=allhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://voices.washingtonpost.com/ezra-klein/2010/12/breaking_the_filibuster_in_one.htmlhttp://www.nytimes.com/2012/04/15/sunday-review/coming-soon-taxmageddon.html?pagewanted=allhttp://www.nytimes.com/2012/04/15/sunday-review/coming-soon-taxmageddon.html?pagewanted=allhttp://www.nytimes.com/2012/04/15/sunday-review/coming-soon-taxmageddon.html?pagewanted=allhttp://www.nytimes.com/2012/04/15/sunday-review/coming-soon-taxmageddon.html?pagewanted=all


46/47

McCarty, Nolan M., Poole, Keith T. and Rosenthal, Howard, Political Polarization and Income Inequality (January 27, 2003). Available at SSRN: http://ssrn.com/abstract=1154098 or http://dx.doi.org/10.2139/ ssrn.1154098 -- Polarized America: The Dance of Ideology and Unequal Riches. Cambridge, Mass: MIT Press, 2008McClosky, Herbert, and John Zaller. The American Ethos: Public Attitudes Toward Capitalism and

Democracy. Cambridge, Mass: Harvard University Press, 1984.

Noah, Timothy, "The Great Divergence," 30 September 2010, Slate, 23 January 2012 . -- "White House: Here's Why You Have To Care About Inequality ." 13 January 2012. The NewRepublic. 2 February 2012 .

OECD (2008), Social Expenditures aggregates, OECD Social Expenditure Statistics (database). doi: 10.1787/000530172303 http://dx.doi.org/10.1787/000530172303 (Accessed on 4/30/2012).Osberg, L, and T Smeeding. ""fair" Inequality? Attitudes Toward Pay Differentials: the United States in Comparative Perspective." American Sociological Review. 71.3 (2006): 450-473.Piketty, Thomas and Emmanuel Saez. "Income Inequality in the United States, 1913-1998." Quarterly Journal of Economics 118.1 (2003): 1-39.Pontusson, Jonas and David Rueda. The Politics of Inequality: Voter Mobilization and Left Parties in Advanced Industrial States. Comparative Political Studies. 43. (2010): 675-705.Radcliff, Benjamin, and Patricia Davis. 2000. "Labor Organization and Electoral Participation in Industrial Democracies.American Journal of Political Science. 44 (1): 132-141.Schwabish, Jonathan A, Lars Osberg, and Timothy M. Smeeding. Income Distribution and Social Expenditures: A Cross-National Perspective. Syracuse, N.Y: Maxwell School of Citizenship and Public Affairs, Syracuse University, 2003.Tocqueville, Alexis , Phillips Bradley, Henry Reeve, and Francis Bowen. Democracy in America. New

York: A. A. Knopf, 1945.

Verba, Sidney. Elites and the Idea of Equality: A Comparison of Japan, Sweden, and the United States. Cambridge, Mass: Harvard University Press, 1987. Print.Verba, Sidney, and Gary R. Orren. Equality in America: The View from the Top. Cambridge, Mass: Harvard University Press, 1985.Visser, Jelle. "Changes in State laws during 2005 - Union membership statistics in 24 countries - Analysis of "adjusted"union membership data provides explanatory factors for the differences and trends in unionization."Monthly labor review 129.1 (2006): 38.Wallsten, Peter. "In State of the Union address, Obama puts focus on economic inequality ." 24 January 2012. The Washington Post. 2 February 2012

46
http://www.tnr.com/blog/timothy-noah/99651/white-house-heres-why-you-have-care-about-inequalityhttp://img.slate.com/media/3/100914_NoahT_GreatDivergence.pdfhttp://dx.doi.org/10.2139/ssrn.1154098http://ssrn.com/abstract=1154098http://dx.doi.org/10.1787/000530172303http://dx.doi.org/10.1787/000530172303http://www.tnr.com/blog/timothy-noah/99651/white-house-heres-why-you-have-care-about-inequalityhttp://www.tnr.com/blog/timothy-noah/99651/white-house-heres-why-you-have-care-about-inequalityhttp://www.tnr.com/blog/timothy-noah/99651/white-house-heres-why-you-have-care-about-inequalityhttp://www.tnr.com/blog/timothy-noah/99651/white-house-heres-why-you-have-care-about-inequalityhttp://img.slate.com/media/3/100914_NoahT_GreatDivergence.pdfhttp://img.slate.com/media/3/100914_NoahT_GreatDivergence.pdfhttp://dx.doi.org/10.2139/ssrn.1154098http://dx.doi.org/10.2139/ssrn.1154098http://dx.doi.org/10.2139/ssrn.1154098http://dx.doi.org/10.2139/ssrn.1154098http://ssrn.com/abstract=1154098http://ssrn.com/abstract=1154098


47/47

.

Wilkinson, Richard G, and Kate Pickett. The Spirit Level: Why Greater Equality Makes Societies Stronger. New York:Bloomsbury Press, 2010.
http://www.washingtonpost.com/politics/in-state-of-the-union-address-obama-seeks-t-capitalize-on-incumbency-and-a-hot-issue/2012/01/24/gIQATpM1OQ_story.html?hpid=z2http://www.washingtonpost.com/politics/in-state-of-the-union-address-obama-seeks-t-capitalize-on-incumbency-and-a-hot-issue/2012/01/24/gIQATpM1OQ_story.html?hpid=z2http://www.washingtonpost.com/politics/in-state-of-the-union-address-obama-seeks-t-capitalize-on-incumbency-and-a-hot-issue/2012/01/24/gIQATpM1OQ_story.html?hpid=z2http://www.washingtonpost.com/politics/in-state-of-the-union-address-obama-seeks-t-capitalize-on-incumbency-and-a-hot-issue/2012/01/24/gIQATpM1OQ_story.html?hpid=z2

Exceptionally Unequal: US Income Inequality in Comparative Perspective

Documents