Trump - Accepting-Income-Inequality-2012.pdf

8/20/2019 Trump - Accepting-Income-Inequality-2012.pdf

1/27


2/27

Introduction

Income inequality in the United States has increased since the 1970’s: incomes at the very top

of the income distribution have soared, while median incomes have stagnated (Piketty and

Saez 2003, McCall and Percheski 2010). The inequality of market incomes has also increased

in European countries (Atkinson 2003). Increasing inequality has potentially serious political

consequences such as increasing perceptions of social distance among the population (Lupu

and Pontusson 2011), or impacting the political process as the wealthy have more influence

over political outcomes than the poor (Gilens 2012, Hacker and Pierson 2010). Given such

consequences, why do different societies tolerate such varied levels of inequality? How, if at

all, do increases in inequality affect public opinion? In this paper, I argue that public opinion

toward inequality is influenced by actual levels of inequality, leading to a self-reinforcing

effect of changes in inequality. I argue that, to a certain extent, human beings get used to

inequality when it is presented to them and that we base our understandings of fairness on

the environment we live in. The result is a measurable tendency to accept existing inequality

as fair and desirable.

The question “Where do attitudes toward inequality come from?” is not new, of course.

The expectation that the poor can be expected (or ought to) vote in favor of redistribution

because this is in their economic interest in an unequal society, is widespread in everyday

and academic writings on politics. One of the most famous, simple and elegant formulations

of this expectation is the Meltzer-Richard (1981) model, which predicts that with higher

inequality, demands for redistribution ought to increase. Empirically, this straight-forward

prediction tends not to be true. When comparing the U.S. and countries in continental Eu-

rope, more unequal countries tend to exhibit lower levels of redistributive demands (Alesina

and Glaeser 2004), and a cross-national study of 9 countries (Kenworthy and McCall 2008)

has shown that actual levels of inequality do not predict perceptions of inequality, prefer-

2


3/27

ences for redistribution or actual levels of redistribution. In the U.S., Kelly and Enns (2010)

find that increased inequality leads to more conservative public opinion, even among the

disadvantaged. Even when there is increased concern regarding inequality, at least among

American respondents this does not tend to lead to increased preferences for redistribution

(McCall 2007, McCall and Kenworthy 2009, Page and Jacobs 2009).

The lack of empirical support for the Meltzer-Richard prediction has inspired a significant

literature that attempts to identify the conditions under which the prediction may not hold

true. This search for confounding variables has yielded several interesting hypotheses and

findings. Benabou and Tirole (2006) argue that beliefs in the possibility of upward social

mobility decrease demands for redistribution. Iversen and Soskice (2006) point to electoral

systems as a possible intervening variable. The role of ethnic and racial heterogeneity, es-

pecially in the case where minorities are disproportionately poor, is frequently emphasized

(Gilens 2000, Alesina and Glaeser 2004, Roemer et al. 2007, Donnelly 2012). However, sub-

stantial variation in redistributive attitudes (and policies) remains to be explained (Stepan

and Linz 2011). One strand of research now emphasizes cultural/ideological explanations

for preferences over inequality that are not explained by economic position (Alesina and

Giuliano 2009). Perceptions of deservingness and the role of luck in determining economic

fortune do appear to be one of the variables that divide, for example, Europe and America.

Arguing that important variations in attitudes toward inequality originate in perceptions

of deservingness, fairness and luck, is a relatively recent approach within the subfield of polit-

ical economy, but not in political science as a whole. In 1959, Robert Lane argued that beliefs

in merit and opportunity provide rationales that not only lead working class Americans to

accept inequality but to actively fear equality. The belief that the system rewards merit im-

plies that the well-off deserve their fortune; this combination of beliefs dampens preferences

for equality. Hochschild (1981) also emphasises that, particularly in the economic domain

3


4/27

(as compared to the social and the political domains), Americans harbor a widespread belief

in the justifiability of differential returns. Perceptions of fairness and deservingness thus

strongly influence attitudes toward the justifiability of economic inequality. Below, I set out

to explore the malleability and origins of such perceptions of deservingness. Do increases in

inequality push us to perceive larger differences between individuals as desirable, implying

that beliefs in deservingness adjust to existing levels of inequality? What are the psycho-

logical mechanisms by which we evaluate the appropriateness of existing levels of inequality,

and do these mechanisms predispose us to accept the status quo as legitimate? Below, I

use the social psychology of system justification and status quo bias to develop hypotheses

regarding adaptive reactions to economic inequality.

The psychology of inequality

Social psychology has previously been used in the study of attitudes toward inequality, most

prominently in analyses of ethnic and racial heterogeneity as moderators of redistributive

demands (see above). More recently, psychological insights have also been brought into

laboratory experiments regarding redistributive preferences. In traditional ultimatum games,

the resources to be divided are bestowed randomly, with no incorporation of deservingness

or earning the resources. However, in real life redistributive decisions, we take into account

the origins of income and wealth, and consider whether such benefits are justly deserved.

In this spirit, Barber and English (2012) have shown that manipulation of the (perceived)

desert of incomes significantly alters the outcomes in ultimatum games. Perceived desert also

influences preferences for taxation and redistribution in more elaborate experimental settings

(Durante and Putterman, 2009). Even individuals who do not benefit from inequality are

more likely to acquiesce in and even prefer unequal distributions, if they perceive that the

differential rewards are earned.

4


5/27

Here, I am interested in a related but distinct questions. Rather than manipulating the

legitimacy of the resources to be distributed, I am interested in manipulating the distribution

itself. When the distribution is a relatively unequal one, does our tendency to believe in

deservingness lead us to accept such a distribution as legitimate? Can inequality itself

systematically change our perceptions of how much inequality is deserved? I argue that the

answer to these questions is yes. The theoretical approaches I draw on are status quo bias

and system justification theory.

In a now classic piece, Lerner and Miller (1978) argued that human beings are motivated

to believe that the world is just. In order to navigate a complex and unpredictable social

environment, we need to not only simplify the world around us, but also to believe that

it is predictable and controllable (at least to some extent). In a completely unpredictable

world, co-ordinating actions with desired end goals, and planning for the future would be

pointless activities; as such, in order to self-motivate, human beings need to believe in a world

where consequences are deserved and predictable. In other words, humans need to believe

the world is just. The evidence that Lerner and Miller brought to bear on this proposition

was strictly inter-personal: they showed that when their subjects observed another person

(a confederate) encounter misfortune, and the subjects were not in a place to help the

confederate, they derogated the confederate’s personality. This mental adjustment made the

misfortune appear less arbitrary and more deserved, enabling the subjects to retain their

belief in a just world (where individuals deserve what they receive). Initially, the framework

of ’belief in a just world’ predicted only that assessments of individuals changed as a function

of misfortune occurring to them.

The concept of ‘belief in a just world’ was subsequently expanded by Jost and Banaji

(1994, 2004) to apply to assessments of the social world . In this expansion, known as system

justification theory, Jost and Banaji argue that one consequence of the motivation to believe

5


6/27

in a just world is a defense of existing social arrangements, even when such justification occurs

at a cost to oneself or one’s social group. This motivation - to believe that the social world

rewards individuals based on merit and hard work - is important for individual abilities

to plan for the future and delay gratification (Laurin et al. 2010). System justification

theory emphasizes that humans have a tendency to think of the status quo as fair and

justified. We know that this tendency is most pronounced when the system justification

motive is activated: much research on system justification is centered on differences between

individuals with high and low system justification tendencies, or between individuals in

whom the system justification motive has (not) been activated. The question of how much

impact system justification tendencies as a whole have on societal-level processes is not

yet well understood (Jost and Hunyady 2002). System justification theory has been used to

explain adherence to social stereotypes (Jost and Banaji 2004), perceptions of discrimination

(Hafer and Choma 2009) and policy attitudes on affirmative action (Phelan and Rudman

2011). To date, there has not been an exploration of whether system justification tendencies

can also explain attitudes to economic or income inequality, but the evidence from these

previous studies is consistent with the existence of such a link. In the economic realm, the

belief that individuals get what they deserve, and that the economic system is just, would

logically lead to the conclusions that the well-off have deserved their privileged status, and

that the economic inequalities that exist have arisen for good reason. Thus, based on a

system justification motive, I hypothesize that individuals who perceive higher inequality in

the social system they inhabit also think of higher levels of inequality as fair.

The concept of system justification is closely related to the concept of status quo bias.

While system justification is thought of as a motivational force in human beings, status quo

bias focuses on the cognitive mechanisms that can lead to similar outcomes. Status quo

bias refers to the tendency of the human mind to prefer, and rate more positively, known

6


7/27

stimuli than unknown ones. For example, the mere exposure effect states that simply being

exposed to a photo of a human face causes us to later rate the same face as more likeable

than a previously unseen face. For a review of the numerous cognitive mechanisms that

lie behind a general status quo bias in humans, see Eidelman and Crandall (2009). While

the mechanisms emphasized by a cognitive bias toward the status qup are different from

the motivated reasoning suggested by system justification theory, they both point us toward

the same hypothesis: exposure to higher inequality will cause individuals to think of higher

inequality as fair and desirable.

In the following two sections, I present two experimental tests of my hypothesis. The first

experiment is a laboratory, where participants experience two different levels of inequality

and are subsequently asked how they would distribute the resources at hand. The second

experiment is embedded in a survey and uses real-life income differences as the variable of

interest.

Laboratory experiment: manipulating experiences of in-

equality

This laboratory experiment was designed as a test of principle to show whether levels of

inequality influence attitudes toward inequality. More specifically, the game is set up to

explore whether experiencing a relatively high level of inequality in the distribution of re-

wards after a competition can cause individuals to endorse higher inequality than those in

a more equal control condition. The laboratory setting was chosen as it provides the exper-imenter strong control over the environment and enables the manipulation of experiences

of inequality. Since I am interested in the impact of inequality itself (and not the impact

of economic self-interest) on distribution preferences, the experiment is set up so that all

7


8/27

participants are disadvantaged by inequality. I hypothesize that individuals who experience

an unequal division of resources will subsequently recommend a more unequal division of

similar resources.

Method: Participants were recruited for a study that they believed was about experiences

of competitive situations. When the participant arrived at the location of the experiment,

they were told that there are two participants in the experiment, and that the participants

will compete against each other (the ‘other participant’ was in fact a confederate of the re-

searcher). The participants first filled in a background survey that included only the Big Five

personality measures, the Global Belief in a Just World scale, the Social Dominance Orienta-

tion scale and demographic variables including ideology and partisanship. The participants

then ‘competed’ in a 4-minute anagram solving competition. The anagram competition

included the randomly assigned treatment condition: an extra monetary prize, to be dis-

tributed between the winner and the loser of the challenge. In the ‘unequal’ condition, the

winner was to get $9 and the loser was to get $1. In the ‘equal’ condition, the winner was

to get $6 and the loser was to get $4. The researcher verbally pointed out the existence of

a monetary prize in the competition; however, the exact dollar amount was only specified

on the written instructions received by the participant prior to the anagram task. Both the

researcher and the confederate were blind to the experimental condition until the debrief.

The words in the anagram task were neutral with respect to inequality (e.g. ‘rat’, ‘elbow’,

‘ocean’). The anagram task was designed to be challenging, and most participants reported

that they experienced the task to be ‘somewhat’ to ‘very’ difficult. After the anagram task,

participants filled in a second battery of Big Five questions while the researcher scored

the task. The participants scored between 0 and 32 points on the anagram task (roughly

equivalent to solving 0 to 12 anagrams), and the confederate always ‘scored’ 2 points more

than the participant (except in the case of very low participant scores, 0-5 points, in which

8


9/27

case the confederate ‘scored’ 1 point more than the participant). When the scores were

announced, the participants were reminded that they would get the second-place award while

the confederate would get the winner’s award. They then filled in the final questionnaire of

the experiment which was ostensibly about their experience of the competition. Included on

the final questionnaire was a question on the fairness of the pay-off that the participant had

received, and a question on how the participant herself would distribute the $10 between

the competitors, were she to design the game. The participants were then asked whether

they recalled what their monetary payment was going to be (manipulation check), asked for

any suspicions regarding the purpose of the experiment, debriefed, and paid. They kept the

money payment they had been promised during the experiment.

Participants: 65 participants were recruited using the Harvard Psychology Department

Study Pool in February - May 2012. They completed the study for a cash payment of $5

(55 participants) or course credit (10 participants) plus the cash payment earned during the

experiment. 12 participants were excluded from the analysis due to one or a combination

of: guessing the purpose of the experiment, guessing that the confederate was not a true

participant, and/or treatment failure (such as not remembering the pay-off of the winner

and loser). Including these participants does not change the results of the experiment. The

remaining 53 participants were a combination of college students and community members.

Ages ranged from 15 to 56 (mean 30, median 24). 25 were female and 28 male. 55% of

participants were White, 17% were African American, 11% were Hispanic, 13% were Asian

and 2 participants reported ’Other’. 26 participants were in the ‘unequal’ condition ($1-$9)

and 28 in the ‘equal’ condition ($4-$6).

Results: The main dependent variable was the amount of money the participant would

award to the winner of the competition, were they to design the game. The amount of money

awarded to the winner is used as a direct measure of how unequal the participant would make

9


10/27


11/27

$6 condition $9 condition

TotalLow BJWHigh BJW

Money allocation to the winner

5

6

7

8

9

1 0

Figure 2: Results of laboratory experiment. N=53. Responses broken down by below- andabove-mean levels of responses to the Belief in a Just World scale.

11


12/27

such as system justification, but also with cognitive biases such as the mere exposure effect

or anchoring in numbers. The mechanisms behind this result are important to understand,

as they can affect the interpretation of the results with respect to the real world. For exam-

ple, the conditions that may moderate this relationship are different depending on whether

the effect is due to anchoring (possible moderators include time lapsing since exposure to

inequality and the presence of other numeric information in the environment) or system

justification (possible moderators include political rhetoric regarding threats to the political

system, and individual perceptions of dependence on the political system).

The role of beliefs in a just world. One way to explore whether at least part of

this effect is explained by motivated, i.e. system justification, tendencies is to look at the

relationship between individual-level belief in a just world and the effect of the experimental

manipulation. Participants completed the belief in a just world (BJW) scale (Lipkus 1991),

which includes eight items such as ‘I feel that the world treats people fairly’ and ‘I feel that

when people meet with misfortune, they have brought it upon themselves’. Theoretically,

those who believe that the world is just should be more likely to accept the inequality as

justified, and thus be more likely to recommend unequal distributions of the prize money.

This is indeed what we see in Figure 2. In the figure, the experimental results are broken

down into groups of respondents who were below/above the mean score on the belief in just

world scale. The BJW scale ranges from 1 to 6, where higher scores indicate higher beliefs

that the world is just. In my sample, the responses ranged from 1.25 to 5.62, with a mean of

3.23 and median of 3.25. As can be seen in Figure 2, high BJW respondents recommend more

unequal distributions of money in both conditions. It is important to note that this pattern

does not occur with self-reported conservative-liberal ideology; political conviction does not

appear to play a role in producing these results.1 Controlling for experimental condition,

1There were too few (8) self-reported Republicans in my sample to allow a test with partisanship.

12


13/27

those high in BJW would give the winner $0.56 more (p=0.06). When looking at each

experimental condition alone, the numbers of participants ‘per cell’ decrease to between 12

and 17, and the differences are no longer statistically significant, though the numbers trend

in the expected direction. In the $9 condition, those high in BJW would give the winner

$8.15 and those low in BJW would give the winner $7.38 (a difference of $0.77, p=0.17).

In the $6 condition, those high in BJW would give the winner $6.36 and those low in BJW

would give the winner $6.00 (a difference of $0.36, p=0.11). This evidence is consistent with

the interpretation that experimentally induced inequality is accepted to a greater extent by

individuals with a higher belief in a just world than by individuals with a low belief in a just

world.

This experiment demonstrates that existing inequality has an impact on people’s prefer-

ences for inequality. In addition, this effect is present even when the perceived fairness of the

unequal condition is lower. In other words, even when people think of a situation as unfair,

and want to diminish the unfairness of it (in this case, by not offering the winner a full $9),

their attempt at equalizing the outcomes still ‘falls short’ of the answers they would have

given had they been in a more equal environment to begin with. In other words, even when

people perceive a situation as unfair and attempt to correct for that, they may not endorse

a distribution that is as equal as the one endorsed by people who start out with a relatively

equal status quo.

Survey experiment: Attitudes toward income inequality

The laboratory experiment shows that it is possible for experienced inequality to impact

preferences for inequality. However, the situation of laboratory experiment is artificial and

the ‘income’ under consideration was the pay from participating in a competition, and not

13


14/27

income in the sense we usually mean when discussing income inequality. The survey exper-

iment below addresses this by asking people about their beliefs and preferences regarding

real-life income inequality in the United States. As before, the hypothesis will be that

those who find out that their society is more unequal will become more accepting of income

inequality.

Measuring attitudes toward income inequality. The outcome variable in the survey

experiment will be attitudes toward income inequality, and specifically the respondents’

answers to the question of how large the income differences between different occupations

ought to be. Inequality at the societal level is an abstract concept, and it can be difficult to

formulate questions on inequality that are straight-forward and easy to understand for all

respondents. With this in mind, I chose to use a set of questions on occupational earnings

that has previously been used in the International Social Survey Project’s Inequality Module.

The respondents are asked how much money they believe that a list of occupations makes in

a year, and then they are asked how much they believe that these occupations ought to earn

in a year. This provides estimates of the respondents’ perceived level of income inequality

and of their desired level of income inequality.

The occupational groups used in this survey are: unskilled factory worker, skilled factory

worker, owner of a small shop, a doctor in general practice, a member of the federal cabinet

and a CEO of a large national corporation. To develop a uniform measure of perceived and

desired inequality, I use a justice index developed by Jasso (1977, 2000), which has been

previously used to analyse this question as asked in the ISSP (Austen 2002, Gijsberts 2002,

Hadler 2005, Kelley and Evans 1993, Kelley and Zagorski 2004, Kenworthy and McCall 2008,

Osberg and Smeeding 2006, Verwiebe and Wegener 2000). In this index, perceptions and

preferences of inequality are captured by computing ln(income of high prestige occupations

/ income of low prestige occupations). Since I am focusing purely on perceptions of income

14


15/27

inequality, without hypotheses regarding the relative prestige of occupations, I use the highest

earning and lowest earning occupations, as defined by the respondent. For each respondent,

then, the index of perceived and preferred income differences becomes ln(highest specified

income / lowest specified income). This yields two indices for each person: a perceived

income gap index, and a preferred income gap index. This log index is used for computing

the statistical significance of my findings below; for ease of interpretation, the non-logged

ratio of high to low incomes is provided in all figures.

Creating a treatment group. Based on correlational data from the ISSP, it is the case

that almost all respondents underestimate the true extent of income inequality in the United

States (Trump 2012, see also Norton 2010 for a similar observation with respect to wealth

inequality). This finding is replicated in my survey sample: there were no respondents who

guessed the income differences between occupational groups to be as large as they truly are.

From the point of view of experimental design, this is fortunate: no deception is required to

create a treatment that tells individuals that income inequality is higher than they think it

is. My treatment, thus, is simply a presentation of income data for 2010.

Method. I recruited 407 U.S. participants on Mechanical Turk in August 2012 to answer

an “Opinion survey”. The mean age of participants was 30, 63% were female, 55% had a

college degree or higher, 77% were Caucasian, and 20% self-identified as Republican. One

half of my sample (203 participants) were the control group: they answered some basic

demographic questions and the scale on belief in a just world, followed by the questions

on perceived and ideal income inequality. The remaining half (204 participants) received

information regarding the current income inequalities in the U.S. (see the Appendix for an

image of the information treatment). This information was inserted immediately after the

participants gave their guesses for existing income inequality, and immediately before they

gave their responses for how large income inequality ought to be. All respondents then

15


16/27

Percept ions No informat ion condition Information condi tion

Perceived and preferred income ratios

0

1 0

2 0

3 0

4 0

Figure 3: Survey experiment results. The 407 respondents were recruited on MTurk inAugust 2012.

also indicated their occupation, their annual income, and answered some political attitude

questions, including a question on whether it is the responsibility of the government to reduce

income differences.2

Results and discussion: The basic results of the survey experiment are summarized in

Figure 3. The outcome variable is a simple ratio measure of desired income inequality (high-

est recommended income / lowest recommended income). The plotted values are unlogged

ratios; significance is calculated after a log transformation. The first bar shows the mean

perceived level of income inequality; the second bar shows the mean preference for income

inequality in the control group, i.e. when no additional information has been presented. The

2The six political attitude questions were: ‘How often do you trust the government in Washington to dowhat is right?’, ‘Differences in income in America are too large.’, ‘Large differences in income are necessaryfor Americas prosperity.’, ‘It is the responsibility of the government to reduce the differences in incomebetween people with high incomes and people with low incomes.’, ‘The rich pay too much in taxes.’ and‘The government has a responsibility to help the poor.’

16


17/27

third bar shows the mean preference for income inequality in the treatment group, i.e. after

exposure to information about actual levels of income inequality in the U.S. After receiving

the information treatment, the preferred level of income inequality rises from 9 to 14 - a

50% increase from control group preferences. It is important to note that while the pref-

erences regarding inequality move up, no respondents completely accept the current level

of income inequality as justified. Just as all respondents underestimate the true level of

income inequality, they also prefer a reduction in inequality even after receiving informa-

tion on inequality. This tendency is consistent with the observation that not all inequalities

are always thought of as justified, and resistance to inequalities exists in the political world.

Rather than showing that all inequality is automatically accepted, my argument here is more

modest: increased inequality increases, on average, our perception of how much inequality

is acceptable.

As in the laboratory experiment, one way to explore whether the cause of this effect

is a purely cognitive psychological mechanism (such as anchoring) or whether motivated

reasoning is playing a role here is to look at the way that different demographic groups

react to the information treatment. There is no strong reason to suspect, for example,

that the tendency to anchor answers in previously seen numbers should vary either by the

individual’s baseline belief in a just world or by their partisanship. However, if the impact

of the information treatment is partially motivated, either through beliefs in a just world or

politics, then we can expect the experimental effect to vary by these traits. In Figures 4 and

5, the results are presented by above- and below mean levels of belief in a just world and

partisanship, respectively. When the treatment groups are broken down by belief in a just

world (BJW), there is no statistically significant difference between the low- and high-BJW

individuals in the control condition. However, in the treatment condition, those who believe

that the world is just have significantly more unequal preferences than those who do not

17


18/27

Perceptions No information condition Information condition

Total

Low BJW

High BJW


0

1 0

2 0

3 0

4 0

5 0

Figure 4: Survey experiment results. The 407 respondents were recruited on MTurk inAugust 2012. N (Low BJW) = 178; N (High BJW) = 229

believe that the world is just. This difference indicates an (at least partially) motivated

thinking regarding the acceptability of inequality.

In the laboratory experiment, where the sample of Republicans was very small, there were

no differences between partisans in acceptance of inequality. However, as shown in Figure

5, in the survey experiment, the differences between partisans are pronounced. Republi-

18


19/27

Perceptions No information condition Information condition

TotalDemocrat

Republican


0

1 0

2 0

3 0

4 0

5 0

Figure 5: Survey experiment results. The 407 respondents were recruited on MTurk inAugust 2012. N (Democrats) = 269; N (Republicans) = 80. Independent leaners wereincluded in the counts of partisans.

19


20/27

cans prefer somewhat more inequality in the control condition3 Both groups increase their

preference for income inequality after exposure to the information treatment, but this effect

is particularly pronounced for Republicans, whose preference moves from a ratio of 12.11

in the control condition to 33.37 in the treatment condition. When treatment condition,

partisanship and belief in just world are included in a regression predicting preference for

income inequality, the effect of belief in a just world is no longer significant, while partisan-

ship remains significant. This evidence suggests that when the topic at hand has political

implications, adherents of a party that traditionally argues in favor of the incentivizing na-

ture of income differences have an easier time accepting inequality as justified. It should be

remembered, however, that the impact of the information treatment on Democrats was also

significant, albeit smaller in magnitude.

The upward adjustments in respondents’ opinion on what magnitude of income inequality

is acceptable are somewhat removed from direct relevance to policy questions. Indeed, it is

possible that the upward adjustment of preferences (which does not reflect an acceptance of

the full amount of income inequality in society) is overwhelmed by an increase in demands

to put in place redistributive policies designed to reach this new (albeit updated) preference

for inequality. After all, a plausible hypothetical link between increasing income inequality

and increasing demands for redistribution is that, as people are made aware of inequality,

they start demanding that the government do something about it. In order to test for

this impact of my information treatment, my respondents were asked a number of policy

attitude questions at the end of the survey. The degree of agreement with the statement

‘It is the responsibility of the government to reduce the differences in income between the

rich and the poor’ is shown in Figure 6. The responses are presented by partisanship and

experimental condition. As seen in Figure 6, there are no significant differences by treatment

3In the control condition, there is a 8.25 mean preferred inequality ratio for Democrats, 12.11 meanpreferred inequality ratio for Republicans, p=0.05).

20


21/27

group. If anything, it appears that after the information treatment, Republican identifiers

become less likely to agree with this proposition, while Democrats become marginally more

likely to agree with it, but the differences are not statistically significant. The same results

occur with the related propositions ‘Differences of income in America are too large’ and

‘Large differences in income are necessary for America’s prosperity’. In conclusion, while

the information treatment successfully caused respondents to upward revise their thoughts

on how large income differences are acceptable, it did not have a discernible impact on their

attitudes toward the necessity of government redistribution or even whether differences of

income in America are too large. The implication is that individual citizens may habituate to

increased levels of inequality by upward revising their views on how much more the well-off

deserve when compared to the poor, while not changing their minds on questions that more

directly address government redistribution.

Conclusion

Differences in public willingness to accept and tolerate inequality of incomes and wealth are,

so far, only incompletely understood. We know that increases in inequality do not, with any

regularity, result in demands for redistribution. Using the two experiments in this paper, I

argue that one influence on the formation of redistributive attitudes is that people get used

to inequality. In the laboratory experiment, I showed that participants took their cues for

appropriate levels of inequality from the experimental set-up. Even when the unequal set-

up was experienced as more unfair than the equal set-up, suggestions for ameliorating the

inequality did not reach the same level of equality as those in the more equal experimental

condition. In the survey experiments, I replicated the general finding using real-life income

inequality as a referent. When participants were given the information that American in-

come inequality is higher than they believed it to be, they upward revised their suggestions

21


22/27


23/27

for acceptable levels of income inequality. This tendency was particularly pronounced for

Republicans, but it was statistically significant also among Democrats. Despite this change

in attitudes toward the acceptable range of income differences, there was no change in the

respondent’s agreement with the proposition that it is the government’s responsibility to

reduce income differences or even that income differences in America are too large.

These findings can be useful in making sense of some of the variation between times

and places when it comes to public acceptance of income inequality. Cross-nationally, it

tends to be the case that more unequal countries also exhibit more support for inequality

- and if inequality generates its own support, this is not surprising. How strong this effect

is - for example, whether these effects persist over time or are eradicated soon after leaving

the laboratory/survey environment is one important remaining question. A related research

question that deserves future attention is the role of communications and presentation of

data in the creation of this effect. If all information regarding income inequalities has this

effect on public opinion, the findings ought to give pause to left-wing organizations that seek

to change minds regarding the acceptability of current levels of inequality in America. If, on

the other hand, there are important differences in how this information is received depending

on source and presentation, these questions become important for better understanding the

over-time development of acquiescence as well as resistance to increases in income inequality.

23


24/27

References

Alesina, A. F. and P. Giuliano (2009). Preferences for Redistibution.

Alesina, A. F. and E. Glaeser (2004). Fighting Poverty in the U.S. and Europe: A World of Difference . Oxford: Oxford University Press.

Atkinson, A. (2003). Income Inequality in OECD Countries: Data and Explanations. Eco-nomic Studies 49 (4), 479–514.

Austen, S. (2002). An international comparison of attitudes to inequality. International Journal of Social Economics 29 (3), 218–237.

Barber, B. S. I. and W. English (2012). Divide Our Dollars, Not Divide The Dollar: Redis-tribution, Fairness, and the Ultimatum Game.

Benabou, R. and J. Tirole (2006, May). Belief in a Just World and Redistributive Politics.

Quarterly Journal of Economics 121 (2), 699–746.Donnelly, M. (2012). Identity and Interests in the United Kingdom. In Fifth Annual NYU-

CESS Conference , New York, pp. 1–37.

Durante, R. and L. Putterman (2009). Preferences for Redistribution and Perception of Fairness: An Experimental Study.

Eidelman, S. and C. S. Crandall (2009). A Psychological Advantage for the Status Quo. InJ. T. Jost, A. C. Kay, and H. Thorisdottir (Eds.), Social and psychological bases of ideology and system justification , Chapter 4, pp. 85–106. Oxford: Oxford University Press.

Gijsberts, M. (2002, December). The Legitimation of Income Inequality in State-socialistand Market Societies. Acta Sociologica 45 (4), 269–285.

Gilens, M. (2000). Why Americans hate welfare: race, media, and the politics of antipoverty policy . Chicago: University of Chicago Press.

Hacker, J. S. and P. Pierson (2010, May). Winner-Take-All Politics: Public Policy, PoliticalOrganization, and the Precipitous Rise of Top Incomes in the United States. Politics & Society 38 (2), 152–204.

Hadler, M. (2005, June). Why Do People Accept Different Income Ratios? A Multi-levelComparison of Thirty Countries. Acta Sociologica 48 (2), 131–154.

Hafer, C. L. and B. L. Choma (2009). Belief in a Just World, Perceived Fairness, andJustification of the Status Quo. In Social and psychological bases of ideology and system

justification , Chapter 5, pp. 107–125. Oxford: Oxford University Press.

24


25/27

Iversen, T. and D. Soskice (2006). Electoral Institutions and the Politics of Coalitions:Why Some Democracies Redistribute More Than Others. American Political Science Review 100 (2), 165–181.

Jasso, G. (2000). Trends in the Experience of Injustice: Justice Indexes About Earnings in

Six Societies, 1991 1996. Social Justice Research 13 (2), 101–122.Jasso, G. and P. H. Rossi (1977). Distributive Justice and Earned Income. American

Sociological Review 42 (4), 639–651.

Jost, J. and O. Hunyady (2002). The psychology of system justification and the palliativefunction of ideology. European Review of Social Psychology 13 (1), 111–153.

Jost, J. T. and M. R. Banaji (1994). The role of stereotyping in system-justification and theproduction of false consciousness . British Journal of Social Psychology 33 , 1–27.

Jost, J. T., M. R. Banaji, and B. A. Nosek (2004). A Decade of System Justification

Theory: Accumulated Evidence of Conscious and Unconscious Bolstering of the StatusQuo. Political Psychology 25 (6), 881–919.

Kelley, J. and M. D. R. Evans (1993, July). The Legitimation of Inequality: OccupationalEarnings in Nine Nations. American Journal of Sociology 99 (1), 75.

Kelley, J. and K. Zagorski (2004). Economic Change and the Legitimation of Inequality:the Transition From Socialism To the Free Market in Central-East Europe. Research in Social Stratification and Mobility 22 , 319–364.

Kelly, N. J. and P. K. Enns (2010, July). Inequality and the Dynamics of Public Opinion:The Self-Reinforcing Link Between Economic Inequality and Mass Preferences. American Journal of Political Science 54 (4), 855–870.

Kenworthy, L. and L. McCall (2008, December). Inequality, public opinion and redistribu-tion. Socio-Economic Review 6 (1), 35–68.

Laurin, K., S. Shepherd, and A. C. Kay (2010, August). Restricted emigration, systeminescapability, and defense of the status quo: system-justifying consequences of restrictedexit opportunities. Psychological science 21 (8), 1075–82.

Lipkus, I. (1991). The Construction and Preliminary Validation of a Global Belief in a JustWorld Scale and the Exploratory Analysis of the Multidimensional Belief in a Just World

Scale. Journal of Personality 12 (11), 1171–1178.

Lupu, N. and J. Pontusson (2011, May). The Structure of Inequality and the Politics of Redistribution. American Political Science Review 105 (2), 1–21.

McCall, L. (2007). The Undeserving Rich: Beliefs about Inequality in the Era of RisingInequality.

25


26/27

McCall, L. and L. Kenworthy (2009, August). Americans’ Social Policy Preferences in theEra of Rising Inequality. Perspectives on Politics 7 (03), 459.

McCall, L. and C. Percheski (2010, June). Income Inequality: New Trends and ResearchDirections. Annual Review of Sociology 36 (1), 329–347.

Meltzer, A. H. and S. F. Richard (1981). A Rational Theory of the Size of Government.Journal of Political Economy 89 (5), 914–927.

Norton, M. I. and D. Ariely (2010). Building a Better America - One Wealth Quintile at aTime. Perspectives on Psychological Science .

Osberg, L. and T. Smeeding (2006, June). ”Fair” Inequality? Attitudes toward Pay Differen-tials: The United States in Comparative Perspective. American Sociological Review 71 (3),450–473.

Page, B. I. and L. R. Jacobs (2009). Class War: What Americans Really Think About

Economic Inequality . Chicago: University of Chicago Press.

Phelan, J. E. and L. A. Rudman (2010, August). Reactions to ethnic deviance: The roleof backlash in racial stereotype maintenance. Journal of personality and social psychol-ogy 99 (2), 265–81.

Piketty, T. and E. Saez (2003, February). Income Inequality in The United States, 19131998.Quarterly Journal of Economics 118 (1), 1–39.

Roemer, J. E., W. Lee, and K. van der Straeten (2007). Racism, xenophobia, and distribution:multi-issue politics in advanced democracies . New York: Russell Sage Foundation.

Stepan, A. and J. J. Linz (2011, December). Comparative Perspectives on Inequality andthe Quality of Democracy in the United States. Perspectives on Politics 9 (04), 841–856.

Trump, K.-S. (2012). Perceptions of Income Inequality and Preferences for Income Distri-butions: Do People Who Perceive Higher Inequality Reject or Accept It? In Midwest Political Science Convention , Chicago.

Verwiebe, R. and B. Wegener (2000). Social Inequality and the Perceived Income JusticeGap. Social Justice Research 13 (2), 123–149.

26


27/27

A Appendix: Information treatment in survey exper-

iment

Graph shown to treatment group in the Mechanical Turk Survey experiment.

Occupation

Chairman of a large national corporation

Member of the cabinet in the federal government

Doctor in general practice

Owner of small shop

Skilled factory worker

Unskilled factory worker

Sources: Bureau of Labor Statistics, AFL-CIO,Payscale.com

Mean annual salary in 2010

$11,400,000

$199,700

$173,860

$74,580

$33,770

$24,240

Figure 7: Information treatment in Mechanical Turk experiment

27

Documents

Trump - Accepting-Income-Inequality-2012.pdf