What’s Fair in International Politics?Equity, Equality, and Foreign Policy Attitudes
Kathleen E. Powers,∗ Joshua D. Kertzer,† Deborah J. Brooks,‡ and Stephen G. Brooks§
August 13, 2020
Abstract: How do concerns about fairness shape foreign policy preferences? In this article,we show that fairness has two faces — one concerning equity, the other concerning equality— and that taking both into account can shed light on the structure of important foreignpolicy debates. Fielding an original survey on a national sample of Americans in 2014, weshow that different types of Americans think about fairness in different ways, and that thesefairness concerns shape their foreign policy preferences: individuals who emphasize equityare far more sensitive to concerns about burden sharing, are far less likely to support USinvolvement abroad when other countries aren’t paying their fair share, and often supportsystematically different foreign policies than individuals who emphasize equality. As longas IR scholars focus only on the equality dimension of fairness, we miss much about howfairness concerns matter in world politics.
Word count: 9828 words
∗Assistant Professor, Department of Government, Dartmouth College. [email protected]://kepowers.com†Paul Sack Associate Professor of Political Economy, Department of Government, Harvard University. jk-
[email protected]. http://people.fas.harvard.edu/ jkertzer/.‡Associate Professor, Department of Government, Dartmouth College. [email protected].§Professor, Department of Government, Dartmouth College. [email protected].
From alliance politics to climate change, many of the central challenges in international politics today
relate to questions of fairness. Superpowers who contribute more to collective defense complain about
burden-sharing (Oneal, 1990), while rapidly growing economies like China bristle at the prospect of
suffering disproportionate economic harm to protect the global environment. Fairness concerns pervade
territorial disputes (Goddard, 2006), peace negotiations (Albin and Druckman, 2012), international
cooperation (Kapstein, 2008; Kertzer and Rathbun, 2015; Efrat and Newman, 2016), and the domestic
politics of crisis bargaining (Gottfried and Trager, 2016). Among policymakers, fairness appears to be
a bipartisan principle: President Obama declared that “free riders aggravate me” and warned British
Prime Minister David Cameron to “pay your fair share” in military spending or risk the “special
relationship.”1 President Trump echoed Obama’s concerns that NATO allies’ reliance on U.S. defense
spending is simply “not fair”2 while also decrying China’s “unfair trade practices.”3
What makes a foreign policy action “unfair”? We argue that fairness has two faces. International
relations (IR) scholars tend to discuss fairness primarily in terms of equality — in which something
is fair if everyone receives the same outcome (see, for example, Baldwin, 1993; Albin and Druckman,
2012; Kertzer and Rathbun, 2015; Gottfried and Trager, 2016). This is consistent with a voluminous
body of research on the ultimatum game, which finds that players often reject offers that deviate from
a 50-50 resource division because unequal allocations are perceived as unfair (Guth, Schmittberger and
Schwarze, 1982; Thaler, 1988; Fehr and Schmidt, 1999; Knoch et al., 2006). The IR literature on relative
gains reaches similar conclusions, finding that actors dislike agreements that cause them to gain less
than the other side (Grieco, 1988; Mutz and Kim, 2017).
Yet equality is not the only criterion used to judge what’s “fair” — many actors are also motivated
to maintain equity. Equity implies that differential rewards are fair if they are proportional to actors’
relative contributions (Adams, 1965). Capturing this distinction is especially important given ideological
divides in American politics. Although most Americans report a commitment to fairness in the abstract
(Graham, Haidt and Nosek, 2009), they disagree on what fairness looks like in practice, with liberals
expressing more concern about equality than conservatives, for example (Haidt, 2012; DeScioli et al.,
2014; Jost, Federico and Napier, 2009; Meegan, 2019). As Hochschild (1981) and Fiske and Tetlock
(1997, 276) note, the tension between these two fairness conceptions animates many of the key debates
1Barack Obama, quoted in Jeffrey Goldberg, April 2016, “The Obama Doctrine,” The Atlantic.2Donald J. Trump. Twitter Post. July 9, 2018. 7:55AM. https://twitter.com/realDonaldTrump/status/10162896205967892483Donald J. Trump, “President Donald J. Trump is Confronting China’s Unfair Trade Policies,” The White House.
1
in American political culture. Yet apart from research on inequity aversion in international political
economy (IPE) (Lu, Scheve and Slaughter, 2012; Bechtel, Hainmueller and Margalit, 2017), IR scholars
who invoke fairness have almost exclusively focused on equality rather than equity, thereby neglecting
fairness’ second face.
In this article, we argue that both equity and equality have important implications for the study
of international politics, and we seek to make three important contributions to research on fairness in
foreign policy. First, we introduce a new way to measure individual differences in equity and equality
concerns that we believe will be useful for future research on fairness in both IR and political science
more generally. Since respondents can have different principles in mind when they report that fairness
matters using standard survey items (Rathbun, Powers and Anders, 2019), we elicit moral judgments
about specific equity or equality violations in everyday life. In an original national survey of 1,005
American adults, we find that the two faces of fairness form distinct factors: some Americans care
about equity, some care about equality, and some care about both.
Second, we show that the distinction between these two faces of fairness can help explain debates
in the United States about burden sharing. Although burden-sharing is one of the central dilemmas
in contemporary foreign policy, looming large in debates about the future direction of American grand
strategy (Brooks and Wohlforth, 2016), climate policy negotiations (Ringius, Torvanger and Under-
dal, 2002; Bernauer, Gampfer and Kachi, 2014), and global governance more generally, it is strangely
understudied in American public opinion about foreign policy. We show that individual differences in
concerns about equity meaningfully shape Americans’ attitudes about burden-sharing in international
politics, and can help explain the bipartisan aversion to disproportionate U.S. contributions. These
findings are consistent with bottom-up theories of public opinion about foreign policy, offering another
example of how personal values spill over into the foreign policy domain (Rathbun et al., 2016; Kertzer
and Zeitzoff, 2017): the more concerned about equity individuals are in their daily lives, the more they
are bothered by burden sharing imbalances in foreign policy.
Finally, we turn to a broader selection of foreign policy issues. We show that equality concerns are
associated with support for policies that advance joint gains, and equity concerns are associated with
support for policies that maximize relative gains. As a result, the effects of each face of fairness on
foreign policy preferences sometimes diverge: equality predicts support for free trade, for example, while
equity predicts support for protectionism, and these results hold even when controlling for partisanship
or political ideology. Together, our results demonstrate that as long as IR scholars primarily focus on a
2
single equality dimension of fairness, and associate fairness exclusively with prosociality (e.g., Kertzer
and Rathbun, 2015), we miss much about how fairness concerns shape foreign policy.
1 What’s fair in foreign policy?
One of the central puzzles in the study of public opinion about foreign policy is how the mass public
comes to form its judgments about foreign policy issues, despite knowing relatively little about inter-
national politics (Delli Carpini and Keeter, 1996). The political science literature on this subject has
largely fallen into two camps. Some scholars offer top-down models, in which members of the public
overcome their uncertainty about foreign policy issues by taking cues from trusted political elites, usu-
ally the leaders of their preferred political party (e.g., Berinsky, 2009; Baum and Potter, 2015; Guisinger
and Saunders, 2017). Others offer bottom-up models, in which members of the public overcome their
uncertainty about foreign policy issues by drawing on their basic value systems or orientations (e.g.
Hurwitz and Peffley, 1987; Goren et al., 2016; Prather, 2014; Kertzer and Zeitzoff, 2017). Unlike in
top-down models, which assume citizens are partisan but not ideological (Converse, 1964), bottom-up
models argue that citizens have more structured policy preferences than cynics suggest, because the
same values that shape our behavior in our personal lives also shape our foreign policy preferences
(Rathbun et al., 2016). People who care about retribution, for example, are more likely to support
punitive wars (Liberman, 2006; Rathbun and Stein, Forthcoming) and oppose unconditional financial
bailouts (Rathbun, Powers and Anders, 2019). The value commitments that predict our lifestyle choices
or consumption behaviors also predict our foreign policy preferences (Cohrs et al., 2005; Kertzer et al.,
2014; Bayram, 2015; Kreps and Maxey, 2018; O’Dwyer and Coymak, 2019).
One value that occupies a prominent place in this literature — and in the psychology of morality
more generally — is fairness (e.g., Graham, Haidt and Nosek, 2009; Rai and Fiske, 2011; Meegan, 2019).
Concerns about fairness are usually seen as having evolutionary origins: people must be able to detect
and punish cheaters if they want to enjoy the spoils from cooperation and guard themselves against
exploitation (Haidt, 2012), and fairness concerns typically begin to appear in children around the age of
five (Fehr, Bernhard and Rockenbach, 2008). Political scientists have thus linked fairness concerns to a
range of phenomena in international politics, including crisis bargaining (Gottfried and Trager, 2016),
post-war peace negotiations (Albin and Druckman, 2012), diplomacy (Kertzer and Rathbun, 2015),
international humanitarian law (Chu, 2019), international cooperation (Efrat and Newman, 2016), and
3
foreign direct investment (Chilton, Milner and Tingley, 2017).
What IR scholars have neglected, however, is that “fairness” carries multiple meanings, based on
different moral principles (Hochschild, 1981; Rasinski, 1987; Jennings, 1991; Trump, 2013; Brutger and
Rathbun, 2019). While there are debates in both normative and empirical research about the number
of distinct allocation principles (Deutsch, 1975; Wagstaff, 1994; Scott et al., 2001), for our purposes we
follow Rasinski (1987) in focusing on two principles in particular: equality and equity.
Equality implies a concern with egalitarianism of outcomes: an agreement or distribution is fair
when actors attain equivalent end-states. Behavioral economic games routinely find that participants
prefer resources to be distributed equally among players (e.g., Guth, Schmittberger and Schwarze,
1982; Thaler, 1988; Fehr and Schmidt, 1999; Bolton and Ockenfels, 2000). In ultimatum games, for
example, receivers usually “put their money where their mouth is” and reject unequal offers from
proposers (Camerer, 1997, 169), choosing to receive nothing at all rather than accept less than 50% of
the pot. Equality-minded individuals support policies that promote symmetric outcomes without regard
to whether some beneficiaries contribute more resources than others. In American politics, for example,
concerns about equality tend to be linked with support for social programs like welfare (Feldman and
Zaller, 1992). The same pattern applies in an IR context: If the purpose of an alliance is to ensure
equal security for all parties, the “fairest” arrangement might require wealthy members like the U.S. to
spend more than their poorer allies.
Equity, in contrast, shifts the focus from outcomes to inputs. Equity implies a concern with
proportionality: resource allocations should account for beneficiaries’ perceived contributions (Rai and
Fiske, 2011). Individuals who value equity believe that people ought to reap what they sow.4 The
actor who contributes more to a common resource merits a bigger slice of the pie: according to the
equity principle, actors’ payoffs should be proportionate to their effort (Adams, 1965; DeScioli et al.,
2014). Inequity occurs when individuals share a resource but some beneficiaries shoulder more of a
burden for supplying it.5 Demands that welfare recipients work to receive benefits often invoke equity
4As Fiske and Tetlock (1997, 276), note, each face of fairness stems from a different “relational model”: Equalityconstitutes fairness in Equality Matching relationships, which are predicated on in-kind reciprocity and common amongpeers or co-workers. Fairness as equity marks Market Pricing relations, where people interact according to a principle ofproportionality.
5We use the term “inequity” to refer to circumstances where resource allocations are not proportionate to the relativecontributions of beneficiaries. Inequality exists when members of a group receive disparate benefits, irrespective ofcontributions. This distinction highlights one potential source of confusion that arises from importing behavioral economicsto IR: scholars sometimes use the term “inequity aversion” to refer to the human aversion to both inequity and inequality,and test economic models of inequity aversion by assessing how participants “respond to inequalities” (Wilson, 2011,208). This is largely due to the standard structure of laboratory economic games. A participant who enters a lab toplay an ultimatum game typically does not have information about whether she or her partner has contributed more tothe resource endowment. Without a metric to determine the equitable input/outcome ratio, “the equitable outcome, is
4
principles, for example, and research on free-riding demonstrates that equity violations plague social
dilemmas (Ostrom, 1998). Fuhrmann (2019) describes how weaker states in an alliance have incentives
to free-ride because they can benefit from collective deterrence while powerful allies like the U.S. pay the
costs. When policymakers protest that it is unfair for some NATO members to dedicate the requisite
2% of their GDP to defense while others spend less but receive the same security benefits from the
alliance, they call attention to inequity. Free-riding is common, but inequitable.6
Despite evidence that people evaluate fairness in terms of both equality and equity, IR scholarship
almost exclusively focuses on the former. When Albin and Druckman (2012), for example, find that
“just” civil war settlements are more durable than their unfair counterparts, they focus on the dis-
tributive justice principle of equality. Former belligerents prefer agreements that provide equal rights
for citizens of all parties to the agreement alongside equal political power. Efrat and Newman (2016)
similarly rely on equality when they argue that states will be more likely to defer child abduction cases
to partners whose legal systems are fair. In public opinion, Kertzer et al. (2014), Kreps and Maxey
(2018), and Cram et al. (2018) similarly define fairness in terms of equal treatment for individuals.
Equality’s privileged place in research on fairness in IR is significant for several reasons.7 First,
although we know that equity and equality tend to dominate in different domains in domestic politics
— equality reigns in the family, but equity in the marketplace, for example (Hochschild, 1981; Jennings,
1991), such that splitting food equally tends to be considered fair, while splitting money equally re-
gardless of contribution is not (DeVoe and Iyengar, 2010) — we have little sense of when each principle
dominates in the foreign policy realm.
Second, even those few IR scholars who do study equity tend to treat equity preferences as a
constant, rather a variable. Scholarship on inequity aversion in IPE assumes that everyone bristles
at inequity (Lu, Scheve and Slaughter, 2012; Bechtel, Hainmueller and Margalit, 2017). Yet an ever-
growing body of psychology research tells us that people vary in their commitment to moral principles
given by the egalitarian outcome” (Fehr and Schmidt, 1999, 822). See Starmans, Sheskin and Bloom (2017) for a similardiscussion in the context of economic inequality. We avoid this complication by only using the terms equity and inequityin contexts where relative contributions are known and relevant.
6Although equity concerns can draw on objective metrics — an investor will earn part of the company’s profits inproportion to what she invests — subjective perceptions often shape equity judgments. Many Americans prefer wealthinequality to complete equality, for example. If some people are more capable than others, or exert more effort, equity-minded citizens conclude that fair societies contain inequalities (Starmans, Sheskin and Bloom, 2017; Trump, forthcoming)— but assessments of capability and effort entail subjective judgments.
7In addition to work on inequity aversion, which we describe below, the only other work on equity in IR we areaware of is Gottfried and Trager (2016), who draw on equity theory when presenting their theoretical model. Consistentwith much of the behavioral economic tradition more generally (see e.g., Fehr and Schmidt, 1999), however, they use anexperimental protocol in which neither party has an unambiguous claim to a larger share of the disputed object (Gottfriedand Trager, 2016, 253). The study thus enables the authors to investigate the effects of concerns about equality, but notconcerns about equity.
5
(Graham, Haidt and Nosek, 2009). Consistent with research on the role of equality in foreign policy
public opinion (Kertzer et al., 2014), we can treat equity concerns as an individual difference, asking
not just whether equity matters, but for whom. Like the business-minded leaders in Fuhrmann’s (2019)
research on free-riding, some members of the public might be especially sensitive to imbalances between
inputs and outcomes.
Third, and related, research in social psychology shows that liberals value equality more than their
conservative counterparts (Haidt, 2012; Starmans, Sheskin and Bloom, 2017; Meindl, Iyer and Graham,
2019). When conservatives invoke fairness in domestic politics, they tend to be concerned primarily with
whether those who work hard or pay more taxes reap appropriate rewards, whereas liberals emphasize
both equity and equality, hoping to advance societal well-being by meeting everyone’s basic needs.
This distinction illuminates partisan differences about welfare work requirements, for example, but also
clarifies important areas of convergence such as the widespread support for social security across the
ideological spectrum: working Americans all “pay in” but the program provides some financial stability
for all citizens.8 Disaggregating fairness also enriches our understanding of how fairness shapes foreign
policy attitudes. If equality is primarily a liberal value, but equity matters to Americans across the
ideological spectrum, we can make sense of the cross-partisan nature of complaints about free-riding in
U.S. foreign policy. Moreover, psychologists have also found gender differences in fairness preferences:
whether due to structural societal differences or early childhood socialization, women often display
stronger preferences for equality than men do (Rasinski, 1987; Scott et al., 2001). If fairness attitudes
are correlated with distinctive foreign policy preferences, the tension between these competing fairness
principles could explain part of the gender gap in public opinion about foreign affairs (Brooks and
Valentino, 2011; Mansfield, Mutz and Silver, 2015; Eichenberg and Stoll, 2012; Lizotte, 2019).
Our initial goal, then, is to establish that different types of Americans think about fairness in
different ways: to demonstrate that equity and equality are distinct moral principles, and to map
constituencies that support each face of fairness. These considerations lead to our first hypothesis:
H1: Fairness has two faces, with equality and equity forming distinct dimensions.
Our claim, though, is not simply that equality and equity constitute different dimensions of fairness
attitudes, but that both faces of fairness matter for foreign policy preferences in different ways. IR
8Meegan, Dan (2019). “Conservatives Have a Different Definition of ‘Fair’ And liberals ignore it at their peril.” TheAtlantic. 30 April. As Haidt (2013) notes: “Everyone endorses proportionality, but the left simultaneously endorses equal-ity, even when it is in tension with proportionality. The right has no interest in equality for its own sake. Conservativesprefer proportionality, even when it leads to massive inequalities of outcome”.
6
scholars often portray fairness as a prosocial value that inspires international cooperation by promoting
positive reciprocity (Kertzer et al., 2014; Kertzer and Rathbun, 2015), but this expectation only holds
if we limit our understanding of fairness to equality. Because equality-minded Americans prioritize
outcomes, they are inclined toward foreign policies that maximize jointly enjoyed gains and improve
global conditions. In pursuit of parity, they set aside egoistic concerns about the U.S.’s return on their
foreign policy investments — like whether the U.S. gains more relative to other states or contributes
more resources than its partners in pursuit of just, egalitarian ends.9 The U.S. can contribute to
egalitarian outcomes by working through international institutions like the UN, providing foreign aid to
developing countries, or helping to clean up the global environment so that everyone has equal access
to clean air. Each policy is compatible with an equality principle: The fact that the U.S. might be
required to contribute more to institutions, aid, or environmental protection than other states does not
undermine fairness when it is defined as equality. This logic explains why previous research reports a
relationship between fairness values and cooperative internationalism (Kertzer et al., 2014), or between
equality-oriented predispositions like Social Dominance Orientation and support for trade agreements
that maximize joint rather than relative gains (Mutz and Kim, 2017).
By contrast, equity-oriented individuals attend to the cost side of the equation. Just as research on
retribution demonstrates how concerns about justice can lead to aggression (Liberman, 2006; Rathbun
and Stein, Forthcoming), we argue that equity-minded Americans will oppose international cooperation
on the basis of fairness. Equity does not imply prosociality. An equity principle demands that actors
receive rewards that reflect their relative contributions, a situation that rarely obtains when the U.S.
responds to distant global problems.
We therefore expect that compared to equality, equity concerns will be strongly associated with
negative evaluations of burden-sharing problems in international politics in particular, and with oppo-
sition to policies that do not strike a balance between the price the U.S. pays and the direct benefits
it receives in general. Insofar as concerns about free-riding dominate debates about everything from
NATO contributions to humanitarian interventions, to financial bailouts, and climate change negoti-
ations, we miss out on important dynamics in world politics if we measure only one face of fairness.
Moreover, when a policy promises global or indirect benefits at a high cost to the U.S., the two faces of
9Although people who conceive of justice in terms of equality are more likely to endorse separate other-regardingmoral beliefs (Graham, Haidt and Nosek, 2009), equality is not a proxy for general prosociality. Research from the moralfoundations tradition shows that the fairness and harm/care foundations co-vary — but they remain empirically andconceptually distinct moral systems (Haidt, 2012). Equality, like equity, taps beliefs about justice (Meindl, Iyer andGraham, 2019), whereas harm/care refers to individuals’ concerns about others’ suffering. Caring taps compassion, notfairness.
7
fairness will diverge — equality will be associated with support whereas equity will be associated with
opposition.
Together, these theoretical insights lead to two additional hypotheses:
H2: Equity is associated with concerns about burden-sharing in foreign policy.
H3: Equality is associated with support for policies that maximize jointly enjoyed gains,
while equity is associated with support for policies that maximize relative gains.
2 Methods and results
To demonstrate the value of studying both faces of fairness in IR, we conducted an original survey in
August 2014 on a national sample of 1,005 Americans recruited through Survey Sampling International
(SSI). Participants, 69.4% of whom identified as white and 51.3% of whom identified as female, ranged
in age from 19-95 (median: 49) and reported a median household income of $50-60,000. SSI employs an
opt-in method to recruit a panel of participants targeted to census quotas for sex, age, race, and region;
Table 1 in Appendix §1.2 shows the sample matches census targets on key demographic characteristics.10
We present our analysis in three stages. First, we introduce our measurement strategy to show
that Americans differentiate between equity and equality when they make judgments about right and
wrong, and that Americans from both ends of the political spectrum value equity. Second, we show that
variations in Americans’ equity commitments can explain the polarized debates in the United States
about burden sharing in American foreign policy. Third, we analyze a series of broader foreign policy
issues to show that the effects of equity and equality sometimes diverge, with concerns about equality
associated with support for policies that maximize global gains, and concerns about equity associated
with policies that maximize relative gains. These results highlight why scholars need to consider both
faces of fairness in research on public opinion in foreign policy.
2.1 Differentiating between two faces of fairness
Although political scientists sometimes use the Moral Foundations Questionnaire (MFQ) to study the
extent to which individuals care about people being “treated fairly” (Graham, Haidt and Nosek, 2009),
our purposes preclude this and similar scales because the meaning of “fairness” here is itself ambiguous:
10The sample closely matches the U.S. population on key variables including sex, age, race, and region. Highly educatedAmericans are slightly over-represented in our sample. In the appendix, we show that our substantive conclusions holdwhen we reweight the data to more closely match population parameters for educational attainment.
8
participants could interpret ‘fair’ treatment as equity, equality, or both — leaving us uncertain of which
construct we are measuring.11 A key goal of this paper is to develop an alternative measurement scale
that can usefully distinguish between equity and equality for future political science research.
We therefore build on Clifford et al. (2015, 1179), who develop a series of “moral foundations
vignettes” to assess individuals’ moral commitments. Each vignette describes a situation in which an
actor violates some moral principle, and asks respondents to evaluate whether the action feels morally
wrong. The vignettes are designed to depict violations of one value at a time, such that ideal items
will discriminate between even closely related moral principles, and to distinguish moral violations
from social norms. For example, the fairness vignettes avoid references to (1) physical or emotional
harm, which taps the harm/care foundation, (2) hierarchical relationships, lest they invoke authority
values, and (3) “race, gender, or structural equality” because these characteristics are more likely to
tap other, non-moral, attitudes (Clifford et al., 2015, 1181).12 Clifford et al. (2015, 1179) notes that the
moral foundations scale “relies on respondents’ rating of abstract principles, rather than judgment of
concrete scenarios,” and that abstract endorsements may not always translate into political attitudes.
The vignettes allow us to probe concrete moral judgments, but in scenarios taken from everyday life,
rather than politics or foreign policy.
Table 1 displays the seven vignettes we employ to measure fairness attitudes, building on the
fairness inventories established by Clifford et al. (2015), Iyer (2010), and Meindl, Iyer and Graham
(2019). For each vignette, participants indicated the extent to which the situation felt morally wrong
on a 7-point scale from “not at all wrong” to “extremely wrong.” The first three vignettes in Table
1 depict clear equity violations, where individuals receive outcomes that are not proportionate to the
inputs they provide. For example, “A student copies another student’s work and gets the same grade”,
or “A runner takes a shortcut on the course during a marathon” both describe outcomes that do not
accurately correspond to individuals’ contributions. These scenarios are not problematic on equality
principles (we don’t care about guaranteeing equality of outcomes when evaluating assignments, or on
a race course) but are problematic on equity principles (we expect that individuals who perform better
will be rewarded as such).
The last two vignettes capture clear equality violations, where actors fail to attain equivalent
11Rathbun, Powers and Anders (2019) also highlight the ambiguity inherent in the MFT fairness scale, and showthat the effect of fairness depends on individuals’ commitment to negative reciprocity/retribution. See also Brutger andRathbun (2019).
12To validate the vignettes and establish a database for future use, Clifford et al. (2015) eliminate an item if 40% ofrespondents did not classify it as violating the intended moral foundation. See Appendix §1.3 for a broader discussion ofscale construction.
9
Tab
le1:
Fai
rnes
sV
ign
ette
s:T
ow
hat
Exte
nt
Does
Each
Sce
nari
oF
eel
Mora
lly
Wro
ng?
Vio
late
sF
act
or
load
ings
Sce
nar
ioE
qu
ity
Equ
ali
tyM
ean
(SD
)P
A1
PA
2
1A
stu
den
tco
pie
san
oth
erst
ud
ent’
sw
ork
on
an
un
gra
ded
ass
ign
men
t.(C
opie
sU
ngr
aded
)Y
N5.6
9(1
.54)
0.6
79
2A
stu
den
tco
pie
san
oth
erst
ud
ent’
sw
ork
an
dget
sth
esa
me
gra
de.
(Copie
sG
raded
)Y
N6.1
2(1
.33)
0.8
62
3A
runn
erta
kes
ash
ortc
ut
onth
eco
urs
ed
uri
ng
am
ara
thon
.(M
ara
thon
Short
cut)
YN
6.0
9(1
.36)
0.8
3
4T
wo
bro
ther
sw
inth
elo
tter
yw
ith
ati
cket
they
bou
ght
toget
her
,b
ut
the
earn
ings
aren
’td
ivid
edev
enly
.(L
ott
ery
Div
isio
n)
YY
5.7
5(1
.57)
0.5
66
0.2
12
5A
girl
take
sal
lof
the
Hal
low
een
can
dy
from
ab
owl,
leav
ing
non
efo
rot
her
s.(H
all
ow
een
Can
dy
)Y
Y5.7
6(1
.49)
0.6
53
0.1
7
6A
rais
eis
give
nto
aw
orke
r,w
hen
anot
her
work
ern
eed
sit
more
.(R
ais
efo
rO
ne
Work
er)
NY
3.7
2(1
.9)
0.8
3
7A
nem
plo
yee
earn
sa
lot
ofm
oney
wh
ile
an
oth
erea
rns
ver
yli
ttle
.(E
m-
plo
yee
Earn
sM
ore
)N
Y4.1
0(1
.96)
0.8
13
Note
:A
llit
ems
ran
ge
from
1-7
,w
ith
hig
her
valu
esin
dic
ati
ng
resp
on
ses
that
the
situ
ati
on
feel
sm
ore
mora
lly
wro
ng.
Th
ese
con
dan
dth
ird
colu
mn
sin
dic
ate
wh
eth
erth
esc
enari
oin
volv
esan
equ
ity
vio
lati
on
(in
wh
ich
the
ou
tcom
eis
not
pro
port
ion
al
toth
ein
pu
ts),
an
equ
ality
vio
lati
on
(in
wh
ich
the
ou
tcom
esare
ineg
alita
rian
),or
both
;th
efi
rst
thre
esc
enari
os
vio
late
equ
ity
bu
tn
ot
equ
ality
,th
ela
sttw
osc
enari
os
vio
late
equ
ality
bu
tn
ot
equ
ity,
an
dtw
osc
enari
os
(LotteryDivision
,an
dHalloweenCandy
)vio
late
both
.T
he
plo
tsh
ow
stw
okey
fin
din
gs.
Fir
st,
scen
ari
os
that
incl
ud
ean
equ
ity
vio
lati
on
are
seen
as
sub
stanti
ally
more
mora
lly
wro
ng
than
scen
ari
os
that
do
not,
show
casi
ng
the
imp
ort
an
ceof
equ
ity
inou
rm
ora
lju
dgm
ents
.S
econ
d,
the
resu
lts
from
an
exp
lora
tory
fact
or
an
aly
sis
fin
ds
that
atw
o-f
act
or
solu
tion
map
sonto
ou
rth
eore
tica
lco
din
gs,
wit
hon
efa
ctor
refe
rrin
gto
equ
ity,
an
dth
eoth
erto
equ
ality
.In
tere
stin
gly
,th
etw
osc
enari
os
that
incl
ud
eb
oth
equ
ity
an
deq
uality
vio
lati
on
ssh
ow
som
ew
eak
cross
-load
ing,
bu
tu
ltim
ate
lylo
ad
on
the
equ
ity
fact
or
rath
erth
an
the
equ
ali
tyfa
ctor,
sugges
tin
gth
at
the
equ
ity
con
cern
sare
more
salien
t.
10
end-states. It might be considered fair from an equity perspective when “An employee earns a lot
of money while another earns very little” (since under equity principles, earnings can be guided by
merit), but this asymmetry in wages is considered unfair from an equality perspective (Meindl, Iyer
and Graham, 2019). Finally, two of the vignettes depict a violation both of equity principles and of
equality principles. In the lottery division scenario (“Two brothers win the lottery with a ticket they
bought together, but the earnings aren’t divided evenly”), the uneven division of the winnings not only
represents an equality violation, but also an equity violation, since the two brothers bought the ticket
together. In the halloween candy scenario (“A girl takes all of the halloween candy from a bowl, leaving
none for others”), the girl who absconds with the candy not only fails to split it evenly with others (an
equality violation), but given the lack of information indicating she was the the one who provided all of
the candy in the first place, likely takes a haul that is disproportionate to her contributions (an equity
violation).
Two additional points are worth noting about Table 1. First, the table presents basic descriptive
statistics for each vignette. Although the means for most of the items are relatively high – consistent
with extensive research showing that fairness is an important moral principle for most Americans
(Graham, Haidt and Nosek, 2009; Haidt, 2013) — some are substantially higher than others: scenarios
that include an equity violation are seen by our respondents as substantially more morally wrong than
scenarios that do not, showcasing the importance of equity in our moral judgments.
Second, we conducted an exploratory factor analysis (EFA) on the fairness vignettes. If fairness is
unidimensional — if moral judgments about fairness violations depend on a single principle — we would
find that one factor explains most of the common variance in the data. Instead, parallel analysis and
model fit statistics suggest a two factor solution, which fits the data well (TLI=0.94, RMSEA=0.09).13
We therefore estimate a two factor solution using principal axis factor analysis with oblimin rotation,
producing the factor loadings in the two right-hand columns of Table 1. Consistent with H1, the results
show that the first three vignettes, which violate equity principles, load on a single factor; the last two
items, which violate equality principles, load on the other factor. And, of particular interest, the two
items that feature both equity and equality violations cross-load on both factors, but ultimately display
stronger loadings on the equity factor than the equality factor. This finding suggests that respondents
confronted with both equity and equality violations found the equity concerns to be more salient, and
13In addition to parallel analysis, the Kaiser criterion suggests keeping all factors with eigenvalues greater than 1, andthe second factor approaches this cutoff with a value of 0.96, while the third drops to 0.11. Moreover, a 3-factor solutionproduces Heywood cases, which suggest a misspecified model and threaten the validity of the results (Fabrigar et al.,1999).
11
reinforces the importance of taking fairness’ second face into account.
Figure 1: Strong concerns about equity across demographic groups
Female Male
Liberal Moderate Conservative
Democrat Independent Republican
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
0.00 0.25 0.50 0.75 1.00
Density
Fairness type
Equality
Equity
Note: N=1,005. Figure 1 displays density distributions for equity and equality across party identification, ideology, andgender. The plot shows that both the left and right value equity, whereas left-leaning respondents place a heavieremphasis on equality than right-leaning ones do. Women care more about both types of fairness than men do.
To obtain respondent-level measures for sensitivity to different types of fairness concerns, we extract
factor scores from the factor analysis and produce latent measures for concerns about equity (mean =
0.82, sd = 0.19) and concerns about equality (mean = 0.51, sd = 0.26).14 Figure 1 displays the density
distributions for equity and equality across partisan and ideological categories. What is striking is just
how widespread concerns about equity are. For each constituency, equity trumps equality, and the
results are also relatively stable across partisan and ideological subgroups: conservatives value equity
14Both are rescaled from 0 to 1 in the analysis below for for ease of interpretability.
12
more than liberals do (t = −2.57, p < 0.05), but the substantive size of the difference is relatively small.15
In contrast, we find both ideological and partisan divides on equality: Democrats care more about
outcome-oriented equality violations than their their Republican counterparts (t = 6.71, p < 0.01), and
liberals more than conservatives (t = 3.78, p < 0.01). This pattern is consistent with what psychologists
now recognize as the line that divides ideas about fairness in American politics (Haidt, 2013; Clifford
et al., 2015; Meegan, 2019). Multiple group factor analysis in Appendix §2.3 further confirms that
Democrats and Republicans think about fairness in slightly different ways: although we obtain the
same factor solution in both groups, we also find that the two latent factors are moderately correlated
among Democrats (r = 0.30), but not among Republicans (r = 0.01): Democrats who care more about
equity tend to also care more about equality, but the same is not true for their Republican counterparts.
Finally, we find even starker differences with respect to gender. Women care significantly more
about fairness violations than men do, both for equity (t = −4.419, p < 0.01) and equality (t = −4.27,
p < 0.01). This highlights the cost of privileging equality in research on fairness in IR. Equality alone
cannot capture how Americans think about fairness and in fact, appears to be a significantly weaker
concern than equity, even among liberals. The question we turn to next is how these different types of
fairness concerns are associated with foreign policy preferences.
2.2 Equity predicts opposition to burden-sharing violations
Our initial findings show that most Americans value equity. We therefore turn to a specific class of
foreign policy issues where we expect this type of moral principle to loom particularly large: burden-
sharing issues in foreign policy. Despite the fact that burden-sharing concerns animate contemporary
debates about the most consequential foreign policy issues from NATO to the Paris Climate Agreement,
they are rarely included in standard measures of foreign policy attitudes (e.g., Hurwitz and Peffley, 1987;
Chittick, Billingsley and Travis, 1995; Rathbun et al., 2016; Gravelle, Reifler and Scotto, 2017). We
argue that when members of the public decry actions in which the U.S. bears disproportionately large
costs, they transfer their general concern for equity in their daily lives up to the foreign policy domain.
We designed four dependent variables that solicit individual reactions to common scenarios in which
the U.S. makes a substantial contribution to resolve a collective problem. Participants responded to four
hypothetical foreign engagements and rated the extent to which each poses a problem for the U.S. on a
scale from 1 (“not a problem at all”) to 7 (“a very big problem”). Each item, listed in Table 2, describes
15Importantly, these differences in equity are ideological rather than partisan, since Republicans do not value equitysignificantly more than Democrats do (t = −1.593, p < 0.112).
13
Table 2: Burden sharing scenarios: How much of a problem is each for the United States?
Countries such as Germany, Canada, and Japan devote a far smaller share of their economy todefense spending than the United States does, because they are US allies and America has pledgedto defend them. (Defense Budgets)
Western allies give less foreign aid to provide for education and health care for women and childrenin the Middle East, because they know the US will foot the bill. (Foreign Aid)
The US provides all of the needed troops and money for a peacekeeping mission while othercountries do not contribute any troops or money to the mission. (Peacekeeping)
The US provides all of the needed resources and personnel for cleaning up toxic waste contaminationfrom a sunken ship in Antarctica while other countries do not contribute any resources or personnelto the effort. (Environment)
a different substantive domain in which the U.S. plays a dominant role in resolving a collective problem.
Defense Spending, for example, implicates U.S. allies who earmark fewer resources for defense, while
Environment introduces a hypothetical scenario in which the U.S. foots the bill for an environmental
disaster in international waters. The scenarios each highlight the concerns about proportionality that
plague foreign policy questions across issue areas. In creating the four scenarios, we aim to capture
a set of problems that implicate equity and demonstrate its important role in foreign policy public
opinion. We turn to a broader set of foreign policy problems that implicate both equity and equality
in a subsequent section below.
In addition to factor scores for equity and equality, our primary independent variables, we control for
three widely-used scales for foreign policy orientations — militant internationalism, cooperative inter-
nationalism, and isolationism (Wittkopf, 1990). Militant internationalism (MI) refers to an inclination
to use force to achieve foreign policy goals, and emphasizes the importance of demonstrating military
resolve. This measure of military assertiveness thus taps the familiar distinction between foreign policy
hawks and doves. Cooperative internationalism (CI) captures the extent to which individuals want
the U.S. to work with other states and international institutions to solve global problems like climate
change (Wittkopf, 1990). It entails a commitment to global participation but not to military force. For
isolationism, we include a scale that taps this general preference for disengagement from the world —
a stance that maintains that America should “come home” and scale down its conception of itself as a
leader (Chittick, Billingsley and Travis, 1995).
Finally, we included a battery of demographic questions alongside measures of partisanship and
ideology. Participants report their age, sex, race, education attainment (from less than high school to
14
Post-graduate), income (split into quartiles for analysis), and U.S. region of residence (midwest, north-
east, south, or west). We measure self-reported partisanship with a 7-point scale where a 7 indicates
strong Republican, and ideology on a 7-point scale from extremely liberal to extremely conservative.
Table 3 presents estimates from a series of OLS regression models that predict responses to the four
burden sharing vignettes. The dependent measures and continuous independent variables have been
rescaled from 0 to 1, and higher values indicate that participants rated the scenario a bigger problem
for the United States. Positive coefficients suggest that stronger commitments to the moral principle
are associated with less support for America taking on an “unfair” global burden. Models 2, 4, 6, and
8 include foreign policy orientations and the demographic controls.
Consistent with H2, equity is a statistically and substantively important predictor of attitudes
toward U.S. contributions to global problems. The extent to which individuals believe that input/output
ratios dictate whether an action should be deemed moral or immoral predicts their view of whether
certain foreign policy activities are justifiable. Indeed, the environment, foreign aid, and peacekeeping
vignettes draw almost exclusively on equity concerns — not equality. In the case of environmental
cleanup in Antarctica, a move from the minimum to the maximum on equity is associated with a 0.45-
unit increase in the policy preferences — a shift in nearly half the 0-1 scale. Since Antarctica does
not belong to just one state, the cleanup arguably benefits all global actors. Like President George
W. Bush, who avowed that he would not “let the United States carry the burden for cleaning up the
world’s air,” equity-minded citizens think that other states should contribute if they expect to reap the
rewards from a clean environment.16
The same logic underlies the strong association between equity and foreign aid. Because allies
know that the U.S. will provide long run development benefits abroad, they shirk the opportunity to
contribute a share proportionate to their GDP or relative interest in the positive externalities associated
with advancing women’s healthcare. As participants’ moral commitment to equity increases, so does
their disdain for this foreign aid arrangement (b = 0.469, p < 0.01). Peacekeeping missions have
similarly concentrated benefits, such that equity values predict negative perceptions of America taking
the primary role in bearing the costs. A two standard deviation increase in equity (sd = 0.19) is
associated with a 0.18-unit increase in judgments that it is a problem for the U.S. if other states to not
contribute to a mission.
Equality plays a much smaller role compared to equity. The coefficient for equality is statistically
16George W. Bush, 2000. “October 11, 2000 Debate Transcript.” Transcript available athttps://www.debates.org/voter-education/debate-transcripts/october-11-2000-debate-transcript/.
15
significant in Model 5 — the environment vignette (p < 0.1), but the effect is substantively small. A
move from the minimum to the maximum predicts a 0.06 increase in the dependent measure, just a
fraction of a step on the 7-point scale. This is striking given the attention paid to fairness as primarily
a value that promotes individual rights in both political science scholarship (Jost, Federico and Napier,
2009) and psychological research on moral foundations (Graham, Haidt and Nosek, 2009).17
Only burden-sharing in security alliances — Models 1 and 2 — seems to draw significant opposition
on the basis of both equity and equality. A move from the minimum to the maximum on the equity
dimension of fairness predicts a 0.325-unit increase in reporting that disproportionate defense spending
is a problem for the United States. Allies contribute relatively less as a share of their GDP, and yet
benefit greatly from U.S. protection. At the same time, there are disparate outcomes to consider if the
U.S. is not equally secure as a consequence of their alliances. The positive coefficient on equality bears
out this relationship.
Models 2, 4, 6, and 8 include three foreign policy orientations frequently identified as the main orga-
nizing structures for foreign policy attitudes (Gravelle, Reifler and Scotto, 2017) alongside demographic
controls. Of the three orientations, only isolationism has a consistent and statistically significant re-
lationship with the dependent variables. Isolationist participants report that it is problematic for the
U.S. to act as the primary contributor to resolving any of the four international problems. Cooperative
and militant internationalism have little bearing on how participants judge America’s unfair burdens.
Importantly, the effects of equity are substantively larger than isolationism. For example, a shift from
the minimum to the maximum on isolationism predicts a 0.18-unit increase in the assessment that it is
a problem for the U.S. to contribute all resources to peacekeeping mission. In contrast, a shift from the
minimum to maximum on equity is associated with a 0.37-unit increase in the DV — twice the size of
isolationism’s effect.
The results presented in Table 3 demonstrate that Americans who care about equity express sig-
nificantly greater concern about foreign policy scenarios in which the U.S. contributes more than they
benefit. This pattern holds across four diverse policy domains and when we control for a range of
foreign policy orientations and demographic variables. The effect of equality, however, is less consistent
— the moral commitment to equal outcomes held by many left-leaning Americans predicts attitudes
toward defense spending but not other forms of cooperation. The fact that equity continues to exert
17Haidt (2012, 196) differentiates between equality of outcomes — a more liberal idea of fairness — and proportionality,noting that their original formulation of moral foundations theory had focused primarily on the former. Separating equalityfrom proportionality paved the way for a new conceptualization in subsequent research, where fairness as equality iscontained in the liberty/oppression foundation and the fairness foundation captures proportionality (Haidt, 2012, 210).
16
Tab
le3:
Am
eric
ans
hig
hin
equ
ity
pre
fere
nce
ste
nd
tob
em
ore
conce
rned
ab
ou
tb
urd
ensh
ari
ng
vio
lati
on
sin
US
fore
ign
poli
cy
Dom
ain
of
bu
rden
shari
ng
vio
lati
on
:
Def
ense
Bu
dget
sP
eace
kee
pin
gE
nvir
on
men
tF
ore
ign
Aid
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Equ
ity
0.3
25∗∗∗
0.2
78∗∗∗
0.4
70∗∗∗
0.3
72∗∗∗
0.4
54∗∗∗
0.3
92∗∗∗
0.4
69∗∗∗
0.3
80∗∗∗
(0.0
44)
(0.0
47)
(0.0
41)
(0.0
43)
(0.0
41)
(0.0
44)
(0.0
41)
(0.0
44)
Equ
ality
0.1
21∗∗∗
0.1
20∗∗∗
−0.0
06
−0.0
10
0.0
55∗
0.0
59∗
0.0
23
0.0
28
(0.0
33)
(0.0
35)
(0.0
30)
(0.0
32)
(0.0
30)
(0.0
32)
(0.0
31)
(0.0
32)
Coop
erati
ve
Int.
−0.0
74
0.0
35
−0.0
50
0.0
15
(0.0
47)
(0.0
42)
(0.0
43)
(0.0
43)
Milit
ant
Int.
0.0
51
−0.0
20
0.0
69
0.1
20∗∗∗
(0.0
47)
(0.0
42)
(0.0
43)
(0.0
43)
Isola
tion
ism
0.1
78∗∗∗
0.1
76∗∗∗
0.1
57∗∗∗
0.1
72∗∗∗
(0.0
39)
(0.0
35)
(0.0
36)
(0.0
36)
Male
−0.0
25
−0.0
40∗∗
−0.0
28∗
−0.0
28∗
(0.0
17)
(0.0
16)
(0.0
16)
(0.0
16)
Wh
ite
−0.0
0003
0.0
17
0.0
19
0.0
39∗∗
(0.0
21)
(0.0
19)
(0.0
20)
(0.0
20)
Age
0.0
01∗∗
0.0
03∗∗∗
0.0
02∗∗∗
0.0
02∗∗∗
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
Som
eC
olleg
e0.0
11
−0.0
20
0.0
23
0.0
26
(0.0
24)
(0.0
22)
(0.0
23)
(0.0
23)
Colleg
e−
0.0
12
−0.0
21
−0.0
05
0.0
19
(0.0
23)
(0.0
21)
(0.0
21)
(0.0
21)
Post
-Gra
du
ate
−0.0
08
−0.0
13
0.0
10
0.0
03
(0.0
31)
(0.0
28)
(0.0
29)
(0.0
29)
Inco
me:
$30-6
0,0
00
0.0
11
−0.0
02
0.0
01
0.0
11
(0.0
22)
(0.0
20)
(0.0
20)
(0.0
20)
Inco
me:
$60-1
00,0
00
0.0
07
−0.0
03
0.0
11
0.0
12
(0.0
23)
(0.0
21)
(0.0
21)
(0.0
21)
Inco
me:
>$100,0
00
0.0
29
−0.0
001
0.0
06
0.0
20
(0.0
26)
(0.0
24)
(0.0
24)
(0.0
24)
Reg
ion
Contr
ols
XX
XX
Con
stant
0.3
16∗∗∗
0.1
97∗∗∗
0.3
61∗∗∗
0.2
16∗∗∗
0.2
97∗∗∗
0.1
55∗∗
0.3
23∗∗∗
0.0
77
(0.0
38)
(0.0
66)
(0.0
35)
(0.0
60)
(0.0
35)
(0.0
61)
(0.0
35)
(0.0
61)
N1,0
03
999
1,0
03
999
1,0
03
999
1,0
02
998
Ad
just
edR
20.0
74
0.0
96
0.1
19
0.1
69
0.1
22
0.1
50
0.1
20
0.1
59
∗p<
.1;∗∗
p<
.05;∗∗∗p<
.01
Note
:T
ab
led
isp
lays
OL
Sco
effici
ents
,st
an
dard
erro
rsin
pare
nth
eses
.H
igh
ervalu
esin
dic
ate
part
icip
ants
rate
dth
esc
enari
oa
big
ger
pro
ble
mfo
rth
eU
S.
All
conti
nu
ou
svari
ab
les,
exce
pt
age,
have
bee
nre
scale
dfr
om
0to
1.
Ref
eren
ceca
tegori
esare
¡$30,0
00,
Hig
hsc
hool
or
less
,an
dW
est.
17
a substantively large effect on burden sharing attitudes even controlling for foreign policy orientations
like isolationism is important: it shows that there is a class of individuals who are not necessarily pre-
disposed to want the US to stay home and focus more on its own problems, but who are aggravated by
other countries not pulling their weight. And, the fact that we fielded this study in 2014, well before
Donald Trump amplified concerns about “unfair” trade deals and alliance arrangements, shows that
our results are unlikely to be the artifact of elite cues from the Trump White House.
In Appendix §2.1, we conduct a variety of additional robustness checks, showing that the stronger
findings for equity in our results are not due to asymmetries in scale length, and that the pattern of
results we report here are not merely an artifact of broader ideological or partisan differences. Equity
retains its substantively large and statistically significant effect on burden sharing concerns even when
controlling for partisanship and political ideology. Together, these findings offer another example of
how personal values spill over into foreign policy preferences (Rathbun et al., 2016): the more individ-
uals are offended by equity violations in their daily lives, the more concerned they are about burden
sharing in foreign policy, whether in terms of allies’ defense budgets or foreign aid, peacekeeping or the
environment.
2.3 Equity and equality shape foreign policy attitudes beyond burden-sharing
Our results show that equity values — but not equality values — are important for understanding
divergent reactions in the United States toward burden-sharing in foreign policy. Given the extent
to which burden-sharing issues feature prominently in contemporary foreign policy debates, but are
somewhat understudied in the academic literature on public opinion about foreign policy, these findings
make an important contribution. However, one concern about this analysis is that the outcome variables
uniquely implicate equity — the DV question itself makes U.S. costs salient by asking people whether
the situation is a problem for the U.S. — and thereby mask the important role of equality. In this
section, we therefore measure support for a broader set of concrete foreign policy proposals and probe
the conditions under which the two faces of fairness complement or contradict each other.
We measure support for three policy proposals, each of which implicates a potential mismatch
between U.S. contributions and the policy’s primary beneficiaries abroad in a different domain of foreign
affairs: international political economy, international cooperation, and defense. The first proposal asks
participants whether they support or oppose decreasing limits on imports of foreign-made products, and
18
signing more free trade agreements like NAFTA (Free Trade).18 We expect that equality will predict
support for this proposal, because free trade agreements level the playing field for foreign companies and
economies by allowing them to compete for American business. Regional trade agreements like NAFTA
can produce larger gains for Mexico than for the U.S. (Caliendo and Parro, 2015), but Americans who
value equality view closing the economic gap between developed and developing states as a desirable
end.19 Equity-minded Americans will instead oppose free trade agreements that could damage some
sectors of the U.S. economy and improve trading partners’ overall welfare to a greater extent than
America’s. In turn, equity will predict less support the Free Trade proposal — our values shape
whether we paint free trade as fair trade.
The second proposal asks participants whether they support an arms control treaty that would
reduce both US and Russian nuclear arsenals (Arms Control). We expect a positive relationship between
equality and support for this proposed treaty, which offers clear global benefits: the potential for
consequential accidents declines alongside the number of nuclear weapons (Sagan, 1995). Equity-minded
participants, however, will attend to the costs associated with arms control. The U.S. and Russia will
each witness a small decrease in their overall power. Some people who value equity might think that
this is a “fair” price to pay for the security the U.S. will gain from a smaller Russian stockpile. But the
proposal lacks information about the U.S. and Russia’s respective starting positions, which could arouse
relative gains concerns among those who view equal reductions as inequitable and drive opposition.
These concerns muddy our expectations about how equity relates to support for arms control.
Finally, we ask participants if they support increasing military spending to allow the US to better
solve international problems (Military Spending). Again, this proposal presents clear global benefits
— to the extent that deploying the U.S. military augurs peace, equality-minded Americans will be
eager to invest. Viewed in a different light, though, the proposal requires the U.S. to invest scarce
economic resources into a program that only benefits national security via indirect routes: “Solving
global problems” can bolster U.S. security in the long run, but the description emphasizes those benefits
that accrue to the world. Much like our burden-sharing scenarios, this proposal asks the U.S. to pay
while others reap the rewards — driving equity-based opposition.
Figure 2 presents estimates from a series of OLS models that estimate the relationship between eq-
18This survey was part of a larger project, and each of these proposals had a differently framed counterpart that wasrandomly assigned to participants. The experimental manipulation restricts the sample size, but otherwise does not affectthe analyses in this section.
19Indeed, Mutz and Kim (2017, 842) find that holding gains for the U.S. constant, people with low-social dominanceorientation — who value group equality — report greater support for trade agreements that present a win-win situation,where the U.S. trading partner also gains jobs from the agreement.
19
uity, equality, and support for each policy proposal. Each model also controls for militant/cooperative
internationalism and isolationism — coefficients included for comparison — and demographic vari-
ables.20 The results point to five key conclusions. First, consistent with H3, the two faces of fairness
are not always complementary. Whereas equality values are associated with more support for military
spending and free trade, equity-minded Americans would rather not sacrifice resources if U.S. contri-
butions outstrip whatever benefits the U.S. stands to gain. Rather than increase foreign competition
for Americans’ business, equity-minded Americans prefer to reserve their home market for domestic
producers. In the case of an arms control agreement, equality increases support for a proposal that
would limit nuclear arsenals while equity is unrelated to arms control preferences, mirroring the pattern
we observed for 3 of the 4 burden-sharing items.
Second, these results underscore our argument that fairness is not inherently prosocial: equity
values discourage international economic agreements, for example. Moreover, we find evidence that
concerns about equality can increase demands for military spending. Focused on outcomes, equality-
minded respondents see promise in bolstering the U.S. defense budget if it might help solve international
problems. Defense spending could therefore draw support from a coalition of militant internationalists
and the equality-minded Americans who otherwise eschew hawkish politics.
Third, although research on foreign policy attitudes tends to divorce security from economics and
assume that international trade attitudes follow a different logic than other foreign policy domains, we
find evidence that equity and equality can shape public support for NATO and NAFTA alike. This
finding complements a growing body of work on the relationship between values and public opinion
about international economic policies (Kaltenthaler and Miller, 2013; Rathbun, 2016; Rathbun, Powers
and Anders, 2019).
Fourth, it illustrates the importance of studying core values in foreign policy preferences more
generally: even though these foreign policy proposals are arguably further removed from fairness con-
siderations than the burden sharing vignettes are, a set of Wald tests find that we experience a sig-
nificant reduction in model fit when we drop equity and equality from the two models where we have
the strongest theoretical expectations (trade, and military spending), even when controlling for a wide
range of other demographic characteristics and foreign policy orientations. Understanding respondents’
differential concerns about each face of fairness thus systematically enhances our understanding of their
foreign policy preferences.
20See Appendix §2.5 for the results in tabular form, which also suggest we should not be concerned about post-treatmentbias, in that our results hold without these covariates as well.
20
Figure 2: Equity, Equality, and Policy Proposals
Arms Control Free Trade Military Spending
-0.5 0.0 0.5 -0.5 0.0 0.5 -0.5 0.0 0.5
Iso
CI
MI
Equity
Equality
Effect size
Note: N=515, 491, 477, respectively. Figures display OLS coefficient estimates and 95% confidence intervals for equality,equity, militant internationalism (MI), cooperative internationalism (CI) and isolationism from models that also includeadditional demographic controls. To facilitate direct comparability, each variable has been rescaled to range from 0 to 1,
and higher values on the DV indicate support for the policy.
Fifth, the proposals that we focus on here each represent policy areas that have been characterized
by pronounced gender gaps: in the United States, women have historically been less supportive of
free trade (Mansfield, Mutz and Silver, 2015), more supportive of arms control (Silverman and Kumka,
1987), and less supportive of defense spending (Eichenberg and Stoll, 2012) than men. Despite extensive
documentation of gender gaps in foreign policy public opinion, “the reason for these differences remains
elusive” (Lizotte, 2019, 126). Given the apparent gender differences in fairness commitments, variation
in concerns about equality could be one explanation for these gaps.
Results from the OLS models, presented in Appendix §2.5, reveal no effect of gender on support
for the three policy proposals. Our interest does not lie in the total effect, however, but in whether
gender has any indirect effects through values.21 We thus estimate a series of nonparametric causal
mediation models (Imai et al., 2011), in which the effect of gender on each of these foreign policy
issues is mediated by concerns about each type of fairness, while controlling for a host of demographic
characteristics. Although care should be taken in interpreting these results given potential confounders,
the results suggest that fairness concerns offer one potential explanation for gender differences in two
21Zhao, Lynch Jr and Chen (2010, 199) contend that it is not necessary to observe “a significant zero-order effectof X on Y... to establish mediation.” Rucker et al. (2011, 361-362) similarly “question the requirement that a total X→ Y effect be present before assessing mediation.” They summarize that “the lack of an effect... does not preclude thepossibility of observing indirect effects.” We therefore rely on the theoretical foundations provided in literature on gendergaps to probe indirect-only mediation despite the absence of a total effect (Rucker et al., 2011, 368).
21
of the three issues: arms control, and trade attitudes.
The average causal mediation effect (ACME) of male on support for arms control, through equality,
is -0.007 (-0.017, 0.00). To the extent that men are less committed to equality than women, they will
in turn be less supportive of arms control. The ACME is small, comprising about 4.5% of the total
effect, but statistically significant. We find similar evidence for indirect-only mediation on support for
free trade. The effect of male channeled through equality is significant and negative, accounting for
33.5% of the total effect. These results suggest that the gender gap in trade attitudes is explained
in part through women’s greater commitment to egalitarian outcomes. We find no evidence that the
relationship between gender and support for these three policy issues is mediated by equity. Importantly,
the absence of significant direct or total effects for gender implies one or more possible suppressors —
other, unmeasured mechanisms that push men toward arms control and free trade (Rucker et al., 2011).
Future research should account for equality values alongside other relevant factors like partisanship,
identity, and other prosocial values to offer a more complete understanding of gender gaps in foreign
policy attitudes.
3 Conclusion
In this article, we sought to contribute to the study of fairness in IR by reminding IR scholars that
fairness is multidimensional (Adams, 1965; Deutsch, 1975; Hochschild, 1981). Whereas the existing
literature on fairness in IR has focused almost exclusively on fairness as equality, we can also understand
fairness as equity. Because personal values spill over into the foreign policy domain (Rathbun et al.,
2016), both faces of fairness are important for understanding the contours of foreign policy preferences.
Although IR scholars typically associate fairness with cooperation, our results demonstrate that
equity values encourage opposition to security cooperation and public goods provision across several
contexts, due in particular to concerns about inadequate burden sharing. We therefore find evidence of a
new psychological microfoundation for isolationism, something existing scholarship has failed to uncover
(Kertzer et al., 2014). Moreover, we fielded our survey in 2014, before Donald Trump highlighted
inequities in the US alliance system, which suggests that our findings are not merely an artifact of
partisan cue-taking. Our results offer further support for the continued importance of core values and
moral judgments in shaping foreign policy preferences (Bayram, 2015; Rathbun et al., 2016; Kreps and
Maxey, 2018), and in studying fairness concerns as a variable rather than a constant. And although our
22
findings are consistent with bottom-up theories of public opinion in foreign policy (Kertzer and Zeitzoff,
2017), they also suggest important implications for theories of elite political behavior. Framing research
in American politics argues that political elites seek to mobilize and persuade voters using value-based
appeals that frame issues in terms of the values that resonate with their audience (Nelson, Clawson and
Oxley, 1997). The fact that Republicans and Democrats alike value equity suggest that it should be a
particularly potent way to frame foreign policy issues.
We also observe important differences in how Republicans and Democrats think about equality.
Our findings thus contribute to a growing literature on partisanship and ideology in foreign policy. IR
scholars have found that Republicans and Democrats tend to conduct systematically different types of
foreign policies, not just because each party’s base has a different set of interests, but because “right
parties have somewhat different values from left parties.” (Palmer, London and Regan, 2004, 1-24;
see also Rathbun, 2004; Clare, 2014; Bertoli, Dafoe and Trager, 2019). We observe differences in how
Democrats and Republicans conceptualize fairness — Democrats place greater value on equality than
Republicans do — that suggest a potential microfoundation for distinct partisan approaches to foreign
policy.
Our findings also relate to research on partisan or ideological differences in moral reasoning generally
(Koleva et al., 2012). Indeed, although our analysis focused on the two faces of fairness in foreign
policy, our findings inform research on public opinion about domestic issues — since the equity and
equality scales can be fruitfully applied in other contexts. As Hochschild (1981) noted nearly four
decades ago, many key debates in American political life involve competing conceptions of fairness.
Understanding individual differences in equity- and equality-based moral judgments can therefore enrich
our understanding of public opinion more broadly.
23
References
Adams, J Stacy. 1965. “Inequity in Social Exchange.” Advances in Experimental Social Psychology
2:267–299.
Albin, Cecilia and Daniel Druckman. 2012. “Equality Matters: Negotiating an End to Civil Wars.”
Journal of Conflict Resolution 56(2):155–182.
Baldwin, David Allen. 1993. Neorealism and neoliberalism: the contemporary debate. Columbia Univer-
sity Press.
Baum, Matthew A. and Philip B.K. Potter. 2015. War and Democratic Constraint: How The Public
Influences Foreign Policy. Princeton, NJ: Princeton University Press.
Bayram, A. Burcu. 2015. “What drives modern Diogenes? Individual values and cosmopolitan alle-
giance.” European Journal of International Relations 21(2):451–479.
Bechtel, Michael M, Jens Hainmueller and Yotam Margalit. 2017. “Policy design and domestic support
for international bailouts.” European Journal of Political Research 56(4):864–886.
Berinsky, Adam J. 2009. In Time of War: Understanding American Public Opinion from World War
II to Iraq. Chicago: University of Chicago Press.
Bernauer, Thomas, Robert Gampfer and Aya Kachi. 2014. “European unilateralism and involuntary
burden-sharing in global climate politics: A public opinion perspective from the other side.” European
Union Politics 15(1):132–151.
Bertoli, Andrew, Allan Dafoe and Robert F Trager. 2019. “Is There a War Party? Party Change, the
Left–Right Divide, and International Conflict.” Journal of Conflict Resolution 63(4):950–975.
Bolton, Gary E and Axel Ockenfels. 2000. “ERC: A theory of equity, reciprocity, and competition.”
American economic review 90(1):166–193.
Brooks, Deborah Jordan and Benjamin A. Valentino. 2011. “A War of One’s Own: Understanding the
Gender Gap in Support for War.” Public Opinion Quarterly 75(2):270–286.
Brooks, Stephen G. and William C. Wohlforth. 2016. America Abroad: Why the sole superpower should
not pull back from the world. New York, NY: Oxford University Press.
Brutger, Ryan and Brian C. Rathbun. 2019. “Fair Play? Equity and Equality in American Attitudes
Towards Trade.” Working paper.
Caliendo, Lorenzo and Fernando Parro. 2015. “Estimates of the Trade and Welfare Effects of NAFTA.”
The Review of Economic Studies 82(1):1–44.
Camerer, Colin F. 1997. “Progress in behavioral game theory.” Journal of economic perspectives
24
11(4):167–188.
Chilton, Adam S, Helen V Milner and Dustin Tingley. 2017. “Reciprocity and Public Opposition to
Foreign Direct Investment.” British Journal of Political Science pp. 1–25.
Chittick, William O., Keith R. Billingsley and Rick Travis. 1995. “A Three-Dimensional Model of
American Foreign Policy Beliefs.” International Studies Quarterly 39(3):313–331.
Chu, Jonathan A. 2019. “A Clash of Norms? How Reciprocity and International Humanitarian Law
affect American Opinion on the Treatment of POWs.” Journal of Conflict Resolution 63(5):1140–
1164.
Clare, Joe. 2014. “Hawks, Doves, and International Cooperation.” Journal of Conflict Resolution
58(7):1311–1337.
Clifford, Scott, Vijeth Iyengar, Roberto Cabeza and Walter Sinnott-Armstrong. 2015. “Moral founda-
tions vignettes: A standardized stimulus database of scenarios based on moral foundations theory.”
Behavior research methods 47(4):1178–1198.
Cohrs, J. Christopher, Barbara Moschner, Jurgen Maes and Sven Kielmann. 2005. “Personal Values
and Attitudes Toward War.” Peace and Conflict 11(3):293–312.
Converse, Philip E. 1964. The nature and origin of belief systems in mass publics. In Ideology and
Discontent, ed. David E. Apter. New York: Free Press pp. 206–261.
Cram, Laura, Adam Moore, Victor Olivieri and Felix Suessenbach. 2018. “Fair is Fair, or Is It? Terri-
torial Identity Triggers Influence Ultimatum Game Behavior.” Political Psychology 39(6):1233–1250.
Delli Carpini, Michael X. and Scott Keeter. 1996. What Americans Know About Politics and Why It
Matters. New Haven: Yale University Press.
DeScioli, Peter, Maxim Massenkoff, Alex Shaw, Michael Bang Petersen and Robert Kurzban. 2014.
“Equity or equality? Moral judgments follow the money.” Proceedings of the Royal Society of London
B: Biological Sciences 281(1797):20142112.
Deutsch, Morton. 1975. “Equity, Equality, and Need: What Determines Which Value Will Be Used as
the Basis of Distributive Justice?” Journal of Social Issues 31(3):137–149.
DeVoe, Sanford E and Sheena S Iyengar. 2010. “Medium of exchange matters: What’s fair for goods is
unfair for money.” Psychological Science 21(2):159–162.
Efrat, Asif and Abraham L Newman. 2016. “Deciding to Defer: The Importance of Fairness in Resolving
Transnational Jurisdictional Conflicts.” International Organization 70(02):409–441.
Eichenberg, Richard C. and Richard J. Stoll. 2012. “Gender Difference or Parallel Publics? The Dynam-
25
ics of Defense Spending Opinions in the United States, 1965–2007.” Journal of Conflict Resolution
56(2):331–348.
Fabrigar, Leandre R, Duane T Wegener, Robert C MacCallum and Erin J Strahan. 1999. “Evaluating
the use of exploratory factor analysis in psychological research.” Psychological methods 4(3):272.
Fehr, Ernst, Helen Bernhard and Bettina Rockenbach. 2008. “Egalitarianism in young children.” Nature
454(7208):1079–1083.
Fehr, Ernst and Klaus M Schmidt. 1999. “A theory of fairness, competition, and cooperation.” Quarterly
Journal of Economics 114(3):817–868.
Feldman, Stanley and John Zaller. 1992. “The political culture of ambivalence: Ideological responses
to the welfare state.” American Journal of Political Science pp. 268–307.
Fiske, Alan Page and Philip E. Tetlock. 1997. “Taboo Trade-offs: Reactions to Transactions That
Transgress the Spheres of Justice.” Political Psychology 18(2):255–297.
Fuhrmann, Matthew. 2019. “When Do Leaders Free-ride? Business Experience and Contributions to
Collective Defense.” American Journal of Political Science Forthcoming.
Goddard, Stacie E. 2006. “Uncommon ground: Indivisible territory and the politics of legitimacy.”
International Organization 60(01):35–68.
Goren, Paul, Harald Schoen, Jason Reifler, Thomas Scotto and William Chittick. 2016. “A unified
theory of value-based reasoning and US public opinion.” Political behavior 38(4):977–997.
Gottfried, Matthew S and Robert F Trager. 2016. “A Preference for War: How Fairness and Rhetoric
Influence Leadership Incentives in Crises.” International Studies Quarterly 60(2):243–257.
Graham, Jesse, Jonathan Haidt and Brian A. Nosek. 2009. “Liberals and Conservatives Rely on Different
Sets of Moral Foundations.” Journal of Personality and Social Psychology 96(5):1029–1046.
Gravelle, Timothy B, Jason Reifler and Thomas J Scotto. 2017. “The structure of foreign policy
attitudes in transatlantic perspective: Comparing the United States, United Kingdom, France and
Germany.” European Journal of Political Research 56(4):757–776.
Grieco, Joseph M. 1988. “Anarchy and the limits of cooperation: a realist critique of the newest liberal
institutionalism.” International Organization 42(3):485–507.
Guisinger, Alexandra and Elizabeth N. Saunders. 2017. “Mapping the Boundaries of Elite Cues: How
Elites Shape Mass Opinion Across International Issues.” International Studies Quarterly 61(2):425–
441.
Guth, Werner, Rolf Schmittberger and Bernd Schwarze. 1982. “An experimental analysis of ultimatum
26
bargaining.” Journal of economic behavior & organization 3(4):367–388.
Haidt, Jonathan. 2012. The Righteous Mind: Why good people are divided by politics and religion. New
York: Pantheon Books.
Haidt, Jonathan. 2013. “Of Freedom and Fairness.” Democracy: A Journal of Ideas 28.
URL: https://democracyjournal.org/magazine/28/of-freedom-and-fairness/
Hochschild, Jennifer L. 1981. What’s Fair?: American Beliefs About Distributive Justice. Cambridge,
MA: Harvard University Press.
Hurwitz, Jon and Mark Peffley. 1987. “How are foreign policy attitudes structured? A hierarchical
model.” The American Political Science Review pp. 1099–1120.
Imai, Kosuke, Luke Keele, Dustin H. Tingley and Teppei Yamamoto. 2011. “Unpacking the Black Box
of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies.”
American Political Science Review 105(4):765–789.
Iyer, Ravi. 2010. “Democrats and Republicans agree that Justice & Fairness are about Equity, not
Equality or Impartiality.”.
URL: http://www.polipsych.com/2010/02/20/democrats-and-republicans-agree-that-justice-fairness-
are-about-equity-not-equality-or-impartiality/
Jennings, M. Kent. 1991. “Thinking about Social Injustice.” Political Psychology 12(2):187–204.
Jost, John T, Christopher M Federico and Jaime L Napier. 2009. “Political ideology: Its structure,
functions, and elective affinities.” Annual review of psychology 60:307–337.
Kaltenthaler, Karl and William J Miller. 2013. “Social psychology and public support for trade liber-
alization.” International Studies Quarterly 57(4):784–790.
Kapstein, Ethan B. 2008. “Fairness Considerations in World Politics: Lessons From International Trade
Negotiations.” Political Science Quarterly 123(2):229–245.
Kertzer, Joshua D and Brian C Rathbun. 2015. “Fair is Fair: Social Preferences and Reciprocity in
International Politics.” World Politics 67(04):613–655.
Kertzer, Joshua D., Kathleen Powers, Brian C. Rathbun and Ravi Iyer. 2014. “Do moral values shape
foreign policy preferences?” Journal of Politics 76(3):825–840.
Kertzer, Joshua D. and Thomas Zeitzoff. 2017. “A Bottom-Up Theory of Public Opinion about Foreign
Policy.” American Journal of Political Science 61(3):543–558.
Knoch, Daria, Alvaro Pascual-Leone, Kaspar Meyer, Valerie Treyer and Ernst Fehr. 2006. “Diminishing
reciprocal fairness by disrupting the right prefrontal cortex.” Science 314(5800):829–832.
27
Koleva, Spassena P, Jesse Graham, Ravi Iyer, Peter H Ditto and Jonathan Haidt. 2012. “Tracing
the Threads: How Five Moral Concerns (Especially Purity) Help Explain Culture War Attitudes.”
Journal of Research in Personality 46(2):184–194.
Kreps, Sarah and Sarah Maxey. 2018. “Mechanisms of Morality: Sources of Support for Humanitarian
Intervention.” Journal of Conflict Resolution 62(8):1814–1842.
Liberman, Peter. 2006. “An eye for an eye: Public support for war against evildoers.” International
Organization 60(3):687.
Lizotte, Mary-Kate. 2019. “Investigating the origins of the gender gap in support for war.” Political
Studies Review 17(6):124–135.
Lu, Xiaobo, Kenneth Scheve and Matthew J Slaughter. 2012. “Inequity aversion and the international
distribution of trade protection.” American Journal of Political Science 56(3):638–654.
Mansfield, Edward D., Diana C. Mutz and Laura R. Silver. 2015. “Men, Women, Trade, and Free
Markets.” International Studies Quarterly 59(2):303–315.
Meegan, Dan. 2019. America the Fair: Using Brain Science to Create a More Just Nation. Cornell
University Press.
Meindl, Peter, Ravi Iyer and Jesse Graham. 2019. “Distributive Justice Beliefs are Guided by Whether
People Think the Ultimate Goal of Society is Well-Being or Power.” Basic and Applied Social Psy-
chology 41(6):359–385.
Mutz, Diana C and Eunji Kim. 2017. “The impact of in-group favoritism on trade preferences.” Inter-
national Organization 71(4):827–850.
Nelson, Thomas E., Rosalee A. Clawson and Zoe M. Oxley. 1997. “Media Framing of a Civil Liberties
Conflict and Its Effect on Tolerance.” American Political Science Review 91(3):567–583.
O’Dwyer, Emma and Ahmet Coymak. 2019. “Basic Human Values and Their Contexts: A Multilevel
Analysis of Support for the Use of Armed Drones in the United States, United Kingdom, and Turkey.”
Political Psychology Forthcoming.
Oneal, John R. 1990. “The theory of collective action and burden sharing in NATO.” International
organization 44(03):379–402.
Ostrom, Elinor. 1998. “A behavioral approach to the rational choice theory of collective action: Pres-
idential address, American Political Science Association, 1997.” American political science review
92(1):1–22.
Palmer, Glenn, Tamar R. London and Patrick M. Regan. 2004. “What’s Stopping You?: The Sources of
28
Political Constraints on International Conflict Behavior in Parliamentary Democracies.” International
Interactions 30(1):1–24.
Prather, Lauren. 2014. “Values at the water’s edge: Social welfare values and foreign aid.”.
URL: http://www.laurenprather.org/uploads/2/5/2/3/25239175/prather -
values at the waters edge.pdf
Rai, Tage Shakti and Alan Page Fiske. 2011. “Moral psychology is relationship regulation: moral
motives for unity, hierarchy, equality, and proportionality.” Psychological review 118(1):57.
Rasinski, Kenneth A. 1987. “What’s fair is fair—Or is it? Value differences underlying public views
about social justice.” Journal of personality and social Psychology 53(1):201.
Rathbun, Brian C. 2004. Partisan Interventions: European Party Politics and Peace Enforcement in
the Balkans. Ithaca, NY: Cornell University Press.
Rathbun, Brian C. 2016. “Wedges and widgets: liberalism, libertarianism, and the trade attitudes of
the American mass public and elites.” Foreign Policy Analysis 12(1):85–108.
Rathbun, Brian C, Joshua D Kertzer, Jason Reifler, Paul Goren and Thomas J Scotto. 2016. “Tak-
ing foreign policy personally: Personal values and foreign policy attitudes.” International Studies
Quarterly 60(1):124–137.
Rathbun, Brian C, Kathleen E Powers and Therese Anders. 2019. “Moral Hazard: German Public
Opinion on the Greek Debt Crisis.” Political Psychology Forthcoming.
Rathbun, Brian C. and Rachel Stein. Forthcoming. “Greater Goods: Morality and Attitudes toward
the Use of Nuclear Weapons.” Journal of Conflict Resolution .
Ringius, Lasse, Asbjøorn Torvanger and Arild Underdal. 2002. “Burden sharing and fairness principles
in international climate policy.” International Environmental Agreements 2(1):1–22.
Rucker, Derek D, Kristopher J Preacher, Zakary L Tormala and Richard E Petty. 2011. “Mediation
analysis in social psychology: Current practices and new recommendations.” Social and Personality
Psychology Compass 5(6):359–371.
Sagan, Scott Douglas. 1995. The limits of safety: Organizations, accidents, and nuclear weapons.
Princeton University Press.
Scott, John T, Richard E Matland, Philip A Michelbach and Brian H Bornstein. 2001. “Just deserts: an
experimental study of distributive justice norms.” American Journal of Political Science pp. 749–767.
Silverman, Jane M and Donald S Kumka. 1987. “Gender differences in attitudes toward nuclear war
and disarmament.” Sex Roles 16(3-4):189–203.
29
Starmans, Christina, Mark Sheskin and Paul Bloom. 2017. “Why people prefer unequal societies.”
Nature Human Behaviour 1(4):1–7.
Thaler, Richard H. 1988. “Anomalies: The ultimatum game.” Journal of Economic Perspectives
2(4):195–206.
Trump, Kris-Stella. 2013. “The Status Quo and Perceptions of Fairness: How Income Inequality Influ-
ences Public Opinion.” Doctoral dissertation.
Trump, Kris-Stella. forthcoming. “When and Why is Economic Inequality Seen as Fair.” Current
Opinion in Behavioral Science .
Wagstaff, Graham F. 1994. “Equity, equality, and need: Three principles of justice or one? An analysis
of “equity as desert”.” Current Psychology 13(2):138–152.
Wilson, Rick K. 2011. “The contribution of behavioral economics to political science.” Annual Review
of Political Science 14:201–223.
Wittkopf, Eugene R. 1990. Faces of internationalism: Public Opinion and American foreign policy.
Duke University Press.
Zhao, Xinshu, John G Lynch Jr and Qimei Chen. 2010. “Reconsidering Baron and Kenny: Myths and
truths about mediation analysis.” Journal of Consumer Research 37(2):197–206.
30
What’s Fair in International Politics?Equity, Equality, and Foreign Policy Attitudes
Supplementary Appendix
1 Survey instrumentation and sample information
1.1 Foreign Policy Orientations
To measure foreign policy orientations, we relied on a series of widely used items that comprise scales forcooperative internationalism, militant internationalism, and isolationism (e.g., Kertzer et al., 2014; Gravelle,Reifler and Scotto, 2017). Participants reported whether they agree or disagree with each item on a scalefrom 1 (“Strongly agree”) to 5 (“Strongly disagree”). Responses were reverse-coded for analysis, such thathigher scores indicate agreement.
Cooperative Internationalism
• The United States needs to cooperate more with the United Nations.
• It is essential for the United States to work with other nations to solve problems such as overpopulation,hunger, and pollution.
• Promoting and defending human rights in other countries is of utmost importance.
• It is important for countries to comply with international law.
Militant Internationalism
• The United States should take all steps including the use of force to prevent aggression by any expan-sionist power.
• Going to war is unfortunate but sometimes the only solution to international problems.
• The United States must demonstrate its resolve so that others will not take advantage of it.
Isolationism
• The US should mind its own business internationally and let other countries get along the best theycan on their own.
• America’s conception of its leadership role in the world must be scaled down.
• We should not think so much in international terms but concentrate more on our own national problems.
1.2 Sample Characteristics
Table 1 compares key demographics from the American population, based on data from the U.S. Censuswith the demographics of our SSI sample. Our sample closely approximates U.S. census figures on keyvariables like sex, age, race, and region. Highly educated Americans — those with at least some college —are slightly over-represented in our data compared to their counterparts who have achieved a high schooleducation or less. To account for this disparity in education categories, we use entropy balancing in Stata toreweight the data such that it more closely matches the distribution of educational attainment in the U.S.(Hainmueller, 2012). Table 2 compares the unweighted models in columns 1, 3, 5, and 7 to the weightedestimates. Including these weights does not change the substantive results; equity is positively related toeach of the dependent variables.
Table 1: Sample Characteristics, N=1,004
SSI Sample Adult PopulationFemale 0.513 0.508Male 0.487 0.49218-24 0.066 0.13025-44 0.363 0.35045-64 0.395 0.34765+ 0.176 0.171High School or less 0.198 0.437Some college/university 0.263 0.196College/university 0.407 0.271Grad/Prof school 0.132 0.096
1.3 Scale construction
To measure attitudes towards equality and equity, we borrow from Clifford et al. (2015), Iyer (2010), andMeindl, Iyer and Graham (2019) in administering a series of fairness vignettes. In addition to the sevenvignettes listed in the main text, respondents were also presented with two distractor items to minimizeacquiescence bias — “A boy makes fun of his brother for being dumped by his girlfriend” and “A teenagerthrows rocks at cows that are grazing in the local pasture” — and two additional fairness vignettes (“Acompany bonus is given without considering the relative contributions of each employee”, and “A bonus isgiven to a work team for good performance and the money is not divided equally.”) Respondents noted theselast two items were unclear and open to interpretation (perhaps the bonus was not divided equally becausedifferent members of the team contributed more, for example), and factor analysis found that these twoadditional items cross-loaded weakly on each of the two factors, so following Osborne and Costello (2009),we thus dropped the two cross-loading items to achieve a simple structure that more cleanly represents theconstructs of interest. As noted in the main text, fit statistics further suggest this two-factor solution fitsthe data well, with a RMSEA value below 0.1 and TLI that approaches 0.95.
2
Table 2: What’s Fair in U.S. Foreign Policy?
Defense Spending Peacekeeping Environment Foreign Aid
(1) (2) (3) (4) (5) (6) (7) (8)
Equity 0.278∗∗∗ 0.285∗∗∗ 0.372∗∗∗ 0.381∗∗∗ 0.392∗∗∗ 0.401∗∗∗ 0.380∗∗∗ 0.343∗∗∗
(0.047) (0.047) (0.043) (0.042) (0.044) (0.043) (0.044) (0.044)Equality 0.120∗∗∗ 0.093∗∗∗ −0.010 0.002 0.059∗ 0.053∗ 0.028 0.039
(0.035) (0.034) (0.032) (0.030) (0.032) (0.031) (0.032) (0.031)Cooperative Int. −0.074 −0.044 0.035 0.008 −0.050 −0.061 0.015 0.048
(0.047) (0.046) (0.042) (0.041) (0.043) (0.042) (0.043) (0.042)Militant Int. 0.051 0.077 −0.020 0.023 0.069 0.109∗∗ 0.120∗∗∗ 0.160∗∗∗
(0.047) (0.048) (0.042) (0.042) (0.043) (0.043) (0.043) (0.044)Isolationism 0.178∗∗∗ 0.159∗∗∗ 0.176∗∗∗ 0.153∗∗∗ 0.157∗∗∗ 0.131∗∗∗ 0.172∗∗∗ 0.145∗∗∗
(0.039) (0.039) (0.035) (0.035) (0.036) (0.036) (0.036) (0.036)Controls X X X XConstant 0.197∗∗∗ 0.218∗∗∗ 0.216∗∗∗ 0.220∗∗∗ 0.155∗∗ 0.171∗∗∗ 0.077 0.098∗
(0.066) (0.063) (0.060) (0.056) (0.061) (0.057) (0.061) (0.058)N 999 999 999 999 999 999 998 998R2 0.111 0.119 0.183 0.208 0.164 0.184 0.173 0.174
∗p < .1; ∗∗p < .05; ∗∗∗p < .01Note: Table displays OLS coefficients, standard errors in parentheses. Higher values indicate partici-pants rated the scenario a bigger problem for the U.S. All continuous variables, except age, have beenrescaled from 0 to 1. Reference categories are ¡$30,000, High school or less, and West. Models 2, 4, 6,and 8 include survey weights that match U.S. census parameters for education.
3
2 Supplementary analysis
2.1 Downsampling analysis: findings not due to asymmetries in scale length
One of our key findings in the manuscript is that equity is consistently a stronger predictor of foreign policyattitudes than equality, consistent with equity violations themselves feeling morally worse than equalityviolations. At the same time, because our equality and equity scores come from a factor analysis wherethere are more items measuring equity than measuring equality, one potential interpretation is that theseresults are due to asymmetries in scale length leading to a higher-quality measure of equity than equality,since measurement error tends to decrease with the number of items in a scale (Ansolabehere, Rodden andSnyder, 2008).
Figure 1: Equity trumps equality even with fewer measures
Equality Equity
Defense S
pendingEnvironm
entForeign A
idPeacekeeping
1 2 FactorScore
1 2 FactorScore
0.00.10.20.30.4
0.00.10.20.30.4
0.00.10.20.30.4
0.00.10.20.30.4
Number of indicators
Effe
ct s
ize
Note: Figure 1 shows the stronger findings for equity versus equality are not due to asymmetries in scale length. The figurepresents coefficient estimates from a series of regression models of each of our foreign policy dependent variables on each typeof fairness, varying the construction of each fairness measure: either using a single item, an additive score of two items, or thefactor score used in the main text. Even when we measure equity and equality with the same number of items, the effect of
equity is always stronger than that of equality.
Although we use factor scoring in the main analysis precisely to reduce measurement error and avoidthis concern, we nonetheless test this possibility in Figure 1, which re-estimates the relationship between
4
equality, equity, and each of our four foreign policy dependent variables (defense spending, environment,foreign aid, and peacekeeping), but varying the number of items used to construct each fairness measure.The left-most set of estimates in each panel depict the coefficient estimate from a linear regression of eachforeign policy attitude on each fairness measure, using only a single item measure for each type of fairness(thus, there are
(21
)estimates for equality, and
(51
)estimates for equity, for each foreign policy attitude).
The next set of estimates in each panel use two-item additive scores (such that there are(22
)estimates for
equality and(52
)estimates for equity for each foreign policy attitude). The final estimate in each panel is
the factor score used in the main text.The analysis reveals two key points. First, on average, as we increase the number of items, the strength
of the effect increases; the same is true as we shift to factor scores. Second, and importantly, even whenwe measure equity and equality with the same number of items, the effect of equity is always stronger thanthat of equality; indeed, for three of the four dependent variables, the single-item measure of equity with theweakest effect still has a stronger effect than even the factor-score measure of equality. Thus, the divergenteffects we report in the main text cannot be attributed to asymmetries in scale length.
2.2 Additive Scales
Table 3 replicates the analyses reported in Table 3 from the main manuscript, using additive scales forequity and equality rather than factor scores. The equity scale comprises scores from the five equity itemsadded together, and the equality scale includes the two equality items. Both are rescaled from 0 to 1. Theestimates reported in Table 3 are consistent with the factor score results: equity values shape support forthese important foreign policy problems, while equality predicts responses to the defense spending vignette.
2.3 Multiple-Group Factor Analysis
We also test whether a pooled model based on the full sample may mask structural heterogeneity amongdifferent segments of the public. Not only do Republicans and Democrats tend to hold different values (Haidt,2012), but they may structure their values in distinct ways. To assess whether Democrats and Republicansorganize their views about fairness differently, we divide the sample by party identification. The threesubsamples include 369 Republicans and Republican leaners, 481 Democrats and Democrat leaners, and 155self-identified independents.
We display the results of principal axis factor analyses with oblimin rotations in Table 4. Using thesame seven vignettes that revealed a simple structure for the full sample, we find consistent support for atwo factor solution among Republicans and Democrats. Five equity items load strongly on the first factor,while two equality measures comprise the second factor. High TLI and low RMSEA values reveal that thestructure presents a good fit for the data in both partisan subsamples. Among independents, we find someevidence for latent equity and equality values, though fit statistics suggest that this structure may not be asgood a fit for the data.
While partisan Americans similarly distinguish these two facets of fairness, our factor analysis revealsone important difference: equity and equality are correlated among Democrats, but not their Republicancounterparts. We employ an oblique rotation on the pattern matrix that allows the two latent factors tocovary, and find a moderate correlation (r = 0.3) between equity and equality in the Democratic subsample.Democrats who consider violations of equity to be morally wrong tend to also find fault in miscarriages ofequality. In contrast Republicans who value equity do not necessarily care about equality — the two latentfactors are weakly correlated (r = 0.01) in the Republican subgroup.
2.4 Party Identification, Ideology and Foreign Policy
The results presented in the manuscript demonstrate that equity predicts foreign policy attitudes, and modelscontrol for a theoretically-motivated panel of important factors like foreign policy orientations and demo-graphic characteristics. In the manuscript, we demonstrate that equity values have cross-partisan support —liberals, conservatives, Republicans, and Democrats agree that fairness depends on the relationship between
5
Tab
le3:
Wh
at’
sF
air
inU
.S.
Fore
ign
Poli
cy?
(Ad
dit
ive
Sca
les)
Def
ense
Sp
end
ing
Pea
cekee
pin
gE
nvir
on
men
tF
ore
ign
Aid
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Equ
ity
0.36
0∗∗∗
0.3
16∗∗
∗0.4
99∗∗
∗0.4
02∗∗
∗0.4
85∗
∗∗0.4
28∗
∗∗0.4
91∗∗
∗0.4
05∗∗
∗
(0.0
44)
(0.0
48)
(0.0
40)
(0.0
43)
(0.0
41)
(0.0
44)
(0.0
41)
(0.0
44)
Equ
alit
y0.
079∗∗
∗0.0
78∗∗
∗−
0.0
34
−0.0
33
0.0
20
0.0
24
−0.0
03
0.0
04
(0.0
28)
(0.0
30)
(0.0
26)
(0.0
27)
(0.0
26)
(0.0
28)
(0.0
26)
(0.0
28)
Coop
erat
ive
Int.
−0.0
72
0.0
36
−0.0
48
0.0
16
(0.0
46)
(0.0
42)
(0.0
43)
(0.0
43)
Mil
itan
tIn
t.0.0
46
−0.0
24
0.0
64
0.1
17∗∗
∗
(0.0
47)
(0.0
42)
(0.0
43)
(0.0
43)
Isol
atio
nis
m0.1
78∗∗
∗0.1
75∗∗
∗0.1
57∗
∗∗0.1
71∗∗
∗
(0.0
39)
(0.0
35)
(0.0
35)
(0.0
36)
Mal
e−
0.0
24
−0.0
39∗∗
−0.0
27∗
−0.0
28∗
(0.0
17)
(0.0
16)
(0.0
16)
(0.0
16)
Wh
ite
0.0
005
0.0
19
0.0
21
0.0
41∗∗
(0.0
21)
(0.0
19)
(0.0
19)
(0.0
20)
Age
0.0
01∗
∗0.0
03∗∗
∗0.0
01∗
∗0.0
02∗∗
∗
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
Som
eC
olle
ge0.0
11
−0.0
19
0.0
23
0.0
27
(0.0
24)
(0.0
22)
(0.0
22)
(0.0
23)
Col
lege
−0.0
12
−0.0
20
−0.0
05
0.0
20
(0.0
23)
(0.0
21)
(0.0
21)
(0.0
21)
Pos
t-G
rad
uat
e−
0.0
07
−0.0
10
0.0
13
0.0
06
(0.0
31)
(0.0
28)
(0.0
28)
(0.0
28)
Inco
me:
$30-
60,0
000.0
10
−0.0
04
−0.0
01
0.0
09
(0.0
22)
(0.0
20)
(0.0
20)
(0.0
20)
Inco
me:
$60-
100,
000
0.0
06
−0.0
04
0.0
10
0.0
11
(0.0
23)
(0.0
21)
(0.0
21)
(0.0
21)
Inco
me:
¿$1
00,0
000.0
29
−0.0
002
0.0
06
0.0
20
(0.0
26)
(0.0
24)
(0.0
24)
(0.0
24)
Reg
ion
Con
trol
sX
XX
Con
stan
t0.
313∗
∗∗0.2
00∗∗
∗0.3
55∗∗
∗0.2
16∗∗
∗0.2
94∗
∗∗0.1
58∗
∗∗0.3
23∗∗
∗0.0
80
(0.0
37)
(0.0
65)
(0.0
34)
(0.0
59)
(0.0
34)
(0.0
60)
(0.0
34)
(0.0
60)
N1,
003
999
1,0
03
999
1,0
03
999
1,0
02
998
R2
0.08
00.1
15
0.1
33
0.1
91
0.1
33
0.1
72
0.1
29
0.1
78
∗ p<
.1;∗∗
p<
.05;
∗∗∗ p
<.0
1N
ote
:T
able
dis
pla
ys
OL
Sco
effici
ents
,st
an
dard
erro
rsin
pare
nth
eses
.A
llco
nti
nu
ou
sva
riab
les,
exce
pt
age,
hav
eb
een
resc
aled
from
0to
1.R
efer
ence
cate
gori
esare
¡$30,0
00,
Hig
hsc
hool
or
less
,an
dW
est.
Equ
ity
an
dE
quali
tyare
ad
dit
ive
scal
es.
6
Table 4: Multiple-Group Factor Analysis
Republicans Democrats IndependentsEquity Equality Equity Equality Equity Equality
Copies Ungraded 0.67 0.67 0.70Copies Graded 0.84 0.88 0.86Marathon Shortcut 0.82 0.84 0.84Lottery Division 0.49 0.63 0.52 0.31Halloween Candy 0.63 0.66 0.63Employee Earns More 0.85 0.76 0.83Raise for One Worker 0.85 0.81 0.79TLI 0.95 0.95 0.88RMSEA 0.08 0.09 0.13BIC -19.4 -12.29 -12.89Note: Table reports the results of a principal axis factor analysis with oblimin rotation. Cell entries indicatethe rotated factor loadings, values below 0.3 are removed for presentation. Republican N=369, DemocratN=481, Independent=155.
inputs and outcomes. Nevertheless, one possible concern is that equity and equality are proxies for parti-sanship or ideology, such that these variables, and not fairness, predicts participant responses to our foreignpolicy scenarios. We therefore estimated a series of additional OLS models that control for partisanship andideology.
Models 2, 5, 8, and 11 in Table 5 include a 7-item party identification scale as an independent variable,rescaled to range from 0 to 1.1 Models 3, 6, 9, and 12 include a 7-point ideology variable (rescaled from 9to 1). We present these results alongside our main results for comparison, in columns 1, 4. 7, and 10. Wefind clear and consistent evidence that our conclusions are robust to partisanship. While Republicans aremore likely than Democrats to report that inequitable defense spending contributions are wrong, the effect ofequity remains statistically significant and substantively larger than the effect of partisanship. Moreover, therelationship between partisanship and the other three foreign policy scenarios is weak and not statisticallysignificant. In the case of ideology, we find no evidence that ideology, rather than values, predicts how peoplerespond to burden-sharing violations in foreign policy.
We repeat these estimates for our policy proposal dependent variables, and report the results in Table6. Models 2, 5, and 8 control for partisanship, whereas models 3, 6, and 9 include ideology. We findsome evidence that Democratic party identification and liberal ideology explain the positive relationshipbetween equality and support for arms control agreements. Negative and statistically significant coefficientson party identification and ideology suggest that Republicans and conservatives, respectively, are less likelyto support arms control agreements. Once we account for this, the relationship between equality and supportweakens. Yet for each of the other policy outcomes — support for free trade and military spending — wefind consistent results for both equity and equality. Equity values predict less support for free trade andfor additional military spending, and equality predicts more support, even when we account for politicalpredispositions.
1We also estimate the models using dummy indicators for Republican, Independent, and Democrat. The results are consistentwith those presented in Table 5: Republicans express greater opposition to the defense spending vignette than Democrats, butwe find no other significant coefficients for party identification.
7
Tab
le5:
Wh
at’
sF
air
inU
.S.
Fore
ign
Poli
cy?
(wit
hP
art
isan
ship
an
dId
eolo
gy)
Def
ense
Sp
end
ing
Pea
cekee
pin
gE
nvir
on
men
tF
ore
ign
Aid
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
Equ
ity
0.2
8∗∗
∗0.2
7∗∗
∗0.2
8∗∗
∗0.3
7∗∗
∗0.3
7∗∗
∗0.3
7∗∗
∗0.3
9∗∗
∗0.3
9∗∗
∗0.3
9∗∗
∗0.3
8∗∗
∗0.3
8∗∗
∗0.3
8∗∗
∗
(0.0
5)
(0.0
5)
(0.0
5)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
Equ
ality
0.1
2∗∗
∗0.1
3∗∗
∗0.1
2∗∗
∗−
0.0
1−
0.0
1−
0.0
10.0
6∗
0.0
6∗
0.0
6∗
0.0
30.0
30.0
3(0
.03)
(0.0
4)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
Part
yID
0.0
7∗∗
∗0.0
20.0
20.0
3(0
.03)
(0.0
2)
(0.0
2)
(0.0
2)
Ideo
logy
0.0
30.0
20.0
30.0
3(0
.03)
(0.0
3)
(0.0
3)
(0.0
3)
Coop
erati
ve
Int.
−0.0
7−
0.0
4−
0.0
60.0
30.0
40.0
5−
0.0
5−
0.0
4−
0.0
40.0
20.0
30.0
3(0
.05)
(0.0
5)
(0.0
5)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
5)
(0.0
4)
(0.0
4)
(0.0
5)
Milit
ant
Int.
0.0
50.0
10.0
4−
0.0
2−
0.0
3−
0.0
30.0
70.0
60.0
60.1
2∗∗
∗0.1
1∗∗
0.1
1∗∗
(0.0
5)
(0.0
5)
(0.0
5)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
5)
(0.0
4)
(0.0
4)
(0.0
5)
Isola
tion
ism
0.1
8∗∗
∗0.1
9∗∗
∗0.1
8∗∗
∗0.1
8∗∗
∗0.1
8∗∗
∗0.1
7∗∗
∗0.1
6∗∗
∗0.1
6∗∗
∗0.1
6∗∗
∗0.1
7∗∗
∗0.1
7∗∗
∗0.1
7∗∗
∗
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
(0.0
4)
Male
−0.0
2−
0.0
3−
0.0
3−
0.0
4∗∗
−0.0
4∗∗
−0.0
4∗∗
−0.0
3∗
−0.0
3∗
−0.0
3∗
−0.0
3∗
−0.0
3∗
−0.0
3∗
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
Wh
ite
−0.0
000
−0.0
1−
0.0
02
0.0
20.0
10.0
20.0
20.0
20.0
20.0
4∗∗
0.0
3∗
0.0
4∗
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
Age
0.0
01∗∗
0.0
02∗∗
∗0.0
01∗∗
0.0
03∗∗
∗0.0
03∗∗
∗0.0
03∗∗
∗0.0
02∗∗
∗0.0
02∗∗
∗0.0
02∗∗
∗0.0
02∗∗
∗0.0
02∗∗
∗0.0
02∗∗
∗
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
Som
eC
olleg
e0.0
10.0
10.0
1−
0.0
2−
0.0
2−
0.0
20.0
20.0
20.0
20.0
30.0
30.0
3(0
.02)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
Colleg
e−
0.0
1−
0.0
1−
0.0
1−
0.0
2−
0.0
2−
0.0
2−
0.0
1−
0.0
04
−0.0
04
0.0
20.0
20.0
2(0
.02)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
Post
-Gra
du
ate
−0.0
1−
0.0
1−
0.0
1−
0.0
1−
0.0
1−
0.0
10.0
10.0
10.0
10.0
03
0.0
03
0.0
03
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
(0.0
3)
Inco
me:
$30-6
0,0
00
0.0
10.0
10.0
1−
0.0
02
−0.0
03
−0.0
03
0.0
01
0.0
003
0.0
004
0.0
10.0
10.0
1(0
.02)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
Inco
me:
$60-1
00,0
00
0.0
10.0
03
0.0
1−
0.0
03
−0.0
04
−0.0
03
0.0
10.0
10.0
10.0
10.0
10.0
1(0
.02)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
Inco
me:
¿$100,0
00
0.0
30.0
20.0
3−
0.0
001
−0.0
01
−0.0
005
0.0
10.0
04
0.0
10.0
20.0
20.0
2(0
.03)
(0.0
3)
(0.0
3)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
(0.0
2)
Reg
ion
Contr
ols
XX
XX
XX
XX
Con
stant
0.2
0∗∗
∗0.1
6∗∗
0.1
8∗∗
∗0.2
2∗∗
∗0.2
1∗∗
∗0.2
1∗∗
∗0.1
6∗∗
0.1
4∗∗
0.1
4∗∗
0.0
80.0
60.0
6(0
.07)
(0.0
7)
(0.0
7)
(0.0
6)
(0.0
6)
(0.0
6)
(0.0
6)
(0.0
6)
(0.0
6)
(0.0
6)
(0.0
6)
(0.0
6)
N999
999
999
999
999
999
999
999
999
998
998
998
R2
0.1
10.1
20.1
10.1
80.1
80.1
80.1
60.1
70.1
70.1
70.1
70.1
7∗p<
.1;∗∗
p<
.05;∗∗
∗p<
.01
Note
:T
ab
led
isp
lays
OL
Sco
effici
ents
,st
an
dard
erro
rsin
pare
nth
eses
.A
llco
nti
nu
ou
svari
ab
les,
exce
pt
age,
have
bee
nre
scale
dfr
om
0to
1.
Hig
her
valu
eson
part
yid
enti
fica
tion
an
did
eolo
gy
ind
icate
Rep
ub
lica
nor
con
serv
ati
ve.
Ref
eren
ceca
tegori
esare
¡$30,0
00,
Hig
hsc
hool
or
less
,an
dW
est.
8
2.5 Equity, Equality, and Policy Proposals
Table 7 presents results from 6 OLS models that predict support for three policy proposals as a functionof equity, equality, and a panel of additional independent variables. As discussed in the manuscript, theresults show how the two dimensions of fairness — equity and equality — are not always complementary.The results from Models 2, 4, and 6 were used to produce Figure 2 in the manuscript.
References
Ansolabehere, Stephen, Jonathan Rodden and James M. Snyder. 2008. “The Strength of Issues: UsingMultiple Measures to Gauge Preference Stability, Ideological Constraint, and Issue Voting.” AmericanPolitical Science Review 102(2):215–232.
Clifford, Scott, Vijeth Iyengar, Roberto Cabeza and Walter Sinnott-Armstrong. 2015. “Moral foundationsvignettes: A standardized stimulus database of scenarios based on moral foundations theory.” Behaviorresearch methods 47(4):1178–1198.
Gravelle, Timothy B, Jason Reifler and Thomas J Scotto. 2017. “The structure of foreign policy attitudesin transatlantic perspective: Comparing the United States, United Kingdom, France and Germany.”European Journal of Political Research 56(4):757–776.
Haidt, Jonathan. 2012. The Righteous Mind: Why good people are divided by politics and religion. NewYork: Pantheon Books.
Hainmueller, Jens. 2012. “Entropy Balancing for Causal Effects: A Multivariate Reweighting Method toProduce Balanced Samples in Observational Studies.” Political Analysis 20(1):25–46.
Iyer, Ravi. 2010. “Democrats and Republicans agree that Justice & Fairness are about Equity, not Equalityor Impartiality.”.URL: http://www.polipsych.com/2010/02/20/democrats-and-republicans-agree-that-justice-fairness-are-about-equity-not-equality-or-impartiality/
Kertzer, Joshua D., Kathleen Powers, Brian C. Rathbun and Ravi Iyer. 2014. “Do moral values shape foreignpolicy preferences?” Journal of Politics 76(3):825–840.
Meindl, Peter, Ravi Iyer and Jesse Graham. 2019. “Distributive Justice Beliefs are Guided by WhetherPeople Think the Ultimate Goal of Society is Well-Being or Power.” Basic and Applied Social Psychology41(6):359–385.
Osborne, Jason W. and Anna B Costello. 2009. “Best Practices in Exploratory Facotr Analysis: FourRecommendations for Getting the Most from Your Analysis.” Pan-Pacficic Management Review 12(2):131–146.
9
Tab
le6:
Fair
nes
san
dP
oli
cyP
rop
osa
ls(w
ith
Part
isan
ship
an
dId
eolo
gy)
Su
pp
ort
Fre
eT
rad
eS
up
port
Arm
sC
ontr
ol
Su
pp
ort
Mil.
Sp
end
ing
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Equ
ity
−0.1
41∗∗
−0.1
32∗∗
−0.1
34∗∗
−0.0
16
−0.0
01
0.0
12
−0.1
13∗
−0.1
15∗
−0.1
19∗
(0.0
65)
(0.0
65)
(0.0
65)
(0.0
60)
(0.0
60)
(0.0
59)
(0.0
67)
(0.0
66)
(0.0
67)
Equ
ality
0.1
40∗∗
∗0.1
26∗∗
0.1
31∗∗
∗0.0
94∗∗
0.0
68
0.0
71
0.1
00∗∗
0.1
11∗∗
0.1
02∗∗
(0.0
49)
(0.0
49)
(0.0
49)
(0.0
46)
(0.0
46)
(0.0
45)
(0.0
50)
(0.0
50)
(0.0
50)
Milit
ant
Int.
0.1
62∗∗
0.2
06∗∗
∗0.2
03∗∗
∗−
0.0
84
−0.0
34
0.0
02
0.7
25∗∗
∗0.6
74∗∗
∗0.6
83∗∗
∗
(0.0
68)
(0.0
72)
(0.0
72)
(0.0
61)
(0.0
62)
(0.0
63)
(0.0
72)
(0.0
76)
(0.0
77)
Coop
erati
ve
Int.
0.1
56∗∗
0.1
13
0.1
17∗
0.6
32∗∗
∗0.5
70∗∗
∗0.5
15∗∗
∗−
0.1
33∗
−0.0
85
−0.0
94
(0.0
66)
(0.0
69)
(0.0
69)
(0.0
59)
(0.0
62)
(0.0
63)
(0.0
70)
(0.0
74)
(0.0
74)
Isola
tion
ism
0.0
70
0.0
67
0.0
76
−0.0
10
−0.0
25
−0.0
01
−0.3
69∗∗
∗−
0.3
65∗∗
∗−
0.3
73∗∗
∗
(0.0
54)
(0.0
54)
(0.0
54)
(0.0
51)
(0.0
51)
(0.0
50)
(0.0
58)
(0.0
58)
(0.0
58)
Part
yID
−0.0
69∗
−0.1
06∗∗
∗0.0
73∗∗
(0.0
36)
(0.0
34)
(0.0
36)
Ideo
logy
−0.0
80∗
−0.1
95∗∗
∗0.0
74
(0.0
47)
(0.0
43)
(0.0
47)
Male
0.0
17
0.0
16
0.0
17
0.0
17
0.0
19
0.0
23
0.0
23
0.0
24
0.0
21
(0.0
24)
(0.0
24)
(0.0
24)
(0.0
23)
(0.0
23)
(0.0
23)
(0.0
25)
(0.0
25)
(0.0
25)
Wh
ite
−0.0
67∗∗
−0.0
54∗
−0.0
61∗∗
−0.0
47∗
−0.0
29
−0.0
31
−0.0
32
−0.0
41
−0.0
36
(0.0
29)
(0.0
30)
(0.0
29)
(0.0
28)
(0.0
29)
(0.0
28)
(0.0
29)
(0.0
30)
(0.0
29)
Age
−0.0
002
−0.0
004
−0.0
002
0.0
01
0.0
01
0.0
01
−0.0
005
−0.0
004
−0.0
01
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
(0.0
01)
Som
eC
olleg
e−
0.0
48
−0.0
48
−0.0
47
−0.0
14
−0.0
10
−0.0
14
0.0
05
0.0
09
0.0
07
(0.0
34)
(0.0
34)
(0.0
34)
(0.0
33)
(0.0
32)
(0.0
32)
(0.0
35)
(0.0
35)
(0.0
35)
Colleg
e−
0.0
44
−0.0
46
−0.0
47
−0.0
18
−0.0
18
−0.0
23
−0.0
35
−0.0
28
−0.0
30
(0.0
33)
(0.0
33)
(0.0
33)
(0.0
30)
(0.0
29)
(0.0
29)
(0.0
33)
(0.0
33)
(0.0
33)
Post
-Gra
du
ate
−0.0
05
−0.0
08
−0.0
05
−0.0
36
−0.0
36
−0.0
34
−0.0
80∗
−0.0
75∗
−0.0
78∗
(0.0
43)
(0.0
43)
(0.0
43)
(0.0
38)
(0.0
38)
(0.0
37)
(0.0
43)
(0.0
43)
(0.0
43)
Inco
me:
$30-6
0,0
00
0.0
92∗∗
∗0.0
95∗∗
∗0.0
92∗∗
∗−
0.0
23
−0.0
20
−0.0
13
0.0
46
0.0
44
0.0
46
(0.0
31)
(0.0
31)
(0.0
31)
(0.0
29)
(0.0
29)
(0.0
28)
(0.0
31)
(0.0
31)
(0.0
31)
Inco
me:
$60-1
00,0
00
0.0
01
0.0
07
0.0
02
0.0
02
0.0
12
0.0
12
0.0
33
0.0
26
0.0
30
(0.0
31)
(0.0
31)
(0.0
31)
(0.0
30)
(0.0
30)
(0.0
29)
(0.0
31)
(0.0
31)
(0.0
31)
Inco
me:
¿$100,0
00
0.0
88∗∗
0.0
95∗∗
0.0
90∗∗
0.0
11
0.0
19
0.0
18
0.0
01
−0.0
08
−0.0
01
(0.0
37)
(0.0
37)
(0.0
37)
(0.0
33)
(0.0
33)
(0.0
33)
(0.0
37)
(0.0
37)
(0.0
37)
Reg
ion
contr
ols
XX
XX
XX
XX
XC
on
stant
0.4
04∗∗
∗0.4
33∗∗
∗0.4
34∗∗
∗0.3
37∗∗
∗0.3
97∗∗
∗0.4
31∗∗
∗0.4
47∗∗
∗0.4
05∗∗
∗0.4
20∗∗
∗
(0.0
90)
(0.0
91)
(0.0
92)
(0.0
87)
(0.0
88)
(0.0
88)
(0.0
93)
(0.0
95)
(0.0
94)
N515
515
515
491
491
491
477
477
477
R2
0.1
30
0.1
37
0.1
35
0.2
48
0.2
63
0.2
79
0.2
74
0.2
80
0.2
77
∗p<
.1;∗∗
p<
.05;∗∗
∗p<
.01
Note
:T
able
dis
pla
ys
OL
Sco
effici
ents
,st
an
dard
erro
rsin
pare
nth
eses
.A
llco
nti
nu
ou
svari
ab
les,
exce
pt
age,
have
bee
nre
scale
dfr
om
0to
1.
Hig
her
valu
eson
part
yid
enti
fica
tion
an
did
eolo
gy
ind
icate
Rep
ub
lica
nan
dco
nse
rvati
ve.
Ref
eren
ceca
tegori
esare
¡$30,0
00,
Hig
hsc
hool
or
less
,an
dW
est.
10
Table 7: Equity, Equality, and Policy Proposals
Support Free Trade Support Arms Control Support Mil. Spending
(1) (2) (3) (4) (5) (6)
Equity −0.133∗∗ −0.141∗∗ 0.040 −0.016 −0.075 −0.113∗
(0.062) (0.065) (0.062) (0.060) (0.069) (0.067)Equality 0.167∗∗∗ 0.140∗∗∗ 0.167∗∗∗ 0.094∗∗ 0.065 0.100∗∗
(0.047) (0.049) (0.048) (0.046) (0.054) (0.050)Militant Int. 0.162∗∗ −0.084 0.725∗∗∗
(0.068) (0.061) (0.072)Cooperative Int. 0.156∗∗ 0.632∗∗∗ −0.133∗
(0.066) (0.059) (0.070)Isolationism 0.070 −0.010 −0.369∗∗∗
(0.054) (0.051) (0.058)Male 0.017 0.017 0.023
(0.024) (0.023) (0.025)White −0.067∗∗ −0.047∗ −0.032
(0.029) (0.028) (0.029)Age −0.0002 0.001 −0.0005
(0.001) (0.001) (0.001)Some College −0.048 −0.014 0.005
(0.034) (0.033) (0.035)College −0.044 −0.018 −0.035
(0.033) (0.030) (0.033)Post-Graduate −0.005 −0.036 −0.080∗
(0.043) (0.038) (0.043)Income: $30-60,000 0.092∗∗∗ −0.023 0.046
(0.031) (0.029) (0.031)Income: $60-100,000 0.001 0.002 0.033
(0.031) (0.030) (0.031)Income: > $100,000 0.088∗∗ 0.011 0.001
(0.037) (0.033) (0.037)Region Controls X X XConstant 0.564∗∗∗ 0.404∗∗∗ 0.579∗∗∗ 0.337∗∗∗ 0.508∗∗∗ 0.447∗∗∗
(0.053) (0.090) (0.054) (0.087) (0.059) (0.093)N 517 515 494 491 479 477Adjusted R2 0.024 0.100 0.023 0.221 0.001 0.247
∗p < .1; ∗∗p < .05; ∗∗∗p < .01
11