Fairness and family background

INSTITUTT FOR SAMFUNNSØKONOMI
DEPARTMENT OF ECONOMICS
SAM 25 2015
ISSN: 0804-6824
October 2015
Discussion paper
Fairness and
family background
BY
Ingvild Almås, Alexander W. Cappelen, Kjell G. Salvanes,
Erik Ø. Sørensen AND Bertil Tungodden
This series consists of papers with limited circulation, intended to stimulate discussion.
Fairness and family background∗
Ingvild Almås
Alexander W. Cappelen
Kjell G. Salvanes
Erik Ø. Sørensen
Bertil Tungodden
Abstract
Fairness preferences fundamentally affect individual behavior and play an
important role in shaping social and political institutions. However, people differ both with respect to what they view as fair and with respect to how much
weight they attach to fairness considerations. In this paper, we study the role of
family background in explaining these heterogeneities in fairness preferences. In
particular, we examine how socioeconomic background relates to fairness views
and to how people make trade-offs between fairness and self-interest. To study
this we conducted an economic experiment with a representative sample of 14-15
year-olds and matched the experimental data to administrative data on parental
income and education. The participants made two distributive choices in the
experiment. The first choice was to distribute money between themselves and
another participant in a situation where there was no difference in merit. The
second choice was to distribute money between two other participants with unequal merits. Our main finding is that there is a systematic difference in fairness
view between children from low socioceconomic status (SES) families and the
rest of the participants; more than 50 percent of the participants from low SES
families are egalitarians, whereas only about 20 percent in the rest of the sample
hold this fairness view. In contrast, we find no significant difference in the weight
attached to fairness between children from different socioeconomic groups.
∗ Department of Economics, NHH Norwegian School of Economics, 5045 Bergen, Norway. Emails:
[email protected], [email protected], [email protected], [email protected],
[email protected] Special thanks to Andrew Williams for in-depth comments. Thanks also
to participants at the Politics, Philosophy and Economics conference in New Orleans on February 7-9,
2014. We would like to thank the municipal school authority in Bergen for their cooperation; Kristin
Risvand Mo for administrative support; Atle Askeland, Bjørn Ivar Grøttå, and Sarah Marie Søiland
for IT-support; and Eli Birgitte S. Bergsmark, Espen Bolghaug, Andreas Tufteland Engelsen, Kasper
Thoring Fellkjær, Sebastian Fest, Ingar Haaland, Sigve Garsjø, Vi-Vi Ton Nu, Kristin Orset, Henrik
Reigstad, Mari Sakkestad, Johanne Amundsen Wik, Kristina Øystese, and Silje Åsnes for research assistance. The project was financed by support from the Research Council of Norway, research grant
185831, and administered by The Choice Lab.
1
1
Introduction
Fairness preferences fundamentally affect individual behavior and play a crucial role
in shaping social and political institutions. Importantly, it is well documented from
economic experiments that most people are motivated by fairness considerations and
are willing to sacrifice personal gains in order to eliminate income inequalities they
view as unfair (Fehr and Schmidt, 1999; Bolton and Ockenfels, 2000; Engelmann and
Strobel, 2004), but it also shown that most people view some income inequalities as
fair (Konow, 2000; Cherry, Frykblom, and Shogren, 2002; Frohlich, Oppenheimer,
and Kurki, 2004; Cappelen, Drange Hole, Sørensen, and Tungodden, 2007). There is
considerable heterogeneity both with respect to how much weight people attach to fairness considerations relative to their self-interest and with respect to what people view
as fair (Cappelen et al., 2007; Cappelen, Sørensen, and Tungodden, 2010; Cappelen,
Konow, Sørensen, and Tungodden, 2013a; Engel, 2011), and in this paper we study the
role of family background in explaining these heterogeneities in fairness preferences.
In order to examine this relationship, we conducted a real effort fairness experiment with a representative sample of 14-15 year-olds and matched the behavioral data
to administrative data on parental income and education. The distribution phase of
the experiment had two parts. In the first part, participants anonymously distributed
money between themselves and another participant, where both had made the same
contribution to the pool of money being distributed. Since it is commonly assumed
that an equal spilt is the fair allocation in such a situation, the share given to the other
participant provides us with a measure of the weight the participants attach to fairness considerations relative to self-interest. In the second part, participants distributed
money between two other participants with unequal merits, i.e. they had made unequal
contributions to the pool of money being distributed. This allows us to identify the participants’ fairness views, specifically whether they are meritocrats finding inequalities
reflecting differences in merit to be fair, or egalitarians.
Our main finding is that there is a large and significant difference in the fairness
views across groups with different family background: participants from low socioeconomic status (SES) families are much more likely to have an egalitarian fairness
view and consider an equal distribution as fair in a situation with unequal merits; more
than 50 percent of the participants from low SES families are egalitarians, whereas
only about 20 percent in the rest of the sample hold this fairness view. We do, however, not find any socioeconomic gradient in the weight attached to fairness between
adolescents with different socioeconomic background.
To the best of our knowledge, this is the first paper to study the relationship between fairness preferences and family background on a sample of adolescents. Our
findings contribute to the literature on how fairness preferences differ between groups
within and across societies (Henrich, Boyd, Bowles, Camerer, Fehr, Gintis, McElreath, Alvard, Barr, Ensminger, Henrich, Hill, Gil-White, Gurven, Marlowe, Patton,
and Tracer, 2005; Engel, 2011; Cappelen, Moene, Sørensen, and Tungodden, 2013b).
The study also contributes to the extensive literature on the development of children’s
2
fairness preferences in psychology (Piaget, 1965; Mikula, 1972; Damon, 1977; Hook
and Cook, 1979; Gilligan, 1982; Kohlberg, 1984; Moore, Hembree, and Enright, 1993;
Eisenberg, Carlo, Murphy, and van Court, 1995) and, more recently, in behavioral economics (Benenson, Pascoe, and Radmore, 2007; Harbaugh, Krause, and Vesterlund,
2007; Sutter and Kocher, 2007; Sutter, 2007; Fehr, Bernhard, and Rockenbach, 2008;
Almås, Cappelen, Sørensen, and Tungodden, 2010; Fehr, Glätzle-Rützler, and Sutter,
2013; Bauer, Chytilová, and Pertold-Gebicka, 2014). This literature has shown that
children tend to become more prosocial with age in early childhood and more meritocratic and inequality accepting in adolescence. The present paper contributes to our
understanding of how fairness preferences are shaped in childhood by documenting
a strong association between children’s socioeconomic background and their fairness
views.
The paper unfolds as follows. Section 2 gives an overview of the experimental
design and Section 3 describes the sample and the background data. Section 4 introduces a simple model of distributive choice. Section 5 provides some descriptives of
the data and analyses the relationship between sosioeconomic background and fairness
preferences. Finally, Section 6 concludes.
2
Experimental design
To study the participants’ fairness preferences we conducted an experiment with a production phase and a distribution phase. In the production phase, the participants were
given limited time to work on a simple task, which consisted of counting the number
of black squares in a matrix shown on their screen. The participants were told the
following before the task:
The task is to count how many black squares there are in a matrix that will be shown to
you on the screen. When you have done this for a matrix, you should enter the number
in the field below the matrix. When you have counted for all the matrices, you hit the
button to continue. You will then get a new set of matrices. You can also choose to
continue even if you haven’t counted all matrices on one page. For each matrix you
count correctly, you will get one point. Your points will be updated on the screen each
time you continue to a new page. On the screen you will also see a timer that tells you
how much time you have left. When time is up, the computer will count your points,
including those from the last page you were on. You will earn money on this task. What
you earn consists of two parts. First, you will earn 50 NOK independent of how many
correct answers you gave. If you get at least as many points as the average of what
the other participants get in this session, you will earn 75 NOK extra. If you get less
points than the average, you will earn 25 NOK extra. When you are done with the task,
you must consider how the money you have earned should be distributed among participants (including yourself). What distributional choices you and others make will
determine your actual payment from this part of the experiment. When you have read
3
these instructions, you can hit the button to start working on the task.
After conducting the task, the participants took part in the distribution phase. The
distribution phase had two parts. In the first part, each participant was anonymously
matched with another participant and asked to decide how the sum of his or her own
fixed payment and the other participant’s fixed payment, in total 100 NOK, should be
distributed between the two of them. The participant was free to choose any distribution of the money. Both participants in a pair made a decision about how the money
should be distributed and were told that one of the two decisions would be randomly
drawn at the end of the experimental session to decide the actual distribution of the
money in the pair.
The key feature of this choice, which we refer to as the stakeholder choice, is that
there are no potential economic benefits of giving money to the other participant. If a
participant is narrowly selfish and only aims to maximize own income, he or she should
allocate everything to him- or herself. However, people typically also care about being
fair. In this situation, where the two individuals have made an equal contribution to
the sum of money being divided, it seems likely that most people consider it fair to
divide the money equally. The participants are thus placed in a situation where there is
a trade-off between fairness considerations, which pull in the direction of giving half
the money to the other participant, and self-interest, which pulls in the direction of
giving nothing. How much a participant decides to give to the other participant can
therefore serve as a measure of how much weight that participant places on fairness
considerations relative to self-interest.1
In the second part of the distribution phase each participant was asked to decide,
as a spectator, how the bonus earned by two other participants should be distributed.
They had to decide on a distribution for a pair where one participant had been less
productive than the mean in the session, and thus had earned the low bonus of 25
NOK, while the other participant had been at least as productive as the mean, and
had thus earned the high bonus of 75 NOK. The spectators were given a binary choice
between a distribution equal to the earnings, giving 25 NOK to the least productive and
75 NOK to the most productive, or an equal division of 50 NOK to both participants.2
The key feature of the spectator choice is that the decision-maker has no stake in
the situation. This implies that there is no trade-off between fairness and self-interest
1 Note
that the stakeholder design is not well suited to identify the fairness view of the participants,
specifically whether they are egalitarian or meritocrats. First, participants in this setting are influenced
by both self-interest and fairness (which means that even egalitarians may not choose an equal split).
Second, both meritocrats and egalitarians agree that an equal split is fair in this situation, since both
participants have made the same contribution.
2 Note that we compare with the mean productivity in the session, not the median productivity. Consequently, we could not guarantee that there were exactly 50 percent high earners and 50 percent low
earners; in the experiment, we had 56.7 percent high earners. When constructing spectator situations,
some of the high earners were assigned to more than one situation. For these participants, we randomly
selected one of the situations to determine their payout from this part of the experiment.
4
and the spectators can therefore be expected to choose the alternative they view as most
fair. We can thus use the spectator choice to identify the fairness view of a participant.
In particular, it enables us to examine whether he or she views inequality as fair when
it reflects differences in merit.
We conducted ten experimental sessions at NHH Norwegian School of Economics,
where each session lasted for approximately two hours, and used a web-based interface.3 All participants received a show-up fee of 50 NOK, in addition to what they
earned in the experimental session. Neither participants nor experimenters could associate decisions with particular participants and this was common knowledge. Special care was taken so that the payment procedure ensured participant-experimenter
anonymity. At the end of the experimental session the computer assigned a payment
code to each of the participants, and a group of assistants, who were not present in the
lab during the experimental session, prepared envelopes containing the payments corresponding to each payment code. The assistants also made sure that it was impossible
to identify the amount of money by simply looking at the envelope. After bringing the
envelopes to the lab, the assistants immediately left and the envelopes were handed out
in accordance with the payment codes.
3
Sample and background data
The participants were recruited among Norwegian adolescents in 9th grade, i.e., 14-15
year-olds. 9th grade is compulsory in Norway and almost all children attend public
schools (97.2 percent). We randomly selected 11 public middle schools in Bergen,
which is the second largest city in Norway and close to the national average of the
Norwegian urban population with respect to the distribution of income, education and
occupation. Two schools later withdrew due to practical circumstances. At each school
we randomly selected two classes, and all the students in the selected classes received
a personal invitation to participate in the experiment. Participation was voluntary and
both students and their parents had to consent to participation. The participation rate
was high; 524 out of 602 invited students took part in the experiment (87 percent).4
In collaboration with Statistics Norway, we matched the data from the experiment
to Norwegian administrative data, which is a linked national administrative high quality data set. We have detailed parental background information on education and income for 483 of the 523 children (477 of the 517 in the stakeholder choice).5 Table 1
shows that our sample of parents is largely representative for the population in Bergen.
3 The experimental session consisted of a sequence of experiments, see Appendix B and Almås,
Cappelen, Salvanes, Sørensen, and Tungodden (Forthcoming) for a discussion of all the experiments.
4 Six participants could not be paired for the stakeholder choice due to uneven numbers in six sessions, and we had to exclude one observation from the experiment for administrative reasons. This
results in an experimental sample of 523 for the spectator choice and 517 for the stakeholder choice.
5 For 28 children, we have some background information, whereas 12 children could not be matched
to the administrative data.
5
[ Table 1 about here ]
Our main definition of a low SES family is that the family is in the bottom fifth
of both the income and the education distribution, where family income is measured
as the total income of the father and the mother and family education is measured as
the total years of education of the father and the mother.6 We focus on the bottom
fifth of the education distribution since this captures a distinct group of families where
one parent has no more than compulsory schooling and the other parent has no more
than high school. In a highly educated society, such a family clearly has a low educational background. Some parents with only compulsory education do well in terms of
income, however, and thus we also impose the restriction that the families are in the
bottom fifth of the income distribution. According to our definition, eight percent of
the participants in our experiment are from low SES families.
4
A model of distributive choice
To fix ideas, we present a simple model of how participants make decisions in the distribution phase of the experiment (see e.g., Cappelen et al. (2007); Almås et al. (2010);
Cappelen et al. (2013b) for related model formulations). In the first part of the distribution phase, the participants make a choice as a stakeholder and we assume that
stakeholders in such situations are motivated by own income and fairness considerations when deciding how to distribute the sum of earnings:
V i (y; ·) = y − β i f (|y − F i |),
(1)
where y is what the participant allocates to him- or herself, and F i is what the participant considers to be his or her fair income. The perceived unfairness is captured by the
function f (|y−F i |), where we assume that the perceived unfairness is increasing in the
absolute value of the difference between own income and fair income, and f (0) = 0.
Participants may differ both in the weight they attach to fairness relative to selfinterest (β i ) and in what they consider to be a fair distribution (F i ). We assume that
the individuals either endorse an egalitarian fairness view (F i = F E ) that find all inequalities unfair or a meritocratic fairness view (F i = F M ) that find fair inequalities
reflecting differences in merit. In a situation where earnings reflect merit, we can write
these positions as:
1
F E = X,
2
F M = xi ,
6 The
(2)
(3)
main results in the paper are robust to alternative cutoffs for low SES, including defining low
SES families as the bottom 30 percent or bottom 40 percent of both the income and the education
distribution.
6
where xi is the decision-maker’s earnings.
In the stakeholder choice, both fairness views justify an equal split of the money
since the two participants have equal earnings. The share of the money given by individual i therefore provides us with a measure of the weight attached to fairness considerations relative to self-interest, β i .
In the spectator choice, however, the two fairness views have different implications,
since the two individuals in the pair have unequal merits. The egalitarian fairness view
(F E ) considers the fair distribution to be an equal division, while the meritocratic view
(F M ) consider the fair distribution to be that the participants receive their earnings.
Since self-interest plays no role for the spectators, the choice in the second part of the
experiment reveals the participants’ fairness view.
5
Results
We start by presenting some descriptive statistics of the behavior in the two choice
situations, before we examine the association between socioeconomic background and
fairness preferences.
The top panels in Figure 1 provide an overview of the decisions made by participants in the two choice situations. The upper left panel in Figure 1 provides the
histogram of the stakeholder choice for all participants. The average share given is
0.31, but we observe considerable heterogeneity in distributive behavior. About half
of the participants, 51.8 percent, chose to share equally, while 33.3 of the participants
chose to take everything for themselves. Most of the remaining participants chose to
give about a quarter of the money to the other participant. The upper right panel in
Figure 1 provides an overview of the decision made by the participants as spectators.
We observe that only 27.0 percent of the participants chose the egalitarian alternative.
Most participants thus find it fair that income is distributed unequally as long as the
inequality reflects differences in merit.
From the second row in Figure 1, we observe that there is a striking similarity in the
distribution of the stakeholder choice for girls and boys. While some previous studies
have found that men on average are more selfish than women (Engel, 2011), this does
not seem to be the case with Norwegian adolescents.7 Nor do we find any gender
difference in the spectator choice. The similarity in behavior for boys and girls could
reflect the fact that Norway for a long time has been implementing gender equalizing
policies.8
[ Figure 1 about here]
7 The
lack of a gender effect on selfish behavior is consistent with the results from a previous study
of fairness preferences among Norwegian adolescents (Almås et al., 2010).
8 In 2012, Norway ranked highest on the gender equality index of the UN comprising measures of
educational attainment, labor market participation, and health. For further details on the gender equality
index, see http://hdr.undp.org/en/media/.
7
The two panels in the bottom row of Figure 1 report histograms of stakeholder
choice by fairness view. We observe that the choice behavior is very similar for egalitarians and meritocrats (and not statistically significantly different), they give away
32.4 percent and 30.2 percent, respectively. There is, in other words, no correlation
between the participants’ fairness views and the weight they attach to fairness, which
suggests that F i and β i in our model capture two distinct aspects of an individual’s
fairness preferences.
We now turn to the relationship between family background and fairness preferences, where we focus on the comparison between low SES children and the rest of
the participants.
[ Figure 2 about here]
In Figure 2 we report the share of the money that was given to the other participant
in the stakeholder situation by SES group. We observe that participants from both
low SES and medium/high SES families place a substantial weight on fairness. On
average the participants gave away about 30 percent of the total earnings and there is
no statistical significant difference between the two groups (see Table 2). This is robust
to the inclusion of gender and productivity as controls (column 3 and 5 respectively).
We observe, from columns 2 and 4, that there is no correlation between gender or
productivity in the task and the share given to the other participant.
[ Table 2 about here ]
Our findings suggest that there is no socioeconomic gradient in the weight attached
to fairness considerations. In Appendix A, we furthermore show that this result is
robust to alternative definitions of low SES (Table A1) and to including continuous
measures of parental income and education (Table A3).
The absence of a social gradient in the share given to the other participant is
striking, since there are two arguments not captured by our model that one might
think would cause the low SES children to give less to the other participant than
medium/high SES children. First, one might think that low SES children have higher
marginal utility of money, which would strengthen the selfish argument and pull in
the direction of giving away less. Second, one might think that the low SES children
would invoke a needs argument in their fairness consideration, which would provide a
fairness argument for allocating more to themselves than to other participant. Interestingly, none of these effects seems to be present in our experiment. The result is also
surprising in light of Bauer et al. (2014), who find in a study in the Czech republic, that
children, aged 4 to 12 year, of parents with low education are more selfish. The opposite findings may, however, reflect that we study older children in a different cultural
context.
In Figure 3 we report the share of participants who chose an equal split by SES
group. We observe a large and statistically significant difference in the average share
8
of egalitarians between low SES children and the rest of the participants. Whereas
more than 50 percent of the low SES children chose the equal split, only about 20
percent of the rest of the participants did so. Column 1 in Table 3 confirms that the
difference in the share of egalitarians by SES is statistically significant (p < 0.01).
This result is robust to the inclusion of gender and productivity in the task as controls
(columns 3 and 5 respectively).
[ Figure 3 about here]
In column 4 in Table 3, we also observe that there is a significant negative correlation between productivity and the likelihood of holding the egalitarian fairness view
(p < 0.05). In other words, participants who would have benefited from an egalitarian
distribution because they had the low bonus, were more likely than others to act as
an egalitarian when making a spectator choice. This result is consistent with a selfserving bias in the participants’ fairness views (Babcock, Loewenstein, Issacharoff,
and Camerer, 1995; Babcock and Loewenstein, 1997; Cappelen et al., 2013b; Dana,
Weber, and Kuang, 2007; Konow, 2000; Messick and Sentis, 1983). However, as
shown in column 5, this self-serving bias does not explain the association between low
SES and egalitarianism.
[ Table 3 about here ]
In Appendix A, we show that the association between low socioeconomic background and fairness is stronger for the father’s background than for the mother’s background, and that having parents with low education has a stronger effect on the likelihood of being egalitarian than having parents with low income (Table A2). In Appendix A, we also report regressions of the likelihood of being egalitarian on parental
income and education separately (see Table A3), where we find that both variables
have a significant negative effect on the share of egalitarians. Finally, we do not find
that other features of family background, such as the number of siblings or whether the
participant is the oldest child in the family, have any significant effect on the likelihood
of being egalitarian (see Table A4).
6
Concluding remarks
In this paper we have examined the distributive behavior of a representative group of
Norwegian adolescents in two choice situations, a stakeholder situation and a spectator
situation, and we have linked the data on their behavior to data on family background.
The behavior in the two choice situations enable us to identify the participants’ fairness
views and the weight they attach to fairness relative to own income. Our main finding
is that there is a systematic difference across family background groups in what is considered to be a fair distribution. Whereas more than 50 percent of the participants from
low SES families are egalitarians, only about 20 percent in the rest of the sample hold
9
this fairness view. In contrast, we find no significant difference in the weight attached
to fairness considerations by adolescents from low SES families compared with the
rest of the sample. An interesting avenue for future research is to study the extent to
which these results also hold in other countries. In particular, whether the effect of
socioeconomic background is different in less egalitarian societies than Norway.
In order to design optimal policies and understand the support for political and social institutions in society, it is important to have knowledge of what people view as
fair and how fairness views differ across groups. The present study suggests that disagreement about redistributive welfare policies across different socioeconomic groups
does not only derive from selfish considerations being sensitive to your position in
the income distribution, it may also stem from different socioeconomic groups having
different ideas of fairness. Specifically, low SES people may find a high level of redistribution fair from their egalitarian fairness viewpoint, while high SES people may
find it unfair from their meritocratic fairness viewpoint.
Our findings contribute to the literature on how fairness preferences differ between
groups and societies (Henrich et al., 2005; Engel, 2011). In particular, our findings
are interesting in light of Cappelen et al. (2013b), who examine differences in adults
fairness views between rich and poor countries. Their study shows that there are systematic differences between individuals in two of the richest countries in the world
(Norway and Germany) and in two of the poorest countries in the world (Tanzania and
Uganda), with individuals from the poor countries being more egalitarian than individuals from the rich countries. The association between low SES and egalitarianism in
the present paper shows a similar relationship among adolescents within a country.
References
Almås, Ingvild, Alexander W. Cappelen, Kjell G. Salvanes, Erik Ø. Sørensen, and
Bertil Tungodden (Forthcoming). “Willingness to compete: Family matters,” Management Science.
Almås, Ingvild, Alexander W. Cappelen, Erik Ø. Sørensen, and Bertil Tungodden (2010). “Fairness and the development of inequality acceptance,” Science,
328(5982): 1176–1178.
Babcock, Linda and George Loewenstein (1997). “Explaining bargaining impasse:
The role of self-serving biases,” Journal of Economic Perspectives, 11(1): 109–126.
Babcock, Linda, George Loewenstein, Samuel Issacharoff, and Colin Camerer (1995).
“Biased judgement of fairness in bargaining,” American Economic Review, 85(5):
1337– 1343.
Bauer, Michal, Julie Chytilová, and Barbara Pertold-Gebicka (2014). “Parental background and other-regarding preferences in children,” Experimental Economics,
17(1): 24–46.
10
Benenson, Joyce F., Joanna Pascoe, and Nicola Radmore (2007). “Children’s altruistic
behavior in the dictator game,” Evolution and Human Behavior, 28(3): 168–175.
Bolton, Gary E. and Axel Ockenfels (2000). “ERC: A theory of equity, reciprocity,
and competition,” American Economic Review, 90(1): 166–193.
Cappelen, Alexander W., Astri Drange Hole, Erik Ø. Sørensen, and Bertil Tungodden
(2007). “The pluralism of fairness ideals: An experimental approach,” American
Economic Review, 97(3): 818–827.
Cappelen, Alexander W., James Konow, Erik Ø. Sørensen, and Bertil Tungodden
(2013a). “Just luck: An experimental study of risk taking and fairness,” American
Economic Review, 103(3): 1398–1413.
Cappelen, Alexander W., Karl O. Moene, Erik Ø. Sørensen, and Bertil Tungodden
(2013b). “Needs versus entitlements: An international fairness experiment,” Journal
of the European Economic Association, 11(3): 574–598.
Cappelen, Alexander W., Erik Ø. Sørensen, and Bertil Tungodden (2010). “Responsibility for what? Fairness and individual responsibility,” European Economic Review,
54(3): 429–441.
Cherry, Todd L., Peter Frykblom, and Jason F. Shogren (2002). “Hardnose the dictator,” American Economic Review, 92(4): 1218–1221.
Damon, W. (1977). The Social World of the Child, San Francisco: Jossey-Bass.
Dana, Jason, Roberto A. Weber, and Jason Xi Kuang (2007). “Exploiting moral wiggle
room: Experiments demonstrating an illusory preference for fairness,” Economic
Theory, 33(1): 67–80.
Eisenberg, Nancy, Gustavo Carlo, Bridget Murphy, and Patricia van Court (1995).
“Prosocial development in late adolescence: A longitudinal study,” Child Development, 66(4): 1179–1197.
Engel, Christoph (2011). “Dictator games: A meta study,” Experimental Economics,
14(4): 583–610.
Engelmann, Dirk and Martin Strobel (2004). “Inequality aversion, efficiency, and maximin preferences in simple distribution experiments,” American Economic Review,
94(4): 857–869.
Fehr, Ernst, Helen Bernhard, and Bettina Rockenbach (2008). “Egalitarianism in
young children,” Nature, 454(7208): 1079–1083.
Fehr, Ernst, Daniela Glätzle-Rützler, and Matthias Sutter (2013). “The development
of egalitarianism, altruism, spite and parochialism in childhood and adolescence,”
European Economic Review, 64(1): 369–383.
11
Fehr, Ernst and Klaus M. Schmidt (1999). “A theory of fairness, competition and cooperation,” Quarterly Journal of Economics, 114(3): 817–868.
Frohlich, Norman, Joe Oppenheimer, and Anja Kurki (2004). “Modeling otherregarding preferences and an experimental test,” Public Choice, 119(1-2): 91–117.
Gilligan, Carol (1982). In a Different Voice, Harvard University Press.
Harbaugh, William T, Kate Krause, and Lise Vesterlund (2007). “Learning to bargain,”
Journal of Economic Psychology, 28(1): 127–142.
Henrich, Joseph, Robert Boyd, Samuel Bowles, Colin Camerer, Ernst Fehr, Herbert
Gintis, Richard McElreath, Michael Alvard, Abigail Barr, Jean Ensminger, Natalie Smith Henrich, Kim Hill, Francisco Gil-White, Michael Gurven, Frank W.
Marlowe, John Q. Patton, and David Tracer (2005). “”Economic man” in crosscultural perspective: Behavioral experiments in 15 small-scale societies,” Brain and
Behavioral Science, 28(6): 795–815.
Hook, J. G. and Thomas D. Cook (1979). “Equity theory and the cognitive ability of
children,” Psychological Bulletin, 86(3): 429–445.
Kohlberg, Lawrence (1984). Essays in Moral Development. Vol II: The Psychology of
Moral Development, San Francisco: Harper and Row.
Konow, James (2000). “Fair shares: Accountability and cognitive dissonance in allocation decisions,” American Economic Review, 90(4): 1072–1091.
Messick, David M. and Keith Sentis (1983). “Fairness, preference and fairness biases,”
in David M. Messick and Karen S. Cook (eds.), “Equity Theory: Psychological and
Sociological Perspectives,” New York: Praeger, pp. 61–94.
Mikula, Gerold (1972). “Die entwicklung des gewinneaufteilungsverhaltenes bei
kindern und jugendlichen: Eine untersuchung an 5-,7-,9- und 11 jährigen,”
Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie, 4(3): 151–
164.
Moore, Colleen F., Sheri E. Hembree, and Robert D. Enright (1993). “The unfolding of
justice: A developmental perspective on reward allocation,” in Barbara A. Mellers
and Jonathan Baron (eds.), “Psychological Perspectives on Justice,” chapter 9, Cambridge University Press, pp. 183–204.
Piaget, Jean (1965). The Moral Judgement of the Child, New York: Free Press.
Sutter, Matthias (2007). “Outcomes versus intentions: On the nature of fair behavior
and its development with age,” Journal of Economic Psychology, 28(1): 69–78.
Sutter, Matthias and Martin G. Kocher (2007). “Trust and trustworthiness across different age groups,” Games and Economic Behavior, 59(2): 364–382.
12
Figure 1: Descriptive statistics
Spectator choice
.6
0
.2
.4
.6
share given
.8
1
0
0
.2
Fraction
.4
Share egalitarian
.2
.4
.6
Stakeholder choice (all)
boys
Fraction
.4
.2
0
0
.2
Fraction
.4
.6
Stakeholder choice (girls)
.6
Stakeholder choice (boys)
girls
0
.2
.4
.6
share given
.8
1
0
.4
.6
share given
.8
1
.2
Fraction
.4
.6
Stakeholder choice (meritocrats)
0
0
.2
Fraction
.4
.6
Stakeholder choice (egalitarian)
.2
0
.2
.4
.6
share given
.8
1
0
.2
.4
.6
share given
.8
1
Note: The figure reports descriptive statistics for the stakeholder choice and the spectator choice; “share given” is the share of13
the available money given to the other participant in the spectator choice, ’share egalitarian’ is the share that divides equally in
the spectator choice.
0
.2
share given
.4
.6
Figure 2: Differences in weight attached to fairness by SES
low SES
m/h SES
Note: The figure reports the average share given to the other participant in the stakeholder choice by family background. A family is defined as low SES if the family
is in the bottom fifth of both the income and the education distribution, where family
income is measured as the total income of the father and the mother and family education is measured as the total years of education of the father and the mother. The
standard errors are indicated.
14
0
.2
share egalitarian
.4
.6
Figure 3: Difference in fairness views by SES
low SES
m/h SES
Note: The figure reports the share of the participants who have an egalitarian fairness
view and chose an equal split in the spectator choice by family background. A family
is defined as low SES if the family is in the bottom fifth of both the income and the
education distribution, where family income is measured as the total income of the
father and the mother and family education is measured as the total years of education
of the father and the mother. The standard errors are indicated.
15
Table 1: Parental background: Comparison of sample and population
Father
Mother
population
sample
population
sample
A. Education
Only compulsory education
0.158
0.151
(0.016)
0.170
0.169
(0.017)
Some secondary education
0.374
0.408
(0.022)
0.334
0.345
(0.021)
0.46.82
0.441
(0.022)
0.496
0.486
(0.022)
B. Income
Mean income
632
624
(22)
407
364
(9)
10th percentile
279
302
(17)
190
173
(10)
25th percentile
394
403
(10)
286
245
(8)
50th percentile
519
527
(16)
381
342
(8)
75th percentile
736
754
(22)
481
433
(9)
90th percentile
1049
1009
(33)
615
541
(28)
At least some college
Note: In panel A, we report the share of individuals in each category of education,
where “population” refers to the full population having children in the 1996 cohort in
Bergen, weighted by the number of such children, and “sample” refers to the parents
of the 483 participants for whom we have data on family background. In panel B,
we report the mean and the distribution of income in thousands 2009 NOK (from
administrative data, using the social insurance definition of income). Standard errors
in parentheses (for the sample only).
16
Table 2: Regressions of share given on background characteristics
1
Low SES
2
-0.028
(0.041)
Female
3
-0.028
(0.041)
0.004
(0.022)
Observations
R2
5
-0.029
(0.041)
0.004
(0.022)
Productive
Constant
4
-0.016
(0.022)
-0.016
(0.022)
0.311∗∗∗
(0.011)
0.306∗∗∗
(0.015)
0.309∗∗∗
(0.016)
0.317∗∗∗
(0.016)
0.319∗∗∗
(0.016)
477
0.001
477
0.000
477
0.001
477
0.001
477
0.002
Standard errors in parentheses
∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01
Note: The table reports robust OLS regressions of the share given to the other participant in the stakeholder choice on the participants’ background characteristics. “Low
SES” is a dummy for the participant’s family being in the bottom fifth of both the income and the education distribution, where family income is measured as the sum of
the income of the father and the mother and family education is measured as the sum
of the years of education of the father and the mother. “Female” is a dummy for the
participant being a girl. “Productive” is a dummy for the participant having at least as
high score as the median in his or her session. Robust standard errors are reported in
parentheses.
17
Table 3: Regressions of fairness view on background characteristics
1
Low SES
2
0.293∗∗∗
(0.083)
Female
3
0.293∗∗∗
(0.083)
0.002
(0.040)
Observations
R2
5
0.288∗∗∗
(0.083)
0.004
(0.040)
Productive
Constant
4
-0.101∗∗
(0.041)
-0.097∗∗
(0.040)
0.245∗∗∗
(0.020)
0.268∗∗∗
(0.028)
0.243∗∗∗
(0.028)
0.323∗∗∗
(0.031)
0.298∗∗∗
(0.031)
483
0.032
483
0.000
483
0.032
483
0.013
483
0.044
Standard errors in parentheses
∗ p < 0.10, ∗∗ p < 0.05, ∗∗∗ p < 0.01
Note: The table reports robust OLS regressions of the indicator value “Egalitarian”,
taking the value one if the participant chose to divide equally in the spectator choice
and zero otherwise, on the participants’ background characteristics. “Low SES” is a
dummy for the participant’s family being in the bottom fifth of both the income and the
education distribution, where family income is measured as the sum of the income of
the father and the mother and family education is measured as the sum of the years of
education of the father and the mother. “Female” is a dummy for the participant being
a girl. “Productive” is a dummy for the participant having at least as high a score as
the median in his or her session. Robust standard errors are reported in parentheses.
18
Table A1: Regressions of share given on SES: alternative definitions of low SES
Father and mother
Only father
Only mother
Only income
Only education
Low SES
-0.028
(0.041)
-0.046
(0.044)
-0.006
(0.047)
-0.032
(0.028)
0.019
(0.027)
Constant
0.311∗∗∗
(0.011)
0.312∗∗∗
(0.011)
0.310∗∗∗
(0.011)
0.314∗∗∗
(0.012)
0.305∗∗∗
(0.012)
477
0.001
470
0.002
475
0.000
477
0.003
477
0.001
N
R2
Standard errors in parentheses
∗ p < 0.1, ∗∗ p < 0.05, ∗∗∗ p < 0.01
Note: The table reports robust OLS regressions of the share given in the stakeholder
choice on the participants’ SES background. The five columns rely on different definitions of low SES. The first column uses our main definition of a low SES family:
that the family is in the bottom fifth of both the income and the education distribution, where family income is measured as the sum of the income of the father and the
mother and family education is measured as the sum of the years of education of the
father and the mother. In the second column, low SES is defined as having a father
in the bottom fifth of both the income and the education distribution. In the third column low SES is defined as having a mother in the bottom fifth of both the income and
the education distribution. In the forth column low SES is defined as having family
income in the bottom fifth of the income distribution. In the fifth column low SES is
defined as having family education in the bottom fifth of the education distribution.
The difference in number of observations is due to missing observations for father’s
income and education and mother’s income and education. Robust standard errors are
reported in parentheses.
A
Appendix: Additional tables
19
Table A2: Regressions of egalitarianism on SES: alternative definitions of low SES
Father and mother
Only father
Only mother
Only income
Only education
Low SES
0.293∗∗∗
(0.083)
0.165∗
(0.089)
0.042
(0.093)
0.111∗∗
(0.055)
0.189∗∗∗
(0.055)
Constant
0.245∗∗∗
(0.020)
0.260∗∗∗
(0.021)
0.266∗∗∗
(0.021)
0.248∗∗∗
(0.022)
0.232∗∗∗
(0.021)
483
0.032
476
0.009
481
0.000
483
0.010
483
0.029
N
R2
Standard errors in parentheses
∗ p < 0.1, ∗∗ p < 0.05, ∗∗∗ p < 0.01
Note: The table reports robust OLS regressions of the indicator value “Egalitarian”
(taking the value one if the participant chose to divide equally in the spectator choice
and zero otherwise) on the participants’ SES background. The five columns rely
on different definitions of low SES. The first column uses our main definition of a
low SES family: that the family is in the bottom fifth of both the income and the
education distribution, where family income is measured as the sum of the income
of the father and the mother and family education is measured as the sum of the
years of education of the father and the mother. In the second column, low SES is
defined as having a father in the bottom fifth of both the income and the education
distribution. In the third column low SES is defined as having a mother in the bottom
fifth of both the income and the education distribution. In the forth column low SES
is defined as having family income in the bottom fifth of the income distribution. In
the fifth column low SES is defined as having family education in the bottom fifth of
the education distribution. The difference in number of observations is due to missing
observations for father’s income and education and mother’s income and education.
Robust standard errors are reported in parentheses.
20
Table A3: Regressions of share given and egalitarian on income and education (running variables)
share given
Income
N
R2
egalitarian
egalitarian
-0.155∗∗∗
(0.050)
0.049
(0.030)
Education
Constant
share given
-0.020∗∗
(0.009)
0.003
(0.004)
-0.318
(0.391)
0.270∗∗∗
(0.065)
2.268∗∗∗
(0.653)
0.549∗∗∗
(0.131)
477
0.006
477
0.001
483
0.017
483
0.011
Standard errors in parentheses
∗ p < 0.1, ∗∗ p < 0.05, ∗∗∗ p < 0.01
Note: The table reports results from robust OLS regressions of share given in the
stakeholder choice and of choosing an equal split in the spectator choice on log of total
income and years of education. Robust standard errors are reported in parentheses.
Table A4: Regressions of share given and egalitarian on number of siblings and indicator for whether participant is first born
share given
Number of siblings
-0.011
(0.011)
First born
Constant
N
R2
share given
egalitarian
egalitarian
0.016
(0.019)
0.024
(0.022)
0.024
(0.041)
0.330∗∗∗
(0.023)
0.298∗∗∗
(0.014)
0.239∗∗∗
(0.041)
0.260∗∗∗
(0.026)
477
0.003
477
0.003
483
0.001
483
0.001
Standard errors in parentheses
∗ p < 0.1, ∗∗ p < 0.05, ∗∗∗ p < 0.01
Note: The table reports results from robust OLS regressions given in the stakeholder
choice and of choosing an equal split in the spectator choice on two characteristics of
the family: the number of siblings and whether the participant is first born. Robust
standard errors are reported in parentheses.
21
Issued in the series Discussion Papers 2014
2014
01/14 January, Kurt R. Brekke, Tor Helge Holmås, and Odd Rune Straume, “Price
Regulation and Parallel Imports of Pharmaceuticals”.
02/14 January, Alexander W. Cappelen, Bjørn-Atle Reme, Erik Ø. Sørensen, and
Bertil Tungodden, “Leadership and incentives”.
03/14 January, Ingvild Almås, Alexander W. Cappelen, Kjell G. Salvanes, Erik Ø.
Sørensen, and Bertil Tungodden, “Willingness to Compete: Family Matters”.
04/14 February, Kurt R. Brekke, Luigi Siciliani, and Odd Runde Straume,
“Horizontal Mergers and Product Quality”.
05/14 March, Jan Tore Klovland, “Challenges for the construction of historical price
indices: The case of Norway, 1777-1920”.
06/14 March, Johanna Möllerström, Bjørn-Atle Reme, and Erik Ø. Sørensen, “Luck,
Choice and Responsibility”.
07/14 March, Andreea Cosnita-Langlais and Lars Sørgard, “Enforcement vs
Deterrence in Merger Control: Can Remedies Lead to Lower Welfare?”
08/14 March, Alexander W. Cappelen, Shachar Kariv, Erik Ø. Sørensen, and Bertil
Tungodden, «Is There a Development Gap in Rationality?”
09/14 April, Alexander W. Cappelen, Ulrik H. Nielsen, Bertil Tungodden, JeanRobert Tyran, and Erik Wengström, “Fairness is intuitive”.
10/14 April, Agnar Sandmo, “The early history of environmental economics”.
11/14 April, Astrid Kunze, “Are all of the good men fathers? The effect of having
children on earnings”.
12/14 April, Agnar Sandmo, “The Market in Economics: Behavioural Assumptions
and Value Judgments”.
13/14 April, Agnar Sandmo, “Adam Smith and modern economics”.
14/14 April, Hilde Meersman, Siri Pettersen Strandenes, and Eddy Van de Voorde,
“Port Pricing: Principles, Structure and Models”.
15/14 May, Ola Honningdal Grytten, “Growth in public finances as tool for control:
Norwegian development 1850-1950”
16/14 May, Hans Jarle Kind, Tore Nilssen, and Lars Sørgard, “Inter-Firm Price
Coordination in a Two-Sided Market”.
17/14 May, Stig Tenold, “Globalisation and maritime labour in Norway after World
War II”.
18/14 May, Tunç Durmaz, “Energy Storage and Renewable Energy”
19/14 May, Elias Braunfels, “How do Political and Economic Institutions Affect
Each Other?”
20/14 May, Arturo Ormeño and Krisztina Molnár, “Using Survey Data of Inflation
Expectations in the Estimation of Learning and Rational Expectations Models”
21/14 May, Kurt R. Brekke, Luigi Siciliani, and Odd Rune Straume, “Hospital
Mergers with Regulated Prices”.
22/14 May, Katrine Holm Reiso, “The Effect of Welfare Reforms on Benefit
Substitution”.
23/14 June, Sandra E. Black, Paul J. Devereux, and Kjell G. Salvanes, “Does grief
transfer across generations? In-utero deaths and child outcomes”
24/14 June, Manudeep Bhuller, Magne Mogstad, and Kjell G. Salvanes, «Life Cycle
Earnings, Education Premiums and Internal Rates of Return”.
25/14 June, Ragnhild Balsvik, Sissel Jensen, and Kjell G. Salvanes, “Made in
China, sold in Norway: Local labor market effects of an import shock”.
26/14 August, Kristina Bott, Alexander W. Cappelen, Erik Ø. Sørensen, and Bertil
Tungodden, “You’ve got mail: a randomized field experiment on tax evasion”
27/14 August, Alexander W. Cappelen, Sebastian Fest, Erik Ø. Sørensen, and
Bertil Tungodden, “The freedom to choose undermines the willingness to
redistribute.”
28/14 August, Marianne Bertrand, Sandra Black, Sissel Jensen, and Adriana LlerasMuney, “Breaking the Glass Ceiling? The Effect of Board Quotas on Female
Labor Market Outcomes in Norway.”
29/14 August, Astrid Kunze, “The family gap in career progression”.
30/14 September, Ragnhild Balsvik and Morten Sæthre, “Rent Sharing with
Footloose Production. Foreign Ownership and Wages Revisited”.
31/14 October, Nicola D. Coniglio and Giovanni Pesce, “Climate Variability and
International Migration: an empirical analysis”
32/14 November, Kurt R. Brekke, Armando J. Garcia Pires, Dirk Schindler, and
Guttorm Schjelderup, “Capital Taxation and Imperfect Competition: ACE vs.
CBIT”
33/14 November, Jan I. Haaland and Anthony J. Venables, “Optimal trade policy
with monopolistic competition and heterogeneous firms”.
34/14 December, Rolf Aaberge, Kai Liu, and Yu Zhu, “Political Uncertainty and
Household Savings”.
2015
01/15 January, Antonio Mele, Krisztina Molnár, and Sergio Santoro, “On the perils
of stabilizing prices when agents are learning”.
02/15 March, Liam Brunt, “Weather shocks and English wheat yields, 1690-1871”.
03/15 March, Kjetil Bjorvatn, Alexander W. Cappelen, Linda Helgesson Sekei, Erik
Ø. Sørensen, and Bertil Tungodden, “Teaching through television:
Experimental evidence on entrepreneurship education in Tanzania”.
04/15 March, Kurt R. Brekke, Chiara Canta, Odd Rune Straume, “Reference pricing
with endogenous generic entry”.
05/15 March, Richard Gilbert and Eirik Gaard Kristiansen, “Licensing and
Innovation with Imperfect Contract Enforcement”.
06/15 March, Liam Brunt and Edmund Cannon, “Variations in the price and quality
of English grain, 1750-1914: quantitative evidence and empirical implications”.
07/15 April, Jari Ojala and Stig Tenold, “Sharing Mare Nostrum: An analysis of
Mediterranean maritime history articles in English-language journals”.
08/15 April, Bjørn L. Basberg, “Keynes, Trouton and the Hector Whaling Company.
A personal and professional relationship”.
09/15 April, Nils G. May and Øivind A. Nilsen, “The Local Economic Impact of
Wind Power Deployment”.
10/15 May, Ragnhild Balsvik and Stefanie Haller, “Ownership change and its
implications for the match between the plant and its workers”.
11/15 June, Kurt R. Brekke, Chiara Canta, Odd Rune Straume, “Does Reference
Pricing Drive Out Generic Competition in Pharmaceutical Markets? Evidence
from a Policy Reform”.
12/15 June, Kurt R. Brekke, Tor Helge Holmås, Karin Monstad, and Odd Rune
Straume, “Socioeconomic Status and Physicians’Treatment Decisions”.
13/15 June, Bjørn L. Basberg, “Commercial and Economic Aspects of Antarctic
Exploration ‐ From the Earliest Discoveries into the 19th Century”.
14/15 June, Astrid Kunze and Amalia R. Miller, “Women Helping Women?
Evidence from Private Sector Data on Workplace Hierarchies”
15/15 July, Kurt R. Brekke, Tor Helge Holmås, Karin Monstad, Odd Rune Straume,
«Do Treatment Decisions Depend on Physicians Financial Incentives?”
16/15 July, Ola Honningdal Grytten, “Norwegian GDP by industry 1830-1930”.
17/15 August, Alexander W. Cappelen, Roland I. Luttens, Erik Ø. Sørensen, and
Bertil Tungodden, «Fairness in bankruptcy situations: an experimental study».
18/15 August, Ingvild Almås, Alexander W. Cappelen, Erik Ø. Sørensen, and
Bertil Tungodden, “Fairness and the Development of Inequality Acceptance”.
19/15 August, Alexander W. Cappelen, Tom Eichele,Kenneth Hugdah, Karsten
Specht, Erik Ø. Sørensen, and Bertil Tungodden, “Equity theory and fair
inequality: a neuroconomic study”.
20/15 August, Frank Jensen and Linda Nøstbakken, «A Corporate-Crime
Perspective on Fisheries: Liability Rules and Non-Compliance”.
21/15 August, Itziar Lazkano and Linda Nøstbakken, “Quota Enforcement and
Capital Investment in Natural Resource Industries”.
22/15 October, Ole-Petter Moe Hansen and Stefan Legge, “Trading off Welfare and
Immigration in Europe”.
23/15 October, Pedro Carneiro, Italo Lopez Garcia, Kjell G. Salvanes, and Emma
Tominey, “Intergenerational Mobility and the Timing of Parental Income”.
24/15 October, David Figlio, Krzysztof Karbownik, and Kjell G. Salvanes,
“Education Research and Administrative Data”.
25/15 October, Ingvild Almås, Alexander W. Cappelen, Kjell G. Salvanes, Erik Ø.
Sørensen, and Bertil Tungodden: «Fairness and family background».
Norges
Handelshøyskole
Norwegian School of Economics
NHH
Helleveien 30
NO-5045 Bergen
Norway
Tlf/Tel: +47 55 95 90 00
Faks/Fax: +47 55 95 91 00
[email protected]
www.nhh.no