Moral distance in dictator games

Judgment and Decision Making, Vol. 3, No. 4, April 2008, pp. 344–354
Moral distance in dictator games
Fernando Aguiar?1, Pablo Brañas-Garza2, and Luis M. Miller1,3
1 Instituto de Estudios Sociales Avanzados,
Consejo Superior de Investigaciones Cientí?cas (IESA-CSIC)
2 Departamento de Teoría Económica, Universidad de Granada
3 Strategic Interaction Group, Max Planck Institute of Economics, Jena
Abstract
We perform an experimental investigation using a dictator game in which individuals must make a moral decision —
to give or not to give an amount of money to poor people in the Third World. A questionnaire in which the subjects
are asked about the reasons for their decision shows that, at least in this case, moral motivations carry a heavy weight
in the decision: the majority of dictators give the money for reasons of a consequentialist nature. Based on the results
presented here and of other analogous experiments, we conclude that dicator behavior can be understood in terms of
moral distance rather than social distance and that it systematically deviates from the egoism assumption in economic
models and game theory.
Keywords: dictator game, moral distance, moral motivations, experimental economics.
1 Introduction
the dictator makes the decision under conditions of abso-
lute privacy and anonymity and has no information about
In standard dictator game (DG) experiments, one of the
the recipient, nearly no one donates anything. Donations
players, the dictator, is given a ?xed amount of money
tend to be very low, around 10% of the pie on average
and a set of instructions. The instructions explain that the
(Hoffman et al., 1994; Hoffman et al., 1996). However,
money has been assigned to two players — the dictator
when the dictator receives reliable information regarding
and the recipient — but that only the dictator is entitled
the recipient, even while keeping anonymity constant, do-
to decide how much money to keep and how much money
nations increase. When no information is available about
to give to the recipient. In this simple situation, there are
the recipient, the dictator may have doubts as to the recip-
at least three sources of information relevant to the de-
ient’s existence and therefore have no reason to share the
cision: i) information related to the dictator herself, i.e.,
money. In contrast, when dictators were shown pictures
whether or not the decision is observed by others; ii) in-
of the recipients in a particular experiment, up to 25% of
formation the dictator receives about the recipient; and
the subjects gave as much as half of the total amount, al-
iii) information derived from the game framing and the
though 58% of them kept all the money for themselves
language used in the instructions.
(Burnham, 2003). In another experiment, the dictators
Previous papers have shown that the outcome of a DG
were told that their donation would be given to the Red
changes considerably depending on the type of informa-
Cross (Eckel & Grossman, 1996). Thirty-one percent of
tion provided by the experimenters. For instance, when
the dictators in this experiment gave part of the money
(17% gave half of the amount), while 10% gave the full
?The authors thank the stimulus and motivation given by the JDM
amount. Offers also increased when the dictators were
editor, Jonathan Baron. We gratefully acknowledge Andrés de Fran-
told the recipient’s surname (Charness & Gneezy, 2003),
cisco and two anonymous referees for constructive advice that induced
major revisions of the article. Financial support received from the
when they received proof of the actual existence of recip-
Instituto de Estudios Sociales Avanzados (IESA-CSIC) is acknowl-
ients, or when they were told that the recipient was a fel-
edged. Pablo Brañas-Garza and Luis M. Miller acknowledge also
low classmate (Frohlich & Oppenheimer, 2001). These
the ?nantial support received from CICYT (SEJ2004-07554/ECO and
SEJ2006-00959/SOCI, respectively). Addresses: Fernando Aguiar,
experiments have shown that if conditions of anonymity
IESA-CSIC, Campo Santo de los Mártires 7, 14004, Córdoba (Spain);
are relaxed, e.g., dictator’s behavior or relevant informa-
email: faguiar@iesa.csic.es; Pablo Brañas-Garza Departamento de
tion about the recipients are revealed, donations will be
Teoría Económica, Universidad de Granada, Campus de la Cartuja,
higher, although the full amount of money will never be
18011, Granada (Spain); email: pbg@ugr.es; Luis M. Miller, Strate-
gic Interaction Group, Max Planck Institute of Economics, Kahlaische
donated — with the exception of the Red Cross experi-
Str. 10, 07745, Jena (Germany); email: miller@econ.mpg.de.
ment.
344

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
345
The language used in the instructions is yet another
tators might be motivated to give a gift for one or more
factor that may determine the outcome of the experiment.
of many reasons: e.g., a sense of moral obligation, plea-
For example, the fact that subjects are told that they are
sure derived from giving, to demonstrate kindness” (283).
participating in an experiment may make them lose inter-
This hypothesis, however, encompasses very distinct mo-
est in their partner, or may cause subjects to feel that they
tivations that must be analyzed separately. Although the
are playing a game — that it is a competition and they
pleasures of giving or demonstrating kindness are greatly
must therefore win, that is, keep all or part of the money
affected by anonymity — in that the recipient not know-
for themselves (Frohlich et al., 2004). The nature of the
ing if the dictator gives or not takes away the pleasure
payments, the presence or absence of the experimenters,
of giving and does not allow the dictator to demonstrate
the possibility of taking part in future experiments — and
that she is kind — moral obligation is not affected by
thus winning more money — may also determine the out-
anonymity but rather by the lack of information about the
come (Bolton et al., 1998).
recipient’s actual situation. In the classic Hoffman et al.
The aim of these experiments is to measure the proso-
(1994) experiment, and in most experiments along these
cial conduct or other-regarding preferences revealed by
lines, the complete lack of information about the recipi-
the subjects, thus challenging the assumption of self-
ent removes the necessary conditions to donate out of a
interested behavior. The supposition that individuals will
sense of moral obligation:
attempt to maximize their own bene?ts, a standard as-
By imposing double anonymity, HMSS [Hoff-
sumption in economics, could lead to the conclusion that
man et al., 1994] have removed virtually all
no one will give anything in the DG since giving noth-
motivation for donating money to one’s part-
ing is the behavior that maximizes the dictator material’s
ner. The decision makers cannot identify each
payoff. However, as we have seen, subjects often deviate
other, nor do they have enough information to
from this prediction in the laboratory setting. In light of
know if their partner is poor or otherwise de-
these experimental results, several hypotheses have been
serving of their generosity; thus there is little
developed to explain these deviations.
or no basis for altruism to play a part in the
Hoffman et al. (1996) show that as anonymity is re-
decision. We assert that fairness and altruism
laxed, donations increase. In the opinion of these au-
require context: the circumstances of the recip-
thors, this is due to the fact that anonymity generates “so-
ient determine what is the fair or appropriate
cial distance,” which they de?ne as “the degree of reci-
charitable action to take. In the Hoffman et al.
procity that subjects believe exists within a social interac-
(1994) study, the information available to the
tion” (Hoffman et al., 1996, p. 654). The social isolation
Proposer is unlikely to motivate charitable be-
or lack of sense of community produced by anonymity
havior (Eckel & Grossman, 1996, p. 183-184).
leads to the predominance of self-interested behavior.
Nonetheless, the hypothesis of social distance has been
If, as Eckel & Grossman have shown, altruistic behav-
widely criticized. The concept of reciprocity, for exam-
ior requires an appropriate context, which is inexistent
ple, is employed in a very broad sense since in reality
in most experiments, what would happen if the dictator
the DG does not pose so much a problem of reciprocity
makes a decision in an easily recognizable moral context;
— in that the recipient can do nothing- as one of subject
a context in which dictators have information, for exam-
identi?ability. Indeed, it is identi?ability that determines
ple, about their partner’s poverty or illness? In Eckel &
behavior (Bohnet & Frey, 1999).
Grossman’s experiment, in which the partner is the Red
Many experiments have shown that it is not so much
Cross, donations increase, but not in a spectacular way.
the degree of dictator anonymity that determines dona-
Donations increase much more when the money that the
tions, but rather the information that the dictator has
dictator can distribute is the result of the joint effort of
about the recipient. Thus, for example, in an exten-
both the dictator and the partner, given that, in this case,
sive article in which several hypotheses are reviewed to
“social pressures, moral imperatives, and the warm glow
explain donations in dictator games and others are pro-
of giving are likely to magnify the reward to skillful or
posed, Bolton et al. (1998) argue that, if I know noth-
deserving recipients” (Ruf?e, 1998, p. 259). As Ruf?e
ing about the other person and if I give her the money,
points out, the results of this experiment contradict Hoff-
I can err in her favor, but I prefer to err in my favor be-
man et al.’s claim that other-regarding preferences have
cause I am not a saint (the “I’m-no-saint hypothesis”).
to do with “an expectations phenomenon” and not with
According to these authors, however, the hypotheses that
“an autonomous private preference for equity” (Hoffman
explain DG donations are unclear when the dictator has
et al., 1994, p. 348).
information about the recipient. It should be noted that
Moral preferences, the sense of moral obligation,
Bolton et al. (1998) do not ?nd evidence for what they
moral imperatives or equity appear repeatedly in discus-
call the “kindness hypothesis,” according to which “dic-
sions on the outcomes of a diverse range of DG exper-

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
346
iments. Yet, in spite of the fact that donations occur in
instructions and a sheet with 341 numerical codes, three
these experiments, we cannot speak with any certainty
of which were marked with a circle. The size of the en-
about dictators’ reasons for giving, as the only thing we
velope permitted the subjects to carry out the operations
know for sure is the preference that has been revealed in
in complete privacy. The money and the questionnaire
the decision, that is, whether the dictator gave or did not
were placed in the small envelope where the centers were
give. Indeed, aside from the theoretical speculations of
marked. The subjects then stood up and placed the enve-
experimenters, we do not know why dictators give. With
lope in a box. The subjects’ names did not appear any-
a view to bridging this gap, in the following section we
where and they kept the large envelope. After they were
present two experiments (a hypothetical one and a real-
given the instructions they were told the following: “Only
payoffs one) in which the moral context is easily recog-
you will know what is in the large envelope” (the money
nizable and in which dictators are expressly asked about
that remained). After allocating the money, the subjects
the reasons for their action. We can then determine if
stood up, placed the small, sealed envelope (containing
the “kindness hypothesis” is to some degree certain: ei-
their choice, the money and the questionnaire) in the box
ther because the dictators take pleasure in giving, give
and left the room. The subjects were not paid for partici-
to demonstrate kindness, or give out of a sense of moral
pating in the experiment. None of the researchers (exper-
obligation. In the third section, we discuss these results
imenters) that ran the experiment were professors at the
and defend the hypothesis that dictators’ responses are
universities where the experiments were performed and
determined by the “moral distance” between the dictator
had not had any previous contact with the students.
and the recipient, a distance that permits us to classify the
In the instructions, the students were informed that the
behaviors that may arise in the dictator game.
341 codes appearing on the sheet referred to centers that
receive medicines in Asia, Africa and South America,
that they had been assigned three of these centers, and
2 Two moral dictator games: De- that for every 5C they donated, a box of medicines would
sign and results
be sent to one of “their” three centers. All of the sub-
jects in the experiment had to write the three codes on
To analyze how much the number of donors and the size
their small envelope and circle the centers — a maximum
of donations increase in a moral context, we need to study
of three — to which they allotted the 5C in medicines.
a DG situation in which the moral structure of the prob-
The subjects were also told that they were entitled to keep
lem is set out in a clear manner. This requires dicta-
5C for participating in the research study on the distribu-
tors having relevant moral information about the recipi-
tion of goods. Logically, if they kept 5C for themselves,
ent, that is, a context in which they are able to identify
one of the centers would not receive any medicines (in
the moral nature of their decision. The instructions must
this case they were told to circle only two of the three
prevent dictators from thinking that they are just play-
centers that had been assigned to them).2 The procedure
ing a game, or that the recipients do not actually exist.
was performed under conditions of complete anonymity
The two experiments reported in Brañas-Garza (2006)
by means of a double-blind mechanism similar to that de-
and presented below ful?l these conditions.
scribed by Hoffman et al. (1996).
The questionnaire included questions regarding sex,
age, the weekly allowance given to them by their parents
2.1 Design and quantitative results of the
(none of the experimental subjects — all of whom were
experiments1
university students — had a paying job), and the reasons
The ?rst experiment was conducted with 77 students from
for their decision. This last question reads: “Could you
the University of Córdoba (Spain) in November 2002.
please tell me the reasons for your decision in the prob-
Subjects were asked to collaborate in a research study
lem posed here?”
on the problems that arise when trying to distribute non-
Using the same instructions and under identical condi-
divisible and ?nite goods. We avoided using the term
tions of anonymity, a second experiment was performed
“experiment” in order to dispel the possible negative ef-
in March 2003 with 98 students from the University of
fects of the word. Once the study was explained, the sub-
Jaén (Spain). The subjects were placed in four groups.
jects received a large envelope containing the following
This experiment was hypothetical as the subjects were
items: a small envelope, three 5C bills, a questionnaire,
not given real money and had to decide how to divide the
1This section (2.1) is simply a summary of the data analyzed in depth
2Since there is no reason to favor one recipient over another, partici-
in Brañas-Garza (2006). This is necessary so that the reader can gain a
pants’ generous behavior may be also motivated by the desire of treating
better understanding of the analysis of the qualitative data that were not
the three recipients equaly. Although this might affect the quantitative
included in the work by Brañas-Garza and that we discuss here. For the
results, it does not affect the main point of this paper, that is, the reasons
statistical analysis of both experiments see Brañas-Garza (2006).
subjects state to justify their behavior.

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
347
experiments considered here are particularly appropriate
Table 1: Two moral dictator game experiments.
since contrary to standard experimental dictator games,
Amount
Hypothetical experiment
Real
our experiments trigger an amazing amount of altruistic
given
payoffs1
behavior. But before analyzing the claims made by the
subjects, let us recall two features of the experimental de-
C
No info
Poverty
Poverty
Poverty
sign that have contributed to the prominence of unsel?sh
+ drugs
+ drugs
behaviors. Whereas the ?rst is a just a methodological
15
0%
40.8%
68.3%
74.6%
question, the second leads us to the kind of philosophical
claim that we shall make in this paper.
10
0%
25.5%
18.3%
12.0%
Firstly, the results are somewhat special because par-
5
28.6%
11.2%
5.1%
10.7%
ticipants did not face the standard continuum choice of
0
71.4%
22.4%
8.1%
2.7%
how to divide a pie, but instead had to decide how to dis-
N
98
98
98
75
tribute three 5C bills among different potential recipients
(see Brañas-Garza 2006). As we can observe in Table
1 There are not signi?cant differences in dona-
1, the majority of participants (75%) allocate the full en-
tions between the real-payoff experiment and the
dowment to the recipients in the real-payoffs experiment.
“Poverty+Drugs” treatment of the hypothetical exper-
This is quite interesting for our analysis as it permits us to
iment (see Brañas-Garza, 2006).
explore the reasons of a large number of participants who
decided to give the money. Secondly, and more interest-
amount hypothetically by responding to a series of ques-
ingly, the experiment used a “moral frame,” the kind of
tions. Furthermore, the decision was made under three
design that allows moral motivations such as fairness and
distinct scenarios which were presented to them in a ran-
altruism to play a role (Eckel & Grossman, 1996). Thus,
dom manner:
although these are rather atypical dictator results, they are
the kind of results that permit us to explore moral motiva-
a) No information about the recipients.
tions in depth. Having clari?ed this, we will now concen-
b) Information revealing that the recipients were poor
trate on the main arguments provided by the participants.
people from the Third World.
c) Similar to b, but informing subjects that the money
2.2.1 The prominence of consequentialist arguments
would be used to buy medicines.
The main results of the experiments are shown in Ta-
Consequentialist reasons predominate among the an-
ble 1. As can be observed, the increase in the number
swers given by participants in both the real-payoffs and
of donors and donations is spectacular compared to other
the hypothetical experiment. The dictators give all the
experiments (see Section 1). As occurs in all the exper-
money because giving it has the most bene?cial outcomes
iments, when subjects lack information about the recip-
in terms of satisfying needs. In the real-payoffs exper-
ient they hardly give anything. However, as in this case
iment, 60% of the participants provide a consequential-
they knew that the recipients were poor, or poor and in
ist argument (see Table 2). The complete list of answers
need of medicines, the majority of dictators gave all the
given by participants in the real-payoffs experiment can
money — in spite of the fact that they were told in the
be found in the Appendix.3
instructions that they could legitimately keep one 5C bill
As we will see below, the immense majority of those
for participating and donate the other two. As we will
who gave all the money appealed to moral, consequen-
see below, some of them did in fact keep the money, but
tialist or deontological reasons.4 The former are more
only a few. Thus, information about partners clearly has
concerned about what would have happened had they not
an effect on donations to the same or to a larger degree
given the money (or what happens if they do), while the
than dictator anonymity. But what drives dictators to do-
latter are more concerned about what they should do. Re-
nate? What motives do they give for doing so — tastes,
sponses based on reasons of solidarity, equity, generosity
customs, traditions, religious motives, or moral reasons?
or a sense of humanity are labelled as deontological. Al-
The next subsection explores the main reasons given by
though we believe that these labels are not particularly
the participants in both experiments.
precise — dictators do not theorize, but express the rea-
sons for their decision — a clear difference can be seen
2.2 Qualitative results
3Responses were categorized by the third author.
4
As we have pointed out in the introduction, in this pa-
Of the 55 cases that justi?ed their donation in moral terms, only
three kept the 5 euros (see Table 4). We do not include them in the “le-
per we are interested in how subjects explain their behav-
gitimacy” category because they do not justify the legitimacy of keeping
ior after taking part in a dictator game. To this end, the
part of the money, but the moral reasons for giving the rest of it.

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
348
Table 2: Reasons given by subjects in both experiments.
Table 3: Moral reasons and behavior.
Real
Hypothetical1
Number of bills given
Reason
N
%
N
%
Having a reason
0
1
2
3
Total
Consequentialist
41
59.4
68
79.1
No reason
2
7
7
4
20
Deontological
14
20.3
4
4.6
Reason
0
0
3
52
55
Didn’t trust the system
4
5.8
3
3.5
Legitimacy
3
4.3
3
3.5
tify giving all the money, although on this occasion the
Random decision
3
4.3
0
0
cases of self-interest increase, while those appealing to
Didn’t trust the experiment 2
2.9
0
0
deontological reasons decrease. This may be due to the
Egoism/Hardship
2
2.9
8
9.3
hypothetical nature of the experiment as it is more likely
Total2
69
100.0
86
100.0
that dictators will keep all or part of the money when the
decision does not have real consequences: as no money
1 The reasons from the hypothetical experiment corre-
is involved, no one who is needy will lose out. The data,
spond to the treatment in which the participants have
however, do not permit us to reach de?nitive conclusions
full information. We report only this treatment to make
on this question.
it comparable with the real-payoffs experiment.
2 The number of answers is restricted to 69 and 86
in the real-payoffs and the hypothetical experiment re-
2.2.2 The importance of having a moral reason
spectively since the rest of participants did not provide
Having a (moral) reason is decisive when facing a DG
any explanation for their action.
decision. In Table 3 the behavior of those who provide a
moral reason — either consequentialist or deontological
— is compared with the behavior of those who do not
between those who underline the bene?cial consequences
provide this kind of reason.
of the action and those who appeal to the duty of giving.
Almost all the participants who give the three 5C bills
The remaining labels re?ect the various justi?cations
provide a moral reason for their behavior. Of course,
provided by the dictators for giving the full amount or
this reveals nothing about the direction of the causality
not giving anything. Thus, for example, some consider
and we cannot claim that their behavior is grounded in
it legitimate to not give all the money (“legitimacy”) be-
moral reasons. But at least we can say that when they
cause the experimental instructions permit them to do so:
have to justify their actual behavior they use a moral ar-
“You are entitled to keep ?ve euros, although this amount
gument. Moreover, the difference in giving behavior be-
will depend on your decision [remember that everything
tween those who provide a moral reason and those who
is absolutely anonymous].” It is interesting to note that
do not is statistically signi?cant (Mann-Whitney U-test:
a large majority of the experimental subjects did not take
z = ?6.65, p < 0.001).
this rule or “right” into account and instead gave all the
money, explaining their action in moral terms of a con-
sequentialist or deontological nature. Others, however,
2.2.3 Behavioral differences between moral reasons
made their decisions in a random manner, did not believe
in the experiment, did not believe that the money would
Having a moral reason to act is important in a dictator
reach its destination and thus did not give the money,
game, but are different reasons related to different types
or justi?ed their decision for clearly self-interested rea-
of behavior? In other words, does the giving behavior of
sons. Nonetheless, the justi?cations given for keeping
those who provide a consequentialist reason differs from
part or all of the money are so varied and disperse that
the behavior of those who provide a deontological rea-
they become quite insigni?cant in comparison to the con-
son? Since all the participants that provide a moral reason
sequentialist or deontological reasons revealed by those
a posteriori give at least two bills, in Table 4 we study if
who gave the full amount.
there are differences between consequentialist and deon-
These results are con?rmed by the ?ndings obtained in
tological subjects when deciding whether to allocate two
the hypothetical experiment. Similar to what occurs in
or three bills to the recipient.
the ?rst experiment, the majority of participants (79%) in
These results clearly show that the behavior of partici-
the second experiment provide a consequentialist reason
pants who provide a consequentialist reason do not differ
similar to that of the real-payoffs experiment (see Table
from the behavior of those who provide a deontological
2). Again the majority provided a moral reason to jus-
reason (Mann-Whitney U-test: z = ?0.319, p = 0.750).

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
349
Table 4: Differences in giving between consequentialists
Table 5: Considering the DG a cheap decision by moral
and deontologists.
reason.
Number of bills given
Number of bills given
Reason
2
3
Total
Reason
Not cheap
Cheap
Total
Consequentialist
2
39
41
Consequentialist
18
23
41
Deontological
1
13
14
Deontological
12
2
14
When carefully analyzing all of the answers given
2.2.4 Moral framing and consequentialist reasons
by consequentialists and deontological subjects, we ?nd
only a minor difference, yet it is one that can be interest-
ing from a philosophical point of view. There is a positive
The “moral framing” used in our experiment is a crucial
correlation between considering that the dictator experi-
part of the experimental design. Above we described how
ment constitutes a “cheap decision”5 and giving a con-
moral framing increases donations in a standard dictator
sequentialist argument (see Table 5). This correlation is
game. In addition, we explained how this moral fram-
statistically signi?cant (?2 = 7.359, p = 0.007).6 Al-
ing triggers participants’ moral reasons, especially con-
though we did not have a priori a theory about why these
sequentialist ones. In this last part of the results we use
two variables should be related, and therefore it is a post
the data from the hypothetical experiment to show how
hoc correlation, this result allows as to go further in our
consequentialist arguments are more common, the more
speculation about subjects’ moral reasons.
morally loaded the framing.
This can be explained by the fact that here we ?nd two
Figure 1 plots the reasons given by subjects to rational-
distinct types of consequentialism (Mulgan, 2001): al-
ize their behavior in the hypothetical experiment. In this
most half of those who give all the money in the real-
case, we have categorized the reasons given by subjects
payoffs experiment (49%) justify their decision in terms
when facing the three different treatments: without moral
of a non-neutral consequentialism and one-third (32%)
framing, knowing that recipients are poor and knowing
justify the reasons for their choice in terms of a neu-
this fact and additionally knowing that the money will
tral or impersonal consequentialism restricted to imper-
be used to buy medicines. Figure 1 chie?y reveals three
sonal values. Neutral consequentialism is clearly pre-
facts. First, when subjects do not receive any informa-
sented by the subjects as the maximization of the imper-
tion about the recipients, almost 70% of them use this
sonal value in statements such as “the more people that
“lack of information” argument to justify their (mainly
have medicines the better” or “this money is more useful
self-interested) behavior. Second, when they know that
in these countries” or “their scarcity.” Non-neutral con-
the money will be sent to poor countries, but do not know
sequentialism is demonstrated in the choice of an action
exactly where the money will end up, 20% of them do not
that produces the best outcome, including one which will
trust the system. In other words, they are not sure if the
bene?t the agent. The experimental subjects manifested
money will be used for a good purpose. Third, the more
this in statements such as “I’m not losing anything,” “the
information they have, the more frequent the consequen-
money isn’t mine, I didn’t earn it,” “they need it more
tialist reasons. Thus, whereas almost no subject gives a
than me,” “it’s of more use to them than to me,” “it’s not
consequentialist reason when having no information, this
going to do me any good,” “it bene?ts them more than
proportion rises to 50% when they know the recipients
me,” or “I didn’t have the money when I came and I still
are poor and to 70% when they know the money will be
don’t.” When ranking the possible outcomes from best to
spent to buy medicines. Consequentialist arguments are
worst, non-neutral consequentialists consider that keep-
backed here by ef?ciency arguments, because morality is
ing all or part of the money does not bene?t them in any
not a matter of wasting resources.
way whatsoever — donating, then, is a cheap decision,
even a morally cheap one.
Donating money for medicines reduces the uncertainty
of the dictator’s decision, improving the probability that
5This information (the decision being cheap) is also derived from
her decision would have good consequences. Therefore,
the open-ended question answered by subjects after taking part in the
consequentialist arguments increase when the money is
experimental sessions.
spent in medicines.
6A Fisher’s Exact Test provides the same p-value.

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
350
Figure 1: Reasons by treatment in the hypothetical experiment.
3 Discussion:
Social distance or it” or “it could be someone who is richer than me,” state-
moral distance?
ments which were repeated time and again to justify their
action. When information about the recipient is lacking,
it is perfectly legitimate to give priority to oneself: if the
Eckel & Grossman’s claim that if dictators are provided
money goes to someone who needs it less than I do, I
the appropriate context they will have reasons to donate
would be doing an injustice to myself. Thus it has less
is more than ful?lled in Brañas-Garza (2006), albeit the
to do with a question of social distance than the fact that
result would have to be supported by further “moral” ex-
dictators make and analyze their decision whilst bearing
periments. On the other hand, it may be too hasty to reject
in mind that they live in an unfair society in which it is
the kindness hypothesis if we focus on one aspect of this
quite likely that the unknown recipient will be better off
hypothesis, that is, dictators’ sense of moral obligation.
economically than they are. In such a case there is no
Let us explore this aspect in greater detail.
moral obligation to give anything.7
First, like in many experiments, Brañas-Garza (2006)
Drawing on Hoffman, McCabe and Smith’s concept of
?nds that dictator anonymity is not a crucial variable to
social distance, we propose a de?nition of moral distance
understanding how much and why dictators donate. The
as the degree of moral obligation that the dictator has to-
key informational variable is the knowledge that dictators
wards the recipient. If the experiment is designed in such
have about their partners. Similar to what occurs in the
a way that donating is an actual obligation, the moral dis-
majority of experiments with DGs, in our hypothetical
tance will be null and the dictator ought to donate all or
experiment, dictators hardly donate anything when the
part of her resources according to her level of well-being.
dictator lacks information about the recipient. But this is
In such a situation the dictator has the actual obligation
not due to a problem of social distance, lack of reciprocity
to donate since it “is an obligation which, in a particu-
or isolation. Those who lacked information explained
lar situation, is not superseded by any other obligation”
their decision with statements such as “I don’t know who
is going to get it, they might have more than me,” “I might
7Of course this does not exclude generosity. In our hypothetical
need it for something more important than the recipients,”
experiment no less than 28.6% of the subjects donate one-third of the
money in spite of having no information about the recipient. In Hoffman
“if I don’t have any information about who the money is
et al. (1994), eight per cent of the dictators donated half or less than half
going to or what it is going to be used for, I won’t donate
of the money.

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
351
(Crawford, 1969, p. 316; Zimmerman, 1996: 26). If the
cipient is null and the donations are also null or
particular situation is such that donating cannot be said to
very low, the dictator behaves in a morally egotis-
be an obligation at all, the moral distance will be maxi-
tical manner.
mum.
The distance between dictator and recipient is not
b) If the moral distance between the dictator and the re-
greater or lesser only as a function of the reciprocity that
cipient is maximum and the dictator donates noth-
subjects belief exist within a social interaction, as in Hoff-
ing at all, the dictator behaves as a reasonably self-
man, McCabe and Smith’s de?nition of social distance
interested individual.
(Hoffman et al., 1996, p. 654). In fact, the concept of
c) If the moral distance between the dictator and the re-
reciprocity depends on the concept of obligation — cases
cipient is maximum and the dictator gives part or all
of reciprocity are those in which if you give me, I should
of her money, she performs an act of generosity.
give you; if I help you, you should help me, etc. But
then social distance is only a case of moral distance, a
d) If the moral distance between the dictator and the re-
special case in which anonymity plays a crucial, nega-
cipient is null and the dictator gives part or all of her
tive role, for the dictator must know what the recipient
money, her behavior is morally right.
has done for him before to donate or not donate — as in
Ruf?e experiment (Ruf?e, 1998). This negative relation
Note that we are speaking about labelling dictator be-
— more anonymity less reciprocity — is not present in
havior, not about the reasons for action. We judge behav-
most instances of moral distance, for it is not true that
ior from the viewpoint of the observer (the experimenter),
more anonymity leads to a lesser obligation to donate
from the outside, from the point of view of one who ob-
when moral distance is null, and it is not true that less
serves the moral distance between the dictator and the re-
anonymity leads to more obligation to donate when the
cipient and if the dictator gives money or not. The reasons
moral distance is maxim. However, anonymity must be
for action are unknown: perhaps the dictator gave all the
ensured in order to exclude any possible in?uence of rep-
money — when, for example, moral distance is null —
utation (I give so that others see that I am giving).
because the experiment was conducted on a Monday, or
Nor is moral distance related to the personal closeness
perhaps she gives part of the money when the moral dis-
between the dictator and recipient: the dictator can do-
tance is at its maximum because she was happy. What-
nate more money to a stranger than to an acquaintance if
ever the reasons may be for action, behavior can be la-
the stranger needs it, however close they may be. Fol-
belled from the viewpoint of the experimenter, as we did
lowing upon Jonathan Glover, Abelson (2005) refers to
so above. Given the manner in which DG experiments are
moral distance as “the emotional closeness between agent
usually conducted, it is not possible to determine the rea-
and bene?ciary” and states that “our moral obligations to
sons for action, that is, the internal perspective of the dic-
aid vary in strength with the emotional closeness to us of
tator. It is possible to observe only the dictator’s conduct
our fellows. When we know whom we are aiding, that
and speculate on her motives. What we can assert, how-
very fact brings our bene?ciaries closer to us than un-
ever, is that ?rst, the behavior of many dictators deviates
known strangers, and we feel more responsible for their
from money maximization; second, that social distance
well being” (Abelson, 2005, p. 35). In the experiments
does not explain why dictators donate and third, that the
analyzed here, there is no emotional closeness between
labels proposed here seem to ?t the range of behaviors
the dictator and the recipient. It has nothing to do with
revealed in all the DG experiments (moral behavior, gen-
knowing who the recipient is, but the recipient’s state of
erosity, reasonable self-interest, sel?sh behavior). These
well being. Given that the recipient is badly off, there
labels respond to moral distance and donations; two ob-
is an obligation to help; an obligation which some ful?ll
jective categories that are independent of motivations.
— the majority in this case — and others do not. It is
It is interesting to note that all of these behaviors ?gure
this obligation, and not emotional ties, what establishes
in the vast literature on DGs, albeit they appear separately
the moral distance between dictator and recipient. Moral
in a variety of experiments. We have seen, for exam-
distance depends on the relevant information that the dic-
ple, how dictators act with reasonable self-interest when
tator has regarding the recipient’s situation and regarding
lacking information given that in a situation of this type
her own situation.
they are morally distant from the recipient. Self-interest
The freedom that the dictator has to give or not give
works here as a social norm (Miller, 1999) that permits
and the greater or lesser moral distance between the dicta-
the dictators to take care of themselves without being la-
tor and the recipient permits the behavior of the dictators
belled as sel?sh, that is, behaving in a morally egotistical
to be labelled in the following way.
manner. Even so, generosity is possible as some dicta-
tors donate part of the money to an unknown recipient
a) If the moral distance between the dictator and the re-
in a double-blind DG. Likewise, when dictators have to

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
352
divide legitimately earned money with the recipient they
not allow us to establish a causal link between such be-
hardly donate anything — 95% of the dictators keep the
havior and the post hoc explanation for this behavior. But
full amount of money (Cherry et al., 2002, p. 1218). In
perhaps this is not the proper way to approach this issue.
such a case, there is no actual obligation to donate since
As Scanlon argues, judgements on what is correct or in-
the dictator obtained the money through her own personal
correct are not normal empirical judgements, but “claims
effort. Here the moral distance with the partner is maxi-
about what we have reasons to do” (Scanlon, 1998, p. 2).
mum, therefore making it legitimate to defend one’s own
The fact that the experimental subjects who donated all
interests.8 For whatever the reason, moral conduct arises
the money coincide in their reasons for action; a coinci-
in experiments in which either the dictator has to divide
dence that occurs without previous agreement or group
the money that she has produced with her partner (Ruf?e,
negotiation and pressure, seems to support the reasons
1998) or if she has to give money to alleviate poverty and
the dictators gave for donating in the speci?c context pro-
buy medicines (Brañas-Garza 2006). On these occasions
posed to them. Although these are undoubtedly post hoc
moral distance is null and there is an actual obligation to
reasons and we cannot af?rm that they cause the action
donate; an obligation that the dictators usually ful?ll.9 In
without committing a fallacy, there can be no doubt that
these cases we also ?nd self-interested behavior. Given
it is quite a different ball game when dictators say they
that the moral distance is null, there are those who keep
give money because their partner needs it than when they
all the money. When we modify the circumstances sur-
make their decision by tossing a coin in the air. The part-
rounding the decision, we will have both different moral
ner’s poverty seems to be suf?cient grounds for the dicta-
distances and different outcomes.
tor to believe that she should donate in a speci?c situation
Unlike what occurs in other experiments, here we have
and for her to indeed donate.
attempted to approach the internal perspective of the dic-
There may be other circumstances in which the dicta-
tator, that is, the reasons that drive the dictator’s behav-
tors give and yet their behavior could be labelled as ego-
ior, to see if behavior and motivations merge, in other
tistical, since as we have seen the decision comes cheap:
words, if dictators donate money for moral reasons when
the money they donate has come at no cost to themselves
moral distance is null. While the quantitative analysis
and as many state, “won’t do them much good.” In such
by Brañas-Garza (2006) found a spectacular increase in
circumstances morally correct behavior is cheap. Perhaps
donations when reducing the moral distance between the
this explains the high number of non-neutral consequen-
dictator and the recipient, the analysis of the qualitative
tialist responses that we detected in both experiments.
data presented here shows that the dictators — without
Whatever the case may be, it is interesting to con?rm the
previous agreement, without communicating with one an-
coherency of the labels we have chosen to use since those
other and under conditions of absolute anonymity- ap-
who appeal to reasons of a deontological nature do not
peal to moral motivations such as solidarity or helping
consider if their decision is cheap or not, but rather the
the needy to justify their decision. To put it another way,
duty of giving. This is yet another indication that these
ful?lling one’s obligation to give does not seem to be a
are true reasons for action; a fact that would in turn sup-
random choice nor the result of non-moral reasons, but
port the hypothesis that it is moral distance — the sense
of moral reasons. Dictators seem to “measure” moral
of obligation of the kindness hypothesis — and not so-
distance well as they refuse to give money when they
cial distance which marks the difference in experimental
have no information about the recipient, but do give it
DGs. What seems to be unquestionable is that the moral
when they know the recipient is poor or is poor and needs
framing determines dictators’ decisions. However, it is
medicine. In our experiments the response (to give or not
necessary to investigate further to understand better the
to give) to the greater or lesser moral distance is generally
in?uence of moral distance in the dictator’s behavior, and
tied to moral motivations.
the relationship of this concept with other concepts and
Because the justi?cation is given following the DG, we
hypothesis used to explain donations in DGs.
can argue that this is a mere post-hoc verbal justi?cation
that does not imply that the behavior we call moral is ac-
tually the result of a moral judgement (Haidt, 2001). As
References
we stated above that, the experiment analyzed here does
Abelson, R. (2005). Moral distance: What do we owe
8Quite another thing is if the dictator obtains the money by her own
to unknown strangers? The Philosophical Forum, 36,
means and is then told that she has to divide it with a poor recipient or
a poor and ill recipient.
31–39.
9This does not mean that those who give nothing are sel?sh: their
Bolton, G., Katok, E., & Zwick, R. (1998). Dictator game
behavior may respond to the fact that they do not believe in the experi-
giving: rules of fairness versus acts of kindness. Inter-
ment or do not believe that the money will reach its destination. In such
a case their conduct would be reasonably self-interested. But in order
national Journal of Game Theory, 27, 269–299.
to determine this we must know the reasons for action.
Bohnet, I., & Frey, B. (1999). Social distance and other-

---

Judgment and Decision Making, Vol. 3, No. 4, April 2008
Moral distances in dictator games
353
regarding behavior in dictator games: Comment. The
Appendix: Reasons given by subjects
American Economic Review, 89, 1, 335–339.
in the real experiment
Brañas-Garza, P. (2006).
Poverty in dictator games.
Awakening solidarity. Journal of Economic Behavior
and Organization, 60, 306–320.
In the following, “Cons” means consequentialist; “De-
ont” means deontological; “Rand” means random de-
Burnham, T. C. (2003). Engineering altruism: a theoret-
cision; “Trust” means “Did not trust the experiment”;
ical and experimental investigation of anonymity and
“Trust2” means “Didn’t trust the system”; “Legit” means
gift giving. Journal of Economic Behavior & Organi-
legitimacy; “Ego” means Egoism/Hardship; “Good”
zation, 50, 133–144.
means good purpose.
Crawford, B. (1969).
On the concept of obligation.
Ethics, 79, 316–319.
Charness, G., & Gneezy, U. (forthcoming). What’s in
Sub- Extract from the answers
Cate- Dona-
a name? Reducing the social distance in dictator and
ject
gory
tion
ultimatum games. Journal of Economic Behavior and
1
The more people with medicines
Cons
3
Organization.
the better
Cherry, T. L., Frykblom, P., & Shogren, J. F. (2002).
2
I don’t believe in the aim of the ex-
Trust
0
Hardnose the dictator. American Economic Review,
periment
92, 1218–1221.
3
Solidarity
Deont
3
Eckel, C., & Grossman, P. (1996). Altruism in anony-
4
We all should give some money to
Deont
3
mous dictator games. Games and Economic Behavior,
put an end to injustice
16, 181–191.
5
This money is more useful in these
Cons
3
Frohlich, N., Oppenheimer, J., & Moore, J. B. (2001).
countries
Some doubts about measuring self-interest using dic-
6
Their scarcity
Cons
3
tator experiments: the costs of anonimity. Journal of
7
I can help them
Cons
3
Economic Behavior and Organization, 46, 271–290.
8
Solidarity
Deont
3
Frohlich, N., Oppenheimer, J., & Kuriki, A. (2004).
9
I can help them
Cons
3
Modeling other-regarding preferences and an experi-
10
They need it
Cons
3
mental test. Public Choice, 119, 91–117.
11
I can help them
Cons
3
Haidt, J. (2001). The emotional dog and its rational tail:
12
I believe being fair is the best de-
Deont
3
A social intuitionist approach to moral judgment. Psy-
cision
chological Review, 108, 814–834.
13
I prefer [. . . ] donating the money
Trust2
1
personally
Hoffman, E, McCabe, K., Shachat, K., & Smith, V.
(1994). Preferences, property rights and anonimity in
14
They need it
Cons
2
bargaining games. Games and Economic Bahavior, 7,
15
I can help them
Cons
3
346–380
16
I can help them
Cons
3
Hoffman, E, McCabe, K., & Smith, V. (1996). Social dis-
17
I can help them
Cons
3
tance and other-regarding behavior in dictator games.
18
I have enough
Cons
3
American Economic Review, 86, 653–660.
19
They need it
Cons
3
20
They need it
Cons
3
Miller, D. (1999). The norm of self-interest. American
Psychologist, 54, 1053–1060.
21
Solidarity, ethic principles
Deont
3
22
I don’t trust in some organizations
Trust2
2
Mulgan, T. (2001). The demands of consequentialism.
23
2
Oxford: Oxford University Press.
24
I’m keeping what’s mine
Legit
2
Ruf?e, B. (1998). More is better but fair is fair: Tip-
25
They need it
Cons
3
ping in dictator and ultimatum games. Games and Eco-
26
Problems to distribute the money
Trust2
1
nomic Behavior, 23, 247–265.
among recipients
Scanlon, T. M. (1998). What we owe to each other. Cam-
27
My decision has been random
Rand
2
bridge, MA.: Harvard University Press.
28
This money is more useful in these
Cons
3
Zimmerman, M. (1996). The Concept of Moral Obliga-
countries
tion. Cambridge, UK: Cambridge University Press
29
Sharing the money that’s not mine
Legit
2

---