Elsevier

Economics Letters

Volume 69, Issue 2, November 2000, Pages 137-142
Economics Letters

Non-reciprocal altruism in dictator games

https://doi.org/10.1016/S0165-1765(00)00283-4Get rights and content

Abstract

We carry out a double blind dictator game experiment where the anonymous recipients are randomly drawn from the Swedish general population, and any donations are mailed to the recipients. About a third of the subjects donate some money.

Introduction

In dictator games, a dictator divides a sum of money between himself and a recipient that must accept the allocation. The standard game theoretic model of payoff-maximization predicts that the dictator will keep all the money himself and leave nothing to the recipient. This prediction has been refuted in numerous experiments (e.g. Kahneman et al., 1986, Forsythe et al., 1994).

A possible explanation for the deviations from payoff-maximization in dictator games is that subjects have a taste for altruism or fairness. This explanation has, however, been disputed by, among others, Hoffman et al. (1994) and Hoffman et al. (1996). They argued that a main motivation for positive donations in dictator games may be expectations of reciprocity, i.e. by donating some money the subject expects to gain in future interactions with the experimenter, the recipient, or others who know the dictator’s decision. They hypothesized that by decreasing the social distance between the dictator and the experimenter donations would decrease. This was tested using a double blind experimental design where the decision of each dictator would be anonymous both towards other subjects in the experiment and towards the experimenter. The double blind procedure significantly decreased the proportion of subjects that donated anything to 36% and the average donation to 9% of the amount allocated. Subsequent experiments with the same double blind design have yielded similar results with average donations ranging between 8% and 16% of the amount allocated (Eckel and Grossman, 1996, Eckel and Grossman, 1998).

Although the double blind experimental design decreased donations, about a third of the subjects still donated some money to the recipients. The reciprocity between the dictator and the recipient may, however, not have been completely removed with the double blind procedure. The student subjects acting as dictators can infer that the anonymous person is someone like themselves participating in the experiment. Therefore, they may not believe that anonymity can be completely guaranteed. This was also suggested as a possible explanation for the remaining level of donations in the experiment by Hoffman et al. (1996).

In this experiment we increase the social distance between the dictators and the recipients by randomly drawing recipients from the adult general population in Sweden. If money is donated, it is mailed to the recipients, who are unaware that they are participating in the experiment. This design guarantees anonymity between dictators and recipients and removes any possible remaining reciprocity in the double blind design used by Hoffman et al. (1994). If donations in dictator games are motivated solely by reciprocity, donations should therefore drop to zero with this experimental treatment. We also replicate the standard double blind procedure used in recent dictator game experiments (Hoffman et al., 1994, Hoffman et al., 1996, Eckel and Grossman, 1996, Eckel and Grossman, 1998). The null hypothesis we test is that the distribution of donations does not differ between the two experimental treatments.

Section snippets

Double blind standard procedure

In this experiment thirty subjects are recruited to room A (the dictator room) and twenty-nine subjects are recruited to room B (the recipient room). The subjects are paid a SEK 50 show up fee, and asked to sit at assigned seats positioned to keep subjects as separate as possible (SEK=Swedish crowns; Exchange rate June 1999 $1=SEK 8.50). Before the experiment starts the subjects are reminded that there should be no talking during the experiment. The subjects are given the experimental

Results

The results are summarised in Table 1, showing the percent of decisions for each amount donated. In the double blind standard procedure 33.34% of the dictators donate some money and the mean donation is SEK 13.33 (13.33% of the amount allocated to the dictators). In the other experimental group with randomly drawn recipients from the general population 31.48% of the dictators donate some money and the mean donation is SEK 8.89 (8.89% of the amount allocated to the dictators). Table 2 contains

Concluding remarks

In our dictator experiment according to the standard double blind procedure about a third of the dictators made positive donations, and the mean donation was 13% of the total amount allocated. This result is similar to the results of previous studies in the US with the same design, where the mean donation has ranged between 8% and 16% of the amount allocated (Hoffman et al., 1994, Hoffman et al., 1996, Eckel and Grossman, 1996, Eckel and Grossman, 1998).

We also introduced a variation in the

Acknowledgements

We are grateful to an anonymous referee for helpful comments.

References (6)

There are more references available in the full text version of this article.

Cited by (45)

  • Building trust through knowledge sharing: Implications for incentive system design

    2021, Accounting, Organizations and Society
    Citation Excerpt :

    Even in settings where individuals cannot receive rewards for their helping behavior, ample prior research demonstrates that intrinsic (nonmonetary) factors motivate individuals to help others. For example, results from dictator games (Eckel & Grossman, 1996; Johannesson & Persson, 2000) and more natural settings (Smith, Organ, & Near, 1983; Constant, Sproull, & Kiesler, 1996; Podsakoff, MacKenzie, Paine, & Bachrach, 2000) suggest that individuals are willing to incur a personal cost to help others even when they will not receive a reward for this help. Scholars attribute this intrinsically-motivated help to such factors as individuals’ utility for performing altruistic acts, fairness concerns, and pro-social behavior more generally (Camerer, 2003; Organ, Podsakoff, & MacKenzie, 2006).

  • Visions of a Martian future

    2020, Futures
    Citation Excerpt :

    Deep altruism by individuals, private donors, or organizations represents an extreme non-reciprocal form of altruism, where the actions taken by the founders are unlikely to confer any direct benefits to themselves or direct descendants. Although other animals exhibit varying degrees of reciprocal altruism (Trivers, 1971), non-reciprocal forms of altruism appear to be unique to humans and perhaps a few other primates (Brinkers & den Dulk, 1999; Johannesson & Persson, 2000; Takahashi, 2007). A benefactor acting out of a sense of deep altruism would forego any personal recognition in favor of a vision for the human species that extends far into the future.

  • Service gratuities and tipping: A motivational framework

    2015, Journal of Economic Psychology
    Citation Excerpt :

    The idea that tippers care about helping servers is also consistent with economic research on dictator games. This research has found that a substantial minority of people, when asked to divide a windfall sum of money between themselves and a stranger in any way they like, give some money to the stranger even when their decisions are completely anonymous (Hoffman, McCabe, & Smith, 1996; Johannesson & Persson, 2000). Furthermore, this research has found that allocator contributions increase under conditions likely to enhance empathy – i.e., when the recipient is identified ahead of the allocation decision (Small & Loewenstein, 2003), named or pictured (Burnham, 2003; Charness & Gneezy, 2008), more needing or deserving of help (Branas-Garza, 2006) and able to communicate with the allocator (Andreoni & Rao, 2011).

  • Norm, network, and commons: The invisible hand of community

    2014, Environmental Science and Policy
    Citation Excerpt :

    For this reason, a second type of game has been employed in the literature, called the dictator game, in which the second player has no choice in the matter. Clearly, in the dictator game, the first player has nothing to fear from rejections and, yet, experiment after experiment, players continue to offer substantial amounts in these situations, often approaching the level of offers in the ultimatum games (Eckel and Grossman, 1996; Hoffman et al., 1996; Johannesson and Persson, 2000). What do these simulations tell us?

View all citing articles on Scopus
View full text