DESARROLLO Y SO
CIEDAD
SEGUNDO SEMESTRE DE 2006, PP. 245-260.
dictators about recipients’ poverty levels. The research dealing with
social framing includes also Frohlich et al. (2001) who analyzes how
the presence of recipients (in front of dictators) increases the credibil-
ity of the experiment and social proximity and, thus, giving. As shown
in Hoffman et al. (1996) proximity between the participants and the
experimenter and between subjects is a key variable in explaining
social behavior. In fact when the two variables are reduced to the
minimun, giving approximates the game theoretical prediction, i.e.
zero (see Camerer, 2003; Frohlich, Oppenheimer and Kurki, 2004 and
Meier, 2006 for more detailed expositions of dictator game results).
However, the social side of altruism is not yet well understood. Al-
though the number of experimental papers studying altruism is over-
whelming, there are very few articles connecting social networks2 and
generous behavior. Note that the social side of altruism, if it matters
at all, affects a key question: if altruism is socially based then any
policy increasing interactions between individuals could be used as a
device for promoting cooperation3. In this paper we address this issue
in the light of existing experimental evidence.
The rest of the paper is organized as follows. In Section I we examine
previous work on the relationship between social networks and giv-
ing and present empirical regularities across experiments. It turns out
that the evidence is somewhat contradictory. Therefore, in Section II
we analyze the features which may explain the differences in the re-
sults and test our conjectures with a regression analysis. In Section III
we discuss our main conclusion: reciprocity and social integration are
the main determinants of giving.
I. Social networks: an emerging literature
The starting point of this literature is a well established experimental
result: the larger the social distance the smaller the level of donations
2 A recent survey on network literature is Jackson (2006).
3 These ideas could be applied in the area of human resources management. If altruism is
desirable within a firm, and provided experimental literature can show that altruism is socially
based, then organizing social acivities for the weekend would be a good business idea.
247