Altruism with Social Roots: An Emerging Literature

Altruism with Social Roots: An Emerging Literature
Pablo Branas-Garza y Maria Paz Espinosa

The interaction between social proximity and integration (column 7)
captures the number of friends when the subject is playing with a friend.
Note that this is an alternative measure of reciprocity (and thus highly
correlated to p_i(j)) which is highly significant in explaining giving.

Table 4.         Giving Regressions.

(*) p-values in parentheses.

These results can be interpreted as follows:

Even though a friendship effect is observed in the experimental data,
this effect is mixed with two other variables: reciprocity (the possibil-
ity of ex-post favor trading) and social integration (the number of
outstanding cooperative links).

When f_i is included in the regression to capture social integration it is
weakly significant. This is because on the one hand, when a subject is
matched to a friend reciprocity is a decreasing function of the number
of links f_i, so that more isolated agents should give more. Thus, giving
induced by strategic reasons (by the possibility of tracing the recipi-
ent and obtaining ex-post favors) is a decreasing function of f_i. On the
other hand, subjects with higher social integration are more likely to
give more since they have outstanding long run cooperation relations.
This second effect goes exactly in the opposite direction: subjects
with higher social integration have more friends and give more.

256

<<   <   9   10   11   12   13   14   15   >   >>