|
Pure public goods have two distinct characteristics: they are non-rival and non-exclusive. One important question related to public goods is the following: how much money are people willing to pay for their provision? Two other questions are: (i) will people take part in collective actions involving public goods if they notice that the provision will take place regardless of individual contributions?; (ii) will individual actions lead to social inefficient results?
The main goal of this paper is to test hypotheses related to public goods provision in a dynamic setting. In doing so, we report the results of multi-round laboratory experiments performed in a business school, involving students majoring in accounting, business administration and economics.
When summarizing the available evidence in the 1980s, Andreoni (1988, p.291) reported three consistent results: (i) there was no significant evidence of free riding behavior in single-shot games; (ii) in experiments involving repeated games, subjects’ provisions for public goods tended to decay with each repetition; (iii) free riding was often approximated after several trials, although exact free riding was seldom realized.
The analysis performed in this paper is based on two stages. In the first stage, we conducted several classroom experiments related to the provision of public goods in a dynamic setting. In the second stage, we divided subjects in random groups of five members. Half of these groups had a fixed composition during distinct rounds (labeled ‘partners’) while the other half had their composition randomly changed in every round of the game (labeled ‘strangers’).
The results obtained demonstrate that, contrarily to previous studies, subjects who major in economics tend to cooperate more and free ride less than non-majors do. We also find a result where subjects who are randomly assigned to groups with a fixed composition (named ‘partners’) tend to cooperate less than those who are reassigned in every round of the experiment (‘strangers’).
|
