Voting For a Cartel as a Sign of Cooperativeness

This paper tests the hypothesis that a (partial) reason why cartels – costly non-binding price agreements – lead to higher prices in Bertrand Pricing Game-experiments could be because participants who form these kinds of agreements are more cooperative and pick higher numbers in general. To test this hypothesis we run an experiment where participants play two consecutive Bertrand oligopoly games: first a standard version without the opportunity to make price agreements; followed by a version where participants can vote, by majority, on whether to have a costly nonbinding agreement to pick the highest number. We find no statistically significant difference between the numbers picked in the first game by participants who vote for and against an agreement in the second game. We do confirm that having a price agreement leads to higher numbers being picked on average. Additionally we find that participants who vote for or against the price-agreement behave differently in response to the existence of the price agreement. In particular, participants who vote for a price agreement react more positively to the price agreement. The difference in numbers picked in the second game between situations with and without a price agreement is larger for participants who voted in favour of the agreement. Voters who voted for the price agreement are more cooperative than voters who voted against but only in situations where there is a price agreement.

本文旨在验证以下假设：导致卡特尔——代价高昂的非约束性价格协议——在伯特兰定价游戏中引发价格上升的部分原因可能在于，那些形成此类协议的参与者更具合作性，且普遍选择更高的数字。为验证此假设，我们进行了一项实验，参与者连续进行两次伯特兰寡头垄断游戏：首先是一次标准版本，其中参与者无机会达成价格协议；其次是一次版本，参与者可以通过多数投票决定是否达成一项代价高昂的非约束性协议以选择最高的数字。我们发现，在第二次游戏中，无论是对协议投赞成票还是反对票的参与者，在第一次游戏中选择的数字之间均无统计学上的显著差异。我们确实证实，存在价格协议的情况下，平均选择的数字更高。此外，我们发现，对或反对价格协议的投票者，在价格协议存在的情况下，其行为反应有所不同。特别是，那些投票支持价格协议的参与者对价格协议的反应更为积极。在有与无价格协议的情况之间，第二次游戏中选择的数字差异对于支持协议的投票者来说更大。支持价格协议的投票者比反对的投票者更具合作性，但这种差异仅在存在价格协议的情况下才会出现。