The Role of Reputation Building in Coordinating
Collective Resistance (w/ Doug Smith)
Our research—building on experimental results of Cason and
Mui (2007)—examines why subordinates in a
leader–subordinate relationship are willing to pay an
economic cost to help pre- vent exploitation by a leader of
another subordinate. Using a stylized game capturing
essential features of the leader–subordinate relationship,
Cason and Mui find that when the leader attempts to
“divide-and-conquer” the subordinates, the subordinate not
targeted for exploitation will still choose to challenge
the leader’s exploitation a significant fraction of the
time, even though challeng- ing is costly and there is no
direct material benefit. As a result, subordinates achieve
coordinated joint resistance more than equilibrium analysis
would predict. While Cason and Mui suggest that coordinated
resistance is due to a fairness norm — the non-targeted
subordinate shows solidarity with the targeted
subordinate—another motive may be at play: an incentive to
gain a reputa- tion for challenging. Each subordinate has
an equal chance of being the victim of future leader
exploitation. As a result, there is an incentive to create
an expectation among leaders that “divide- and-conquer”
style exploitation will be met with coordinated resistance:
each subordinate secures a higher payoff on average if the
leader chooses to not transgress. Furthermore, a
subordinate has an incentive to challenge now if they think
it will increase the likelihood that joint resistance will
be successful in the future. Our experimental design
removes the strategic incentive to reputation- build. It is
designed to test whether the willingness of subordinates to
help resist exploitation of fellow subordinates is
motivated by a solidarity preference specific to
hierarchical relationships or as an attempt to influence
the behavior of the leader — and other subordinates — in
the future.
Status: project funded by NSF and Rackham Graduate
School - writing draft
Solidarity in Hierarchical Relationships (w/ Doug
Smith)
This paper presents results from an experiment aimed at
avoiding the artificiality of the random anonymous
re-matching in the divide-and- conquer game. We modify the
treatment in a simple way: instead of re-matching after
each round, participants participate in 40 rounds in the
same 3-person groups. This change creates a situation more
analogous to actual interactions between leaders and
subordinates in political or economic situations. At the
same time, by maintaining the pre-commitment of all of the
leaders and half the subordinates, we will
be able to measure the degree to which behavior reflects
concern for other agents’ welfare as well as norms of
fairness, in a setting reflecting continuation within
relationships. This will allow us to develop a fuller
picture of the motivations behind subordinate solidarity in
hierarchical relationships.
Status: project funded by Interdiscpilinary Center for Organization
Studies (Michigan) - sessions complete, data
analysis in process
