18
NBS when applied to the unidimensional case. Confronting this theoretically, we noted that
other theoretical solutions do predict compromise even within the unidimensional case. We have
now presented the first phase of experimental data.
The data also do not support the NBS for the unidimensional case. The participants, by
the nature of the joint decision-making process, systematically choose outcomes further into the
core. Since the other theoretical outcomes are also not supported by the data, one must consider
other options.
Under private information, no outcome was overwhelmingly chosen by the participants.
This may well be because they did not perceive the underlying bargaining problem in the time in
which they had to review their (one-sided) payoff schedules. A future experiment still using
private information but giving ordered payoff schedules—rather than the randomized ones used
here—might possibly yield different results.
Under full information, the equity point stood out as the modal choice. This resembles
the observations in various ultimatum experiments in which fairness was a factor overriding
strict economic rational choice. While the Felsenthal-Diskin bargaining solution and the equity
point can be synonymous (when the status quo is far enough away from the core), the particular
payoffs chosen for this experiment meant that the two points were different. And yet the equity
point was still within the range of mutual benefit. Since only one of the sixty-eight observations
was outside this range, it seems that the status-quo-advantaged actor takes the status quo into
account to the extent that he will not accept agreements that make him worse off than the status
quo.
Thus, an empirically driven hypothesis for the unidimensional case may be as follows.
Under full information, if the equity point is mutually improving, the actors will be more likely
Convention