[mannykannot]

The participants knew that this was the rule, of course (if they had not, there would have been no basis at all for saying "there was no incentive for people to overstate..."), which changes how a rational person would play the game: they would want to give some consideration to how other participants might bid.

Perhaps the study should have had several rounds of bidding, with the distribution of bids (or at least the minimum) from the previous round being announced beforehand. You would then get information both from how low the bids went, and who dropped out at each stage, while also getting metadata, about the methodology, in the form of how much the participants changed their bids from one round to the next.