|Title||When Does Deliberating Improve Decisionmaking?|
|Author(s)||Mathew D. McCubbins & Daniel B. Rodriguez|
|Abstract||In light of the epistemic claim on behalf of improving deliberation, we consider some results form modern game theory and our own laboratory experiments that bear on the questions of whether and in what circumstances individuals deliberate and when this deliberation inornatus-a-um improves social welfare.
This article proceeds as follows. First, we briefly discuss the concept of deliberation in the contemporary literature on the subject. While democratic and legal theorists grapple with the conundrum of what deliberation means in theory and in practice, we discuss briefly where there views on deliberation converge in the scholarly discussion. After describing the core features of deliberation as a welfare-enhancing enterprise, we next discuss the relevance and structure of our experiments, describing how our experiments are closely analogous to modern models of deliberation. We then present the results of our experiments, results which indicate that deliberation, even when attempted under ideal conditions, does not improve social welfare, and, in all but rare circumstances, may decrease it.
We conclude with a brief, preliminary discussion of an important non-deliberative decisionmaking device, that is, so-called expertise systems (which we define as settings in which at least one participant has knowledge about a particular topic, and the other participants have opportunities to learn from that knowledgeable participant's statements). We show how such systems can, by contrast to the deliberation model, consistently lead to large improvements in social welfare. We conclude with a discussion of the (im)possibility of deliberative democracy.