Crowdsourcing Societal Tradeoffs

It would be desirable if, as a society, we could reduce the amount of landfill trash we create, the amount of carbon dioxide we emit, the amount of forest we clear, etc. Since we cannot (or are in any case not willing to) simultaneously pursue all these objectives to their maximum extent, we must prioritize among them. Currently, this is done mostly in an ad-hoc manner, with people, companies, local governments, and other entities deciding on an individual basis which of these objectives to pursue, and to what extent. A more systematic approach would be to set, at a global level, exact numerical tradeoffs: using one gallon of gasoline is as bad as creating x bags of landfill trash. Having such tradeoffs available would greatly facilitate decision making, and reduce inefficiencies resulting from inconsistent decisions across agents. But how could we arrive at a reasonable value for x? In this paper, we argue that many techniques developed in the multiagent systems community, particularly those under economic paradigms, can be brought to bear on this question. We lay out our vision and discuss its relation to computational social choice, mechanism design, prediction markets, and related topics.