When designing open multi-agent systems (i.e. those that allow anyone to add other agents to the system), one must consider how these agents will interact, and design protocols that discourage agents from spending time and computation trying to take advantage of others and encourage them to achieve actual domain tasks. Unfortunately, this seems to be possible only in very restrictive domains [8]. Specifically, it is not possible if we situate our agents in an economic society of agents, such as the University of Michigan Digital Library [2](UMDL). Here the agents will be responsible for making their own decisions about when to buy/sell and who to do business with. Market systems (e.g. auctions) will be implemented as part of the UMDL in order to facilitate the transactions. These mechanisms will, sometimes, diminish the benefits that might come from making ``strategic'' decisions. However, the individual agent has to decide when to let its own welfare rest in the hands of the market mechanism, and when it should take over and make more strategic decisions.
In this paper we present our approach to the problem of how an agent, within an economic MAS, can determine when it should behave strategically (i.e. model the other agents), and when it should act as a simple price-taker and buy/sell at the lowest price possible. We will show how, in some circumstances, agents benefit by building and using models of others while other times the extra effort is wasted. Our results point to metrics that can be used to make quantitative predictions as to the benefits obtained by using deeper models.