In this paper we examine information exchange networks in legislative politics and challenge the idea that legislators seek objective information prior to voting on bills. We make the intuitive claim that legislators establish contacts with each other that stand to maximize the value of the information they trade. Additionally, we make the counterintuitive claim that legislators seek information from sources that are predictably biased for or against their preferred outcomes. We test the propositions derived from this argument in the context of the European Parliament, using tools from social network analysis and modeling the network dependence using a multilevel approach. This research makes two primary contributions to the field of legislative politics. First, we demonstrate both theoretically and empirically how legislators use social contacts to their strategic advantage in their pursuit of legislative information. Second, our analytical approach demonstrates how to appropriately account for interdependence of observations in network data.
3The process of lawmaking is an inherently social exercise and scholars have recently begun to use social network analysis to help explain some legislative behaviors (see for example, Fowler 2006). However, it is not yet clear how social networks among lawmakers contribute to legislative outcomes, policy formation, or pivotal activities such as voting. This paper seeks to begin to fill this gap by examining social networks in legislative politics as circuits of information exchange. Specifically, we are interested in examining if legislative offices establish connections amongst each other that maximize the value of the information they trade.We maintain that in an effort to make well-informed policy choices, legislators have incentives to pursue information from sources that are predictably biased; the social connections they establish to collect information about the legislation they enact reflect these incentives. Information provided by sources that are predictably biased allows legislators to compare the information they expect to receive, given the known bias of the source, to the information they actually receive. This is of great value to legislators as they seek to confirm the appropriateness of the policy positions they are predisposed to take toward a given policy proposal. If the information legislators expect matches the actual information they receive, their predispositions are confirmed; in contrast, if the source provides information that deviates from their expectations it is likely to trigger a re-evaluation of their initial policy positions.Prior authors have noted the value of "biased" information for legislators (see Kingdon 1981, 232;Calvert 1985); however, we offer that information has greater value to decision-makers if it is predictably biased. Such information is either in support of or in opposition to the position a legislator is predisposed to take, which means that legislators ought to seek information from both political allies and ...