Structural holes and bad ideas: Liverpool’s Atlantic trade networks in the early-eighteenth century

Abstract

Historians have been complicating their understandings of networks, and especially mercantile networks. Moving on from a rather positivistic approach based on family, religion and ethnicity, research has now started to stress the inherent problems inherent in networks. 2 Furthermore, researchers have started to stress the problems in constructing and maintaining networks of whatever type. 3 However, far less work has been conducted on the problems encountered in developing and under-developed networks, where key actors may be unduly influential because they have access to new and/or highly-prized information. Such key actors (often but not always, Mark Granovetter's 'weak ties') can help to bridge what Ronald Burt has called structural holes in networks; that is, 'the separation between non-redundant contacts'. 4 Actors that stand near structural holes are 'at a higher risk of having good ideas' through the ability to better synthesise the various, different, and new information to which they have access. 5 Moreover, that information is likely to be seen as high value and credited because that information is considered rare. Therefore, actors bridging structural holes are often perceived as positive players within a network. 6 However, the very fact that these bridging actors have information arbitrage to their advantage means that they are also in a great position to act against the interests of the principal(s), especially where there is a lack of good governance. 7 These opportunities could be seen as 'bad ideas', the opposite of Burt's good ones. This situation could lead to bad decision making simply due to a lack of competency, through positive adverse selection, or even fraud. 8 Either of these behaviours could be seen as a type of network failure, which Andrew Schrank and Josh Whitford define as 'the failure of a more or less idealized set of relational-network institutions to sustain 'desirable' activities or impede 'undesirable activities'. 9 They posit two types of network failure: absolute and relative -the latter of which is more relevant to the discussion here. 10 They argue that relative failure occurs when networks fail due to a lack of competencies (involuted), or due to opportunism (contested). The first is due to ignorance and a failure to absorb enough information, and the latter occurs where there is sufficient competence, but safeguards against mistrust and opportunism are absent. The colloquial, if clear way they sum