Power, Bargaining, and Collaboration

Over the past century, collaborative work has become increasingly common in academia [61]. In the sciences, it has continued to spread as a practice across disciplines and countries [50, 51]. In fact, in most scientific fields, the majority of published work is now co-authored [40]. The social sciences, likewise, have seen an increase in collaborative work, particularly in fields like economics and psychology [40, 32, 37]. Even in the humanities, collaboration, though not the norm, is increasingly practiced [37]. There are a number of reasons why this model of academic production is so popular. In some fields, increasing specialization, and the high cost of, for example, expensive instruments, mean that collaboration is necessary to produce meaningful research [40]. Many authors have argued, furthermore, that collaborating provides myriad benefits to individual researchers even when not strictly necessary. For example, collaboration can lead to greater academic productivity, which benefits individual researchers in hiring and promotion decisions [40, 36, 24, 38].1 Co-authored papers in some areas are more likely to be accepted to top journals and are more likely to be cited [35, 28, 4]. Furthermore, it has been argued that collaboration can be useful to academics in defraying risk. When numbers of publications are important, it may make more sense to invest smaller amounts of time in many papers than to invest heavily in one paper that might not be accepted to a good journal [32]. Lastly, collaboration is important to academics for social reasons. Collaborators create stronger social networks in their fields, and may be viewed by peers as insiders [39]. These clear cut advantages to collaborative work raise ethical questions, however [46, 25, 33, 23, 64]. The structure of academia is such that rewards for successful work, such as promotion, prizes, and salary increases, tend to accrue to individuals. Furthermore, author order, which is the main proxy used to determine credit for academic collaborations, is inherently unequal. Co-authors must occupy different places in the author ordering. This means that academics must decide who will be best positioned to receive these individual rewards. Further benefits to collaboration, such as opportunities to present at prestigious conferences, must likewise be divvied up. How are the fruits