Beyond their direct impact on evaluations, connections might also be relevant at the application stage. In particular, if prospective connected candidates are better informed about their chances of success they can adjust the timing of their application decisions and avoid costly errors.
We study the impact of connections on application decisions and evaluation outcomes using the exceptional evidence provided by the Italian national academic qualiﬁcation evaluations. Our database includes information on around 69,000 applications of researchers who pre-registered in 2012 for the ﬁrst round of the national qualiﬁcation evaluation. When the identity of committee members was announced, around 10,000 applications were withdrawn. The remaining 59,000 applications were evaluated by a ﬁve-member evaluation committee and 40% managed to qualify. We consider two possible links between pre-registered candidates and eligible evaluators: prior coauthorship of an academic article (coauthors) or common current aﬃliation (colleagues).
We ﬁnd that the withdrawal rate is signiﬁcantly higher among pre-registered candidates who, by luck of the draw, are assigned to a committee that includes a coauthor or a colleague. The probability that they withdraw their application after the composition of committees is announced is 3 percentage points (22%) higher than in the case of non connected pre-registered candidates. This pattern is driven by connected researchers with a weak research proﬁle. At the same time, while connected researchers are less likely to apply, we ﬁnd that their chances of success unconditional on applying are signiﬁcantly larger. Their success rate is 4.5 p.p. (13%) higher relative to other com-parable researchers who pre-registered for the evaluation. Moreover, information from 300,000 individual evaluations (ﬁve per applicant) shows that, within each committee, connected candidates tend to receive more favorable evaluations from their coauthors and colleagues, relative to the assessments they receive from other committee members. Approximately 37% of researchers who withdrew their application in the ﬁrst round reapply again in the following round of the national qualiﬁcation evaluations, which took place one year later and were carried out by the same committees. We ﬁnd that connected candidates are more likely to reapply and, among those who reapplied, they tend to be more successful. They also tend to receive more favorable evaluations from their connections in the committee relative to the assessments that they receive from other evaluators, suggesting that their previous withdrawal decision was not driven by the fear of a less favorable evaluation. Instead, the presence of a connection in the committee seems to have helped these researchers to optimize the timing of their application and, eventually, to be more successful.