Firm-like behavior of journals?: Scaling properties of their output and impact growth dynamics

In the study of growth dynamics of artificial and natural systems, the scaling properties of fluctuations can exhibit information on the underlying processes responsible for the observed macroscopic behavior according to H.E. Stanley and colleagues (Lee, Amaral, Canning, Meyer, Stanley, 1998; Plerou, Amaral, Gopikrishnan, Meyer, Stanley, 1999; Stanley et al., 1996). With such an approach, they examined the growth dynamics of firms, of national economies, and of university research fundings and paper output. We investigated the scaling properties of journal output and impact according to the Journal Citation Reports (JCR; ISI, Philadelphia, PA) and find distributions of paper output and of citations near to lognormality. Growth rate distributions are near to Laplace "tents," however with a better fit to Subbotin distributions. The width of fluctuations decays with size according to a power law. The form of growth rate distributions seems not to depend on journal size, and conditional probability densities of the growth rates can thus be scaled onto one graph. To some extent even quantitatively, all our results are in agreement with the observations of Stanley and others. Further on, a Matthew effect of journal citations is confirmed. If journals "behave" like business firms, a better understanding of Bradford's Law as a result of competition among publishing houses, journals, and topics suggests itself.