Modern media technologies paved the way to the open access movement. Instead of the traditional academic subscription and publishing model, which allowed few big publishers to charge excessive publishing fees, the open access model raises the hope for a fair system, where scientific content is freely accessible and thus the dissemination of research work becomes possible at little cost. However, previous literature pointed out that big publishers seem to be able to preserve their market power when going from the subscription-based model to the open access model. In this paper, we take a closer look at the differences across disciplines. The publication routines in Social Sciences, Physical Sciences, Life Sciences and Health Sciences differ to a substantial extent. On these grounds, we test whether there are also differences in the explanations for the article processing charges (APC) across these disciplines. For doing so, we combined various data sources such as the dataset of the "Directory of Open Access", the "OpenAPC Initiative" and the "CiteScore Metrics". Our regression results show that the differences across the four fields in terms of publication habits and endowment levels allow publishers to exploit their market power to different extents.