Reviewers are welcome to recommend a particular course of action, but they should bear in mind that the other reviewers of a particular paper may have different technical expertise and/or views, and the editors may have to make a decision based on conflicting advice. The most useful reports, therefore, provide the editors with the information on which a decision should be based. Setting out the arguments for and against publication is often more helpful to the editors than a direct recommendation one way or the other.
Editorial decisions are not a matter of counting votes or numerical rank assessments, and we do not always follow the majority recommendation. We try to evaluate the strength of the arguments raised by each reviewer and by the authors, and we may also consider other information not available to either party. Our primary responsibilities are to our readers and to the scientific community at large, and in deciding how best to serve them, we must weigh the claims of each paper against the many others also under consideration.