This question is mainly focused on the field of mathematics, although it might make sense for other areas too.
would setting up a Mathematical Institute of Refereeing be a good idea ?
To explain the idea, I have noted several things that seem to go in the same direction :
- referees today are not paid for their work (at least by good quality journals, I'm not talking about predatory ones), they do it by goodwill to be active members of their communities ;
- discussing a result with a colleague always speeds up one's understanding ;
- in many cases it is difficult to find a referee for a paper, and the process can easily take 9 months just to get a first report ;
- many people experience a draught at some point in their career.
So what about an institute where members have a temporary affiliation (say a 5-year contract, which they can quit at any time) and are tasked to only do refereeing, any involvement in other time-consuming activities (teaching, grant application, but also one's own research) would be banned (and checked regularly by the administrative staff and citizens).
The members would be sworn to secrecy (like a medical one, with cash and jail sentences if found guilty of breach, say during their membership and the 2 years beyond it), which would allow them to discuss papers they are reviewing with other members on-site.
Funding would be provided by a variety of sources (Universities, Private sector...).
And to make sure that there is no change from traditional anonymous refereing, these referees would not mention in their report that there are members, so it would look like it had been done by a standard referee at a university (but much faster).
The idea would be both to speed up the process, and to lower the pressure to publish mediocre papers (i.e. it would be valued more highly by the community to know that someone is a hard-working member of that institute, rather than someone is painstainkingly publishing small papers every couple year which consume refereeing ressources while adding little value to the area).
[edit: thank you for the answers and comments, many good points there of course. Clearly not all sub-fields of maths would be covered this way, but I thought it might help in some where this is high traffic. Plus, I thought people towards the end of their career, who have taught for 25/30 years and, in some cases, have also lost a bit the will of finding new stuff all the time, might be interested. (If you don't find it lacking courtesy, I think I'd like perhaps not to accept an answer to let it open-ended.)]