Richard, the primary purpose of NEP is to announce new working papers. The categorization I propose jsut piggybacks on NEP. Thus, I do not see it appropriate to burden editors with the task of classifying older papers.

Roland, I see your point that someone with only a single paper could falsely be considered a specialist in a field. But keep in mind that NEP only has a snapshot of the literature and a recent at that. Only the more productive authors will have five or more papers in their field. For example, I consider myself a macroeconomist but would be excluded.

Ekkehart, this is a good point that any good economist should make: define the goal, then we can talk about the tools. Now that we approach 15’000 registered authors, my goal is to provide sublist by field of manageable size. A by-product is that those lists could be used for rankings within fields. There are various ways of doing the latter that I want to discuss later, and we should not be constrained by the construction of rankings.

Ekkehart

Consider a prolific author who has published 100 papers, which have received 300 entries in several reports. Now assume that this author has dispersed interests, so these 300 entries are distributed among, say, 10 reports. With an average of 30 papers, he would not be considered an expert in any of these fields. Moreover, even if he has contributed 70 entries in one of the 10 reports, this would not be sufficient to call him an expert in this field, according to the 25-percent criterion.

Compare this non-expert to another researcher who has contributed just one paper to NEP, which was cited in 4 reports. This renders him an expert in all four fields.

These two cases could easily be handled by the 5-paper criterion alone: the first researcher would then be considered an expert in many of his 10 fields, while the second guy had to wait a little bit longer.

Hence, while applying the 5-paper criterion alone would really makes sense, the 25-percent criterion provides no additional insights. On the contrary, it may rule out very productive writers and benefits the less prolific ones. I guess this was not the intention ;-)

All the best
Roland

5 and 25% are random numbers — as random as any other choice.

I did the obvious thing: Checked my own classification, and the things I know best: Environmental economics, and energy economics.

My specialisations are correctly recognised, although some are by co-authorship. One focus area is not there, but that is because there is no NEP report on water. This is an incentive to start one.

The lists of environmental economics and energy economics have many of the usual suspects.

There are some unexpected names, but in the cases I checked that is because these people did not update their profile for years.

Some of the big names are missing, but in most cases I checked that is because there people did not register. Publishing ranking by field would give them an additional incentive.

A few people are misclassified. In the cases I checked, that is because the NEP report cover only their recent papers. As far as I know, NEP reports do cover recently added historical material. However, it would be good to do a retrospective NEP report on all the material that was published before the first NEP report in an area.