Guest post by Tilman Borgers (JET Lead Editor), Marciano Siniscalchi (JET Editor), and Jason Hartline (JET Associate Editor):
The Journal of Economic Theory (JET) would like to encourage submissions from computer scientists. JET is a leading journal of the economic theory community, and has a broader readership among economists and covers a broader range of topics than other theory journals. JET is also the first field journal of the economic theory community, having been founded more than 50 years ago. Many publications in JET in those 50 years have changed not just the direction of economic theory, but also the direction of economics overall.
The convergence of research interests of computer scientists and economic theorists has been a remarkable development, and JET would like to do more to help facilitate the exchange of ideas across fields. Therefore, we compile here some suggestions for computer scientists who are interested in submitting their work to JET.
The basic standard for a publication in JET is that the paper should be original, make a substantial contribution, and be of interest to a broad group of readers, and this group should include economic theorists. Of course, editors make subjective, and fallible, judgments when assessing whether a paper meets these criteria. Typically, the substantive contribution is a contribution to economic theory, i.e. to our understanding of models of markets, strategic games, mechanisms, etc. Papers may be computer-science centric in its contributions, but then these contributions should be on a topic of interest to economists. For example, new algorithmic results related to game theory or mechanism design may be of interest to JET, if it is the editors’ judgment that these algorithms will be of interest to economists. On the other hand, results on more applied computational problems, such as faster algorithms for winner determination in auctions, or for clearing prediction markets, may be out of scope for JET.
In terms of style, successful JET submissions include an introduction that is accessible to a broad theory audience, and that explains the motivation for the work, overviews the main results, and explains some key intuitions. The introduction, or a separate literature review section, should precisely situate the work relative to the most closely related research. The main body of the paper should explain the model rigorously, and state the results precisely. Proofs which are not very long, and which provide insight, are typically included in the main body of the paper, whereas other proofs are moved to an appendix. It is often useful to paraphrase results in words after stating them formally, and to give explanations of intuitions as well as explanations of proof structures. We encourage authors to make their work as simple as is possible without losing the main message.
There is no length limit per se, but published papers have rarely more than 40 pages, including appendix and references, in print. We value conciseness, and focus on a main theme throughout the paper. Minor results can be left out. On the other hand, we do provide authors with the space needed to be precise and clear.
One general recommendation for computer science authors in preparing manuscripts for economics journals is to have an economist colleague look over the paper before submitting. This is a good way to identify inaccurate assumptions about readers’ knowledge, or omitted relationships to the prior literature in economics. Such advice can also help better motivate the results of the paper from an economic perspective.
To be publishable, if an earlier version of a paper was published as an extended abstract in conference proceedings, then the journal version must make additional contributions beyond the conference version. This additional contribution may include important conceptual aspects of economic interest that were omitted from the original extended abstract, proofs that were omitted from the extended abstract, and additional results that did not appear in the extended abstract. Authors should explain the differences between the conference version and the journal version in a cover letter. Papers that have previously appeared as one or two page abstracts in a conference volume do not need to distinguish themselves. Mentioning these appearances in a cover letter would useful, however.
Turing's Invisible Hand 