Ten Reasons to Submit Preprints to arXiv

arXiv offers functionalities that SSRN, institutional working-paper series, or a personal webpage do not have.

I have been involved with arXiv for a few years now—first as moderator for General Economics in 2019–2021 and then as part of the Economics Advisory Committee. I am surprised that arXiv has not become more popular in economics, despite the many benefits that it offers to individual researchers, and to the field in general.

What is arXiv?

arXiv is a non-profit operated out of Cornell University that provides a free distribution service and an open archive for scholarly papers. It now hosts more than 2.2 million papers in the fields of physics, mathematics, computer science, biology, finance, statistics, electrical engineering and systems science, and, since 2017, in the field of economics.

The economics section of arXiv was launched by a group of econometricians: Victor Chernozhukov, Iván Fernández-Val, Marc Henry, Francesca Molinari, Jörg Stoye, and Martin Weidner. It initially only accepted econometrics papers, but all papers in economics are now welcome. Econometrics papers should be submitted to the Econometrics category (econ.EM). Theory papers should be submitted to the Theoretical Economics category (econ.TH). And all other papers should be submitted to the General Economics category (econ.GN). More categories will open when the volume of submissions warrants it. The current moderators for the three categories are Sami Stouli, Anna Mikusheva, Philipp Strack, Nageeb Ali, Shoshana Vasserman, and Peter Hull.

Why submit preprints to arXiv?

So, why should anyone submit a preprint (that is, a non-refereed, unpublished, working paper) to arXiv instead of posting it on SSRN, sending it to other working-paper series, or simply hosting it on a personal webpage? There are many reasons, which I am going to review below.

I compare arXiv to SSRN because is quite popular in economics. I also compare arXiv to the CEPR and NBER working-paper series because those are the most prestigious working-paper series in economics. There are many other working-paper series, but they suffer from the same limitations as the CEPR and NBER series. Given that arXiv exists, I am not sure why all these working-paper series continue to exist.

1. Openness

First, arXiv is open to anybody. When you sign up for arXiv, the best is to use an institutional email address (for instance, a .edu email address) to expedite the endorsement process.

¹

This is different from the CEPR and NBER working-paper series, which are restricted to members of their respective club.

2. Minimal moderation

The second reason to submit to arXiv is that moderation is minimal. A moderator checks that your submission is indeed a scientific paper, and that’s all. This takes a few hours.

By contrast, SSRN takes days if not weeks to vet papers. CEPR takes a varying number of days to vet papers. NBER bans policy recommendations from their papers.

3. Time stamps

A third reason to submit to arXiv is that it offers a precise time stamp. Each paper comes with the day and time (up to the second) when the paper was submitted. So any issue of scientific priority can easily be resolved.

NBER only gives the month when the paper was circulated. SSRN and CEPR give the day when the paper was posted, but that could be weeks after the paper was submitted. Of course, a personal webpage does not offer a time stamp.

4. Open access

A fourth reason to submit to arXiv is that arXiv papers are completely open access. And papers that are freely available online receive many more citations than papers that are not.

Unlike papers on arXiv, papers by NBER and CEPR are behind a paywall. Papers on SSRN are freely available but can only be downloaded after registering or jumping through hoops.

5. Daily distribution

A fifth reason to submit to arXiv is that it offers a daily distribution service. A paper submitted during the day will be published at night and circulated to the category’s email list at the same time. (There is no distribution on weekends.)

In contrast, CEPR and NBER circulate papers only once a week. Using SSRN or a personal webpage does not give access to an email list.

²

6. Version control

A sixth reason to submit to arXiv is that it is easy to update your working paper as many times as required. Furthermore, all the versions of the paper are permanently available and easily citable. Consider for instance this paper. We circulated four versions of it. The paper can be cited as arXiv:1911.05271 to refer to the paper in general. But it is also possible to cite a specific version if it contains specific results of interest. For instance, version 2 can be cited as arXiv:1911.05271v2. The ID and link will always lead to that specific version. The PDF will always be available there. So even if some results are cut from the published version of the paper, it’s possible to find them and refer to them through arXiv’s version control.

Another advantage of arXiv’s versioning is that the research process is completely transparent. arXiv keeps track of who wrote what when. With such record keeping, it will be easier to attribute scientific credit properly, and it will be harder to poach ideas.

Such sophisticated versioning is unique to arXiv. On SSRN for instance, a new version entirely replaces the old version. It is not possible to cite the old version, or even find it at all. So the paper history is completely opaque.

7. Digital object identifier (DOI)

Another reason to use arXiv is that it provides a DOI with each paper. For instance, for the paper mentioned above, the DOI is https://doi.org/10.48550/arXiv.1911.05271. DOIs are extremely helpful because unlike URLs they never rot. The DOI will always link to the paper’s page. So they are much more robust than traditional URLs. More and more journals require DOIs in bibliographies, so it is an important to have one.

SSRN, the CEPR, or a personal webpage do not provide a DOI for the paper. NBER does, however.

8. Link to code

Yet another great functionality of arXiv is that you can link the paper to its code. This is achieved through a partnership with Paper with Code. You can link the paper at the time of submission to any repository on GitHub or GitLab. For the paper mentioned above, here is the page linking paper to code. You can then see links to code repositories at the bottom of the arXiv page:

Neither SSRN nor the CEPR and NBER working-paper series offer this functionality.

9. Journal reference

A ninth reason to submit to arXiv is that arXiv allows you to add a journal reference and a DOI upon publication. This way, readers will know that your preprint has been published, and they will know where to find it.

CEPR and SSRN do not offer this functionality. NBER does if you are willing to email the webmaster.

10. Authorized prepublication

There is a last reason to use arXiv for people who eventually submit their work to Science, Nature, and PNAS. These journals allow preprints to be posted on arXiv before submission because arXiv is a nonprofit preprint server.

But they do not allow preprints to be posted on SSRN because SSRN is owned by Elsevier, so it is a for-profit server. The NBER and CEPR do not run preprint servers, so they are ruled out too.

How to submit?

How to submit a paper in practice? It is fairly easy—with the caveat that arXiv requires source files for papers written in LaTeX.

arXiv provides detailed instructions on how to submit a paper. IEEE also has a guide for first-time arXiv users. And Federico Echenique has compiled arXiv guidelines. Once you have done it once or twice, it should not take more than 5 minutes to submit your paper.

If you have any question about your arXiv submission, feel free to leave a comment.

1. If you are a member of the CEPR or NBER clubs, it might make sense to circulate your working paper through their networks until arXiv becomes more widely adopted. But it also makes sense to submit your paper to arXiv simultaneously, for the reasons described here.

2. It’s very easy to subscribe to your favorite arXiv email list. For economics, just email econ@arxiv.org, write subscribe Your Name in the subject line, and write add EM (for econometrics) or add TH (for theory) or add GN (for everything else) in the body of the email.