Applying and Working at CNRS

First written: November 2024. Last updated: December 2024.

Summary: CNRS researcher positions (”chargé de recherche”) in France are among the best academic positions in the world, because they provide unrivalled freedom to pursue research. I’ve applied to join CNRS in January 2021 and started my position in February 2022. I’ve written this document to describe my experience applying for the position and getting started, for the benefit of others who are considering applying. This is mostly directed at non-French applicants (like myself), but may be of interest to French ones as well. It may also be of some interest to applicants to INRIA which shares some similarities with CNRS. It is only about the application for chargé de recherche positions, with some considerations being different for the senior roles of directeur de recherche.

What is CNRS?

CNRS is short for Centre National de la Recherche Scientifique (National Center for Scientific Research). It is a national institution and employs about 11 000 permanent researchers. In typical universities in France, academics are organized into research units (which might be equivalent to, e.g., a “computer science department” at universities elsewhere), which we usually call labs. Within each lab:

Some of the permanent staff are employed by the university. Their job is to both teach and to do research. Junior staff have the title maître de conférences and senior staff have the title professeur des universités.
Other permanent staff are employed by CNRS instead of the university, but they still belong to the lab and have their office there. Their job is only to do research. Junior staff have the title chargé de recherche (CR) and senior staff have the title directeur de recherche (DR).

(Note that only one of these four types of permanent research staff is called “professor”. This is in part because school teachers are known as “professors” in France, so I guess it is not necessarily a more prestigious job title than “maître de conférences”/“chargé de recherche”. But it does make it awkward to translate the other job titles into English.)

Officially, such labs are called unité mixte de recherche (UMR) to underline that its staff is a mixture of university and CNRS employees. (There are also other forms of labs such as UPRs.)

For example, I work at the LAMSADE lab at Université Paris Dauphine. LAMSADE is a UMR consisting of 38 university researchers and 13 CNRS researchers (plus PhD students, postdocs, support staff). My office is in the building of Université Paris Dauphine, and I have a university card from Dauphine, but I’m not employed by it.

What does a CNRS researcher do?

The job description is to do scientific research. That’s it. There is no obligation to do teaching or help with administering teaching or the university. The research you do is your choice; you don’t have a boss directing your research. Researchers are periodically evaluated, with a yearly form to fill out online (the “CRAC”, listing publications and other scientific activites such as supervision, grants, public outreach), and with 5-yearly evaluations by a scientific jury (which are still light touch). All other activities are voluntary.

Personally, I have found it easy to avoid taking on too many other responsibilities, and essentially all my time is spent on research activities, except that I give about 12 hours of lectures per year (paid extra) and I’m part of a lab committee that meets infrequently.

CNRS researchers are fonctionnaires, civil servants of the state, and enjoy very high job security. One becomes titulaire after one year, which could be compared to being tenured in the US. Because they work for a nation-wide entity, it is not difficult to move to any other lab in France if desired (and generally labs will be happy to accept CNRS arrivals since they don’t have to pay you). It is similarly possible to change disciplines. There are also possibilities to earn additional income through teaching or consulting, or to leave the position for a few years to do something else, with guaranteed re-entry. CNRS also has some international offices and labs that allow its researchers to work in other countries.

Application

Applications for CNRS positions are accepted once a year, with a deadline typically in early January. The available positions are usually announced in early December. They are very similar from year to year, but in some years there are some special focus areas or interdisciplinary calls available. Check the CNRS website on the external competition for researchers for details. That website also includes the official guide for candidates and information about deadlines.

The available positions are divided by subject area, each managed by a so-called Section of the Comité national de la recherche scientifique (CoNRS). Note that CoNRS is a body that is independent from CNRS. CoNRS sections are groups of academics of a particular discipline whose task is to evaluate applications. To know more about the application materials expected for your subject, find out what is your section, then search for the website of that section (a list of such websites is available on the C3N website). For me in computer science (section 6), this would be https://cn6.fr/. Note that some sections put informal limits on the seniority of applicants for chargé de recherche (CR) positions (e.g., at most 7 years of research experience including PhD).

Note (2024/2025): The sections are in the process of being renumbered. For example, the computer science sections 6 and 7 will become sections 2 and 3 from 2025/2026.

Success rates will vary widely between sections. In section 6, there are usually about 100-130 applications for 8 positions. About 30 applicants are interviewed. Thus, if your application is strong enough to get an interview, the success rate is above 25%. If you browse the rankings of admissible/admitted applicants from past years, you’ll see that many people apply multiple times, with their ranking often improving over time.

Step 1: Identify the labs you want to join

As part of your application, you need to identify at least two (and ideally three) labs that you are interested in joining. In your application, you will need to present an “integration plan” where you explain why your future research fits well within the lab and any possibilities for collaboration. The lab will usually write a letter to the jury endorsing your application. Thus, it is important to contact these labs early in the application process, to discuss your application. It makes sense to visit labs in person, have meetings with relevant researchers, and to maybe present a seminar talk.

Note that you’re applying for a fixed number of positions available nation-wide. Thus, unlike most job applications, your future colleagues are not really in the adversarial position of evaluating you – the evaluation is done by a nation-wide jury. Instead, your future colleagues’ interests are aligned with yours: if your application is successful, they will get a new valuable colleague that they don’t have to pay for. Thus, your future labs will actively want to help with your application, including by giving tips and feedback.

The usual recommendation is that at least one of the labs you are proposing should be located outside Paris/Île-de-France. I did not follow this recommendation and only indicated two Paris labs (since I had a strong preference for Paris and would likely have preferred to work in a different country rather than elsewhere in France). How important this is may also vary by discipline.

Step 2: Write your application materials

CNRS positions are unusual in that they are research-only. Thus, the requested application materials also focus more heavily on research than on, for example, teaching. There are two main documents you will need to write, one about the past (describing your research performed thus far) and one about the future (laying out a research program that you would follow as part of CNRS for the next several years).

For detailed instructions on what to write in these documents (and recommended page counts), look for instructions from the relevant section. I’ll just note some general thoughts. It is important to keep in mind that your application will be read by people who are not experts in your field. Thus, you should write in an accessible way. Minimize jargon, and consider adding figures and examples. I’d recommend setting out your past and future research in a broad way that would be interesting to a broad range of readers from your discipline. For example, mention both theoretical and more practical approaches. Your application will be read in detail by 2-3 members of the jury, so you should write in such a way that each of these jury members finds something to be excited about. Consider sending your application documents to people from outside your focus area to get feedback about comprehensibility.

To be eligible, you must also submit a sworn translation of your PhD certificate. I used an online service for this, which cost me about €40 and took 4 days.

Step 3: Interview

If you are shortlisted, you will be invited to an interview by the CoNRS section. My interview happened via zoom due to COVID-19, but most interviews happen in person in Paris. Exceptions are typically made for those who are based far away, but it’s worth making an effort to attend in person. During my interview, I was given the opportunity to briefly present my research and my integration plan, and we then went into an extended question period. In my case, there were three main jury members who had prepared questions, based on reading the papers I had highlighted in my application document about past research. These questions were interesting and friendly, though I also got some more skeptical questions about whether I have the competences to fully implement my proposed research program.

One general tip about preparing presentations for interview settings: I recorded my talk together with my slides as a video, uploaded it to YouTube (marked as unlisted), and sent the link to several people asking for feedback. In contrast to scheduling a mock live interview, this makes lesser demands on people’s time, and thus increases the amount of feedback you can gather. Consider sending your video to people outside your focus area.

The official CNRS candidate guide says (in bold font) that interviews “will be conducted in French. Under certain conditions, they may take place partially in English.” But my interview was fully held in English (indeed, in French I wouldn’t have been able to talk about much else besides the weather). I would imagine that most sections handle things this way for foreigners – this is in the interest of everyone since CNRS wants to recruit the best researchers, francophone or not.

Step 4: Admissibility jury and admission jury

A few days after the interviews, the jury you talked to during the interview (the “admissibility jury”, consisting of members of the CoNRS section) publishes a ranking of candidates. Candidates who were interviewed but are not on the ranked list are not admissible, so will not get the job. In my case, I was ranked first (together with several other candidates), which was very exciting and led to many people sending congratulations.

However, this ranking is not the final ranking. This is because there is a second jury (the “admission jury”, which is part of CNRS instead of CoNRS) that may make changes to the ranking, with the final rankings announced several weeks of painful uncertainty later. It is a bit mysterious to me what this jury does, and there are no explanations when changes are made to the ranking.

François-Xavier Coudert maintains webpages containing the rankings of the admissibility and admission juries, highlighting changes made by the latter: 2024 - 2023 - 2022 - 2021 - 2020 - 2019

The job

Accepting the offer

My interview happened on 8 April 2021, and the admissibility ranking was published on 15 April. The admissions jury met on 23 May, and I was informed by email on 26 May that my application succeeded. The email also indicated the lab I would be attached to (which was my first choice!). I then received a form where I could indicate my acceptance as well as the desired start date, with the latest option being February 2022. I chose February 2022 (March would have been even better but was not possible). I used the summer and the “extra” four months to do a short postdoc in Canada.

In early August, I received a series of forms to fill out, and a request for all of my work contracts so that my seniority level could be determined.

Moving and getting started

I then simply showed up for work on 1 February. I was slightly worried that I forgot to do something, because at that point I had no document that explicitly said that I did indeed have a job. Because of the civil servant status of the position, there is also no contract. But that’s apparently just how it works, and I received a document confirming I indeed have a job on 16 November 2022, a full 288 days after starting:

Moving to a new country is always an administrative adventure, because there are cyclic dependencies, such as banks wanting proof of residence and landlords wanting proof of bank account. For me, things that worked were Revolut for a bank account and Free for a SIM card (obtained by going to one of their shops and using the kiosk, that way you don’t need a pre-existing French number and you don’t need proof of address). I got my first place to stay from Chez Nestor which specializes on medium-term rentals for foreigners; they were always great to deal with but are a bit expensive. I can’t give advice about visas, since I don’t need one as an EU citizen. I also had to sign up in person for health insurance (which I found stressful for language reasons) and which involved a lot of waiting until I received my social security number and carte vitale.

Pay

As CNRS researcher, you will not be an employee but a civil servant (fonctionnaire de la fonction publique de l’État). This means that you will not sign a contract; instead you will be appointed. The conditions of your work are defined by laws and decrees, not a personal contract. In particular, there is no possibility of negotiating the conditions of employment, including the salary. (However, under certain circumstances one might be able to obtain different kinds of bonuses.)

Your salary is computed based on your seniority, which is the amount of time you have been employed to perform research. For many people, this clock will start when they begin their PhD and include the time of doing the PhD plus postdocs. For others, it might also include work done during research internships beforehand. You can see the payscale for chargé de recherche here: https://www.emploi-collectivites.fr/grille-indiciaire-etat-charge-recherche-classe-normale/0/6330/10579.htm

For example, I started with 4 years of PhD plus 2.5 years of postdoc. This places me in salary echelon 4, because the first three echelons take 1 year, 2 years, 2.25 years to complete, respectively, and 6.5 years > 5.25 years. Your seniority and therefore your salary increase automatically with time.

The salary is determined as a number of points (the “indice majoré” column), which is 605 points for echelon 4. These points are then multiplied by €4.92278 to give EUR 2 978 per month. The multiplier is raised occasionally, but it hasn’t kept up with inflation (there is an instructive graph about this on wikipedia). To this amount are added some bonuses, including a 3% bonus if you work in an expensive city (such as Paris). From this amount, a large chunk of social contributions (pension, health, family support) and taxes is subtracted.

There are some additional perks. Notably, 75% of your public transport expenditures will be reimbursed.

If you are curious, here is my most recent pay slip, after about 2.5 years in the job:

The right-most columns are employer contributions. Lines 1000 and 1060 are the salary and location bonus. Line 2210 is a bonus that everyone gets. Lines 5xxx and 6xxx are social contributions, including pension, health, family support. Line 8036 is income tax. Some of these lines will vary with your family structure, with married couples having tax advantages. I receive a bank transfer of €2616.50.

Career advancement

In France, to be allowed to solely supervise PhD students as well as to advance to the positions of directeur de recherche and professeur des universités, you need to become habilitated, that is, obtain the title of HDR (habilitation à diriger des recherches). Before then, you’ll co-supervise students with senior colleagues (which is anyway better, including for the student). To obtain HDR, you’ll need to gather a jury, write a habilitation dissertation, and do an oral defense.

Once you have HDR, you are eligible to apply to become a directeur de recherche (DR), which comes with higher salary. Passing from CR to DR is not automatic: the application procedure for DR positions is the same as for CR positions, so it involves preparing a full application package and going through the two juries. This process is somewhat competitive: in section 6 (computer science), in the 2024 application cycle, there were 32 applications for 8 positions (but of course people reapply). The CRs who were promoted to DRs entered their CR positions in 2010, 2011, 2012, 2015, and 2017 (I only found dates for 5 out of 8). The person who started 2017 had recently obtained an ERC Starting Grant. So it looks like there is typically about a 10-12 year delay between CR and DR for those who succeed in getting to DR.

Besides DR, another option is to advance into the status of CRHC (chargé de recherche hors classe). This is usually awarded to those who have gone through all the echelons available in the normal-class CR scale (which lasts for 21.5 years after beginning research, which in my case would be about 16 years after starting the CR position).

Some pros and cons

Finally, I will list a few upsides and downsides of working as a CNRS researcher and at French universities, compared to academic jobs elsewhere (such as tenure-track US faculty positions).

Pros

Labs with focus topics. Many research units in France have some focus topics, with several researchers working on related problems. Indeed, one of the selection criteria for CNRS jobs is the quality of your integration plan, where it helps to identify possible collaborators in your future lab. Having colleagues with similar interests can greatly improve productivity and is more fun. In contrast, in many other countries, departments aim for high coverage, with their faculty studying many different topics. Thus, there it can be a downside to have someone in the target department who does similar work to you.

Group grants. A significant fraction of research funding in France comes in the form of grants for consortia of many researchers spanning several universities. Examples are ANR AAPG grants or the Campus France PHC partnerships funding international collaborations. In my experience, this fosters a collaborative spirit, and it can allow you to have access to (some) funding even at a very junior career stage.

The academic environment. This might be particular to Paris, but the academic environment here is extremely stimulating, with a constant stream of visitors who are attracted both by tourism and the high density of universities. I don’t think this is matched by other cities.

Free time. Because I don’t have to teach (and choose to teach little), I have plenty of time to devote to activities that are “public goods”, yet that many other researchers don’t find time for. This includes writing software, writing books, editing Wikipedia, building websites that explain research to the public, writing group grant applications, etc. In addition, I usually have enough time to “think ahead” about what research I want to do in the future.

Donations are tax incentivized. Donations to charities are treated extremely generously by the French tax system: 66% of your donations can be deducted from your tax burden (not your taxable income, but your tax!). Another way of looking at this is that the French government matches your donation twice: you donate €100, France donates €200 to the same charity. If you’re a utilitarian or an Effective Altruist, that’s a great opportunity to help others.

Cons

Low pay. The salary paid to academics in France is signficantly lower than in many peer countries such as Germany, and very much lower than in the US. If you live in central Paris, you’ll probably pay at least €1100/month in rent. This limits what you can do in your life. (One reason for the low pay are transfers to the old; my take-home pay is lower than my pension contributions.)

Language. French is needed to have a comfortable life in Paris and in France. English works within academia, and the PhD students in my lab are quite international. Among permanent researchers in my lab, English is fine but some of my colleagues would prefer to stick to French in social contexts. But this may vary a lot between labs. The administration in the university operates in French (so I spend a lot of effort on translating emails), as does daily life. As far as I can tell, there isn’t much in terms of expat communities in Paris. So it clearly makes sense to invest into learning the language once you have a permanent position, but in the mean time, language is a barrier. On the other hand, I’m still far from being a fluent speaker, and things are going okay.

Bureaucracy. People are complaining about bureaucracy in all countries, though they do for different reasons. I’ll mention my main issues in France so you can see whether they seem bearable to you.

For travel expenses, your travel needs to be registered in advance, and you need to book all flight/train tickets and all hotels all at the same time so they can be approved by the administration. This causes me significant stress; I would usually want to book flight tickets far in advance, and then book hotels and local travel when my plans become clearer. Not possible here. In contrast, in the UK and the US I just paid for everything myself and asked to be reimbursed later.
All kinds of tasks (travel, job offers, job contracts) need to be “validated” by an entire chain of people in the right order. Frequently, some parts of the chain are on vacation or otherwise unavailable, which could mean you cannot post a job offer or cannot get a conference trip approved on time.
The administration is closed for 4-6 weeks in summer and another 3-5 weeks in winter (when the books are closed). Nothing can be done in those periods.

Dominik Peters