Michel Volle left us in June 2024 at the age of 83, and those who knew him recognise that this is a great loss. He leaves behind a body of work of major importance for Official Statistics, the main lessons of which still resonate today. He examined both the profession of statistician and the fundamental tools of the profession, without shying away from the fundamental question of statistical use and adaptation to the opportunities and needs of the moment.

Michel Volle’s reading and interests were wide ranging. He had the distinction of having developed expertise and written benchmark works in INSEE’s three core professions: statistics, economics and IT. However, throughout his eclectic career, he also took an interest in topics such as history, philosophy, telecoms and theology, and regularly published entries on all these topics on his blog (Ouvrir dans un nouvel ongletwww.volle.com) from 1998 to 2018. The table of contents for his blog posts takes up around fifty pages on its own! Yet he liked to remind people that it was learning about statistics and their production that had given him discipline, a framework and a demanding nature that would be useful to him for the rest of his life.

We are limiting ourselves here to subjects that are relevant to the Courrier des statistiques, which means excluding several major aspects of his work, especially everything that relates to the field of telecoms, and addressing only a small part of the rich intellectual heritage that he leaves behind.

Although he left INSEE when he was only around 40 years old, he still leaves a lasting impact on generations of statisticians. We are of course thinking of his soberly titled course “Analyse des données” (Data Analysis), which, in itself, is precise, ordered and clear. People are less familiar with “Le métier de statisticien” (The Profession of Statistician), although it is a benchmark for how to produce statistics, many elements of which are still relevant in this age of data science: the importance of units, codes or classifications; registers and questionnaires; collection, verification, extrapolation; use of administrative information sources; merging files; publishing; etc.

He was a co-author, alongside Bernard Guibert and Jean Laganier, of the article that defines the concept of classifications, an article that today remains the benchmark on this subject. Subsequently, in his writings, in the same way as Alain Desrosières, he would continuously stress the importance of this subject, its intrinsic difficulty and its structural nature for statistics.

With a penchant for introducing a historical dimension in his writing, he would always point out that he was not a historian by profession, and that there are people much more erudite than him; but this would not prevent him from contextualising statistical activity by recalling the stages since the fourteenth century in Florence, or from producing an exciting Histoire de la statistique industrielle (History of Industrial Statistics) in 1982, based on work that he was able to carry out freely in the INSEE Research Unit between 1975 and 1978 and which led to the theory he advances in Histoire. This rich history, which is often cited as a reference point and is illustrated with numerous descriptions, begins in the 1930s and focuses for the most part on the period 1950–1975, the period of the pioneers who introduced business statistics.

Although he was interested in the past, including over a long period, as a demonstration of his interest in the issue, he also looked at the future. For example, in 1989, he published a “Rapport général sur l’évolution à moyen terme de l'appareil statistique français” (General Report on the Medium-Term Evolution of the French Statistical System), which begins with this fundamental question: “À qui et à quoi sert la statistique ?” (For whom and for what purpose are statistics used?). His initial observation that the “client” of statistics is not easy to define remains valid. Using this question as a starting point, his report was structured around four cross-cutting issues:

The discourse is sufficiently general to be only slightly dated: its questioning of the use of statistics is still relevant, and increasingly so in view of the “data deluge”.

As an economist, he became aware of the importance of technologies in the economy at a very early stage, writing “Économie des nouvelles technologies” (Economy of the New Technologies) in 1999, examining production functions with increasing yields, then publishing “e‑conomie” (e‑conomy) in 2000. He showed the fundamental nature of the phenomenon of computerisation, which he believed to be insufficiently studied within the economic literature. He delved further into his examination of predatory practices in economics in “Prédation et prédateurs” (Predation and Predators) in 2008, a book in which he modelled the key mechanisms of predation and money laundering. In 2014, he wrote a benchmark book on the concept of iconomy, that is to say the society brought about by the “third industrial revolution”, that of computerisation, and he co-produced the associated dictionary. He also spoke on this subject at a recent Conseil national de l’information statistique (National Council for Statistical Information, CNIS) roundtable event in 2018.

His major contribution in the field of computer science may seem more surprising: as a statistician-economist, computer science was not necessarily his reference universe initially. Many of his articles on the blog he created in 1998 focus on information systems, and this blog has long been a goldmine for people studying this subject, as it brings readability and theoretical hindsight to often arduous and technical matters. His blog articles are impactful, generally simple to read and effective, usually brief, to illustrate an idea (see, for example, “L’expression des besoins et le système d’information” (The Expression of Needs and the Information System), or “Éloge du demi-désordre” (Praise for Semi-Disorder)), although they are sometimes more thorough (“Évolution du rôle du système d’information : du concept au processus” (Evolution of the Role of the Information System: from Concept to Process)). They form so many building blocks that would allow him to create a crowning achievement for which there are few or even no equivalents: “De l’informatique” (Information Technology) in 2006, again published by Economica. It addresses a very wide range of topics: physical characteristics of computers, modelling, quality of service, development, project management, workflow, history of IT and the Internet, internal marketing, organisation, networks, business strategy, etc.

As is plain to see, to Michel Volle’s mind there were never “noble” subjects in contrast to “less noble” subjects. His approach, on the contrary, was a persistent attempt to understand complex phenomena by addressing all of their aspects, giving them a clear general formal structure, often based on mathematics, a solid historical context and intentionally simple insights. He could also convey a certain forcefulness in some texts and was not afraid to express his irritation or anger.

We have lost an eminent personality, a tireless pedagogue who was curious about ideas and beings, always open to debate and fervently committed to public service.