Mais les établissements publics (EPIC) comme était EDF avant sa transformation en société anonyme en 2004, ou la RATP jusqu'à aujourd'hui, sont bien dedans ?

Oui et de même si elles font des pertes et qu'elles sont renflouées par l'Etat. Par contre si elles s'endettent auprès d'autres acteurs ce n'est donc pas compté

Si non, est-ce que cela signifie que la transformation de la Poste ou d'EDF, par exemple, d'administration publique ou d'établissement public à entreprise à capitaux publics aurait changé le calcul du déficit ?

The summary exercise has forced us to make choices, I hope the paper remains readable and not too arbitrary Our goal is to show that a rigorous and fruitful use of the notion of exploitation is possible, but also to stimulate debate, so all reactions are more than welcome!

We conclude by an emphasis on the empirical relevance of the multiscalar "chains of exploitations" mentioned above.

We discuss four main forms of exploitation in the contemporary society: within the production unit, in market transactions (including rentier and financial relations), in the domestic sphere, and by the state.

2. We then map this conceptual framework on the existing empirical literature in social sciences, showing that this literature often speaks of exploitation even when it does not use the term.

We extend it by stressing the variety of non-exploitative institutions that can be considered in such a framework and the need to apply it at various distinct scales so as to be able to identify layered networks of exploitation relations.

To overcome these, we build on Roemer's counterfactual approach that identifies exploiters as those who would lose from the transition to a pre-defined socialist order.

1. On the theoretical level, we stress the limits of the classical marxist approach (exploitation as appropriation of surplus labour) as well as of the neoclassical approach (exploitation as rent extraction).