Based on the idea of ‘non-decision-making’ approach, such a concept of structural power includes not only a component of intentional agenda setting, but also a reference to an impersonal ‘mobilisation of bias’ where social structures systematically favour certain agents. Such an understanding of power is common currency in *dependency writings, both Marxist and non-Marxist, and in present Gramscian approaches.

Such a conceptualisation has been criticised for deducing power from rewards, the so-called ‘benefit fallacy’ of power. We usually do not call a free-rider powerful who certainly profits from a certain systemic arrangement, but who basically remains at its mercy. But the benefit fallacy exists only within a causal framework itself. To say that a system benefits certain people does not mean that they caused that benefit or that they control it. It just means that for understanding power in a social system, it seems odd not to take into account the effects of such a social system which can systematically advantage some actors. In other words, in terms of the second link between rule and outcome, systematic benefits are relevant.

And here it is very important, within which framework of analysis such systematic interest-furthering is conceived. Rational choice approaches can, of course, account for the privileges and advantages that stem from the social position of actors, but for which they did nothing. They can do so, but then only as ‘systematic luck’ (Dowding). By reducing a systematic bias to a question of luck, this approach leaves out of the picture the daily practices of agents that help to reproduce the very system and positions from which these advantages were derived. For this reason perhaps, Dowding rephrases his approach and now explicitly includes systematic luck not into the concept of power but into a more general power analysis. Such choices are not innocent as the last section will develop.