Why programmes resemble each other while electorates differ

In a competition where one must gather support beyond one's base, an offer has an interest in not straying too far from the bulk of the distribution of preferences. This reasoning, whose best-known formulation comes from Anthony Downs, predicts a convergence towards the middle when competition is played out mainly on one dimension.

Convergence has a practical limit: moving too close to the opponent discourages one's own base, whose mobilisation is necessary. Offers therefore oscillate between the attraction of the middle and the need to distinguish themselves, which produces a differentiation more marked in tone than in measures.

Part of the content of programmes is determined by constraints that do not depend on preferences: international commitments, legal framework, budgetary situation, timetables already under way, administrative capacities. On these grounds, the real space of choice is narrower than the space of discourse.

This produces an asymmetry that feeds disappointment. Debate bears on orientations, where the difference is large, while implementation bears on margins, where it is small. Noting this gap says nothing about the sincerity of the actors, only about the structure of the problem.

To this must be added the effect of administrative learning: once in power, teams discover the same files, the same services and the same constraints as their predecessors, which mechanically brings decisions closer together.

The most stable differences are not in the announced objectives but in four less visible places: the distribution of the effort demanded, the instrument chosen for a given objective, the order of priorities when everything cannot be done, and the conception of the level at which the decision should be taken.

That is why comparing programmes on their intentions gives an impression of similarity, whereas comparing them on those four points makes clear gaps reappear. It is also what a multidimensional profile measures, since it asks about trade-offs rather than intentions.

Part of the divergence between electorates bears on subjects no offer takes up. A strong disagreement in the population may remain absent from the competition, for want of an actor to carry it, and it therefore does not appear in the programmes.

That is one of the sources of the feeling that nothing distinguishes the offers: what distinguishes the voters is not what is being debated. The mismatch is not a lie, it is a problem of subject selection.

To compare two programmes, four questions are worth more than a complete reading. Who pays, and who receives? By what instrument? What is given up, since not everything can be financed? And at what level is the decision placed?

These questions have an advantage: they bear on elements that are verifiable and comparable, where declarations of intent almost always resemble one another.