Erik Moberg ã:
A Theory of Democratic Politics


The second hypothesis, that parliamentarism gives political parties strong incentives for cohesion and discipline, has not, to my knowledge been systematically discussed anywhere. I will therefore start from scratch with the logical foundations of the hypothesis as I see them.

Using the concepts of decisive and blocking sets as defined in part 2.1 it seems natural that legislators in general are trying to form, or become members of, decisive sets in favor of their own ideas or positions. If they do not succeed with that they will at least try to bring about sets which are able to block proposals they are against.

Now, it is an important property of a parliamentary system that most decisive constellations necessarily consist of the executive, or at least has the executive as a constituent part. It is thus not possible that the executive is excluded from most decision-making constellations since that would be tantamount to regular no-confidence votes, and thus against the parliamentarian principle. The legislators thus have incentives to make the system work, and therefore, since stable, cohesive and disciplined parties according to part 5.3 is a prerequisite for that, to form such parties.

The incentives may however be stronger in a proportional parliamentary system than in a majoritarian one, in particular if the majoritarian system (because of the operation of Duverger's law) comes close to being a two party-system. This is so because the majority, in a two party situation and for statistical reasons, often is considerably bigger than the minority. The majority which supports the executive therefore can afford some defectors and still deliver the necessary confidence. In proportional systems, on the contrary, where coalition executives are the rule the executive's marginals are usually much narrower and discipline therefore more important. (The nature of these reasons, both the statistical mechanisms in the majoritarian case, and the mechanisms behind the narrow margins in the proportional case, will be treated in more detail later on.)

Anyway, the existence of incentives among the legislators is not enough for cohesive and disciplined parties to appear. Without some further mechanism, such as reasonably effective means for enforcing discipline at the party leaderships' disposal, the parties are likely to remain rather loose and unconsolidated organizations. This is so since the parties, in spite of being parties, are likely to harbor, within themselves, considerable differences of opinions, which it is difficult to erase. The establishment of common clear-cut party positions is likely to require some kind of organizational power. That topic will be dealt with in parts 6.3 and 6.4.