Erik Moberg ã:
A Theory of Democratic Politics


Using the hypotheses about party-shaping mechanisms it is now possible to derive conclusions, also hypothetical, about the constellations of main actors in the four constitutional systems. The result is shown in the figure below.

The upper, left part of the figure illustrates the main actor constellation in parliamentary systems with proportional elections. Here, there is no strong constitutional mechanism reducing the number of parties, and hence we are likely to find considerably more than two parties, perhaps five to ten. Furthermore the incentives for upholding party discipline are strong, since the system is parliamentarian, and effective means towards that end are also available, due to the proportional elections. The parties are thus likely to be disciplined and cohesive. Looked at from outside they may be considered as unitary actors. This is the reason why the party-actors are indicated by solid circles in the figure.

In the upper, right part of the figure we have parliamentary systems with majoritarian elections. The number of parties shall, on the average, and because of the number-reducing tendency of the majoritarian elections, be smaller than in the former group of constitutions. The figure shows a case for which the reduction has been very effective, and where, as a result, there are only two parties. Obviously, and even with the number-reducing mechanism operating, there may however be more than two parties. Anyway, the parties are likely to be less disciplined than in the former case. One reason is that the incentives given by parliamentarism, according to the hypothesis in the preceding part, are weaker than in case of proportionalism. Another reason is that the means for enforcing discipline are weaker than in a proportional system. Since the discipline is less than perfect the party actors are indicated by broken circles in the figure.

Turning then to the lower parts of the figure, which show presidential systems, it may at first be noted that the actors are placed in two levels rather than one. The presidential executive is placed at an upper level, whereas the actors in the legislature are indicated at a lower level. The basic reason for placing the president at a separate level is that the rules regulating his or her actions usually differ in important respects from the rules for the legislative actors - particularly by giving the president much more power. Furthermore, as will be developed in the following, the president's general incentives may differ considerably from those of the legislative actors. The president may, for instance, sometimes be endowed with incentives for favoring the common good rather than partisan interests.

Now, the lower, left part of the figure, shows the constellation of actors in presidential systems with proportionalism. Here, in the elections for the legislature, there are no strong constitutional forces reducing the number of parties, and consequently we may expect quite a lot of them, as indicated. As for incentives for discipline, there are obviously no ones of the same nature as those in a parliamentary system, but perhaps there are other ones. Anyway, the means for discipline are there, since we are dealing with a proportional system. The parties may thus be disciplined and cohesive, but it may also be the other way round. In order to settle this issue additional assumptions about the incentives for discipline are required. So far, I have however not introduced any such assumptions, and the resulting ambivalence is illustrated in the figure by some broken, and some solid, circles.

The lower, right part of the figure, finally, shows the constellation of actors in presidential systems with majoritarian elections. Here, the majoritarian elections will certainly reduce the number of parties. At the same time, however, and since there are no assumed incentives for party discipline, and neither any means, the parties will lack discipline, and in that sense be relatively unimportant. The individual legislators will be free to act on their own to a very considerable extent. The main actors will thus be individuals rather than parties. This is indicated by the many small circles in the figure.