Return to: Search Page or to: Table of Contents Vol. 20, Issue 4

Allan Sikk and Rein Taagepera, "How population size affects party systems and cabinet duration," Party Politics, 20 (July, 2014), 591-603. [Available at http://ppq.sagepub.com/content/vol20/issue4/ ]

First paragraph:
Do party system characteristics differ in small and large countries, all other conditions being the same? If they do, then what is the specific average relationship to the population, and why? One may expect that tiny countries would not have room for many parties. With fewer parties to divide the pie, the largest party's share might be larger than is the case in larger countries. Also, the governmental cabinets might have fewer parties, which might make them more stable.

Figures and Tables:
Table 1. Population, seat allocation rule, mean district magnitude (M), assembly size(S), largest seat share effective number of legislative parties (N), and cabinet duration (C) for 25 one-seat and 11 multi-seat systems.
Table 2. Actual geometric means of system characteristics and predictions based on mean population, 24 M 1/4 1 countries
Table 3. Actual geometric means of system characteristics and predictions based on mean population, 11 M > 1 countries.
Figure 1. The largest seat-share versus population for M 1/4 1 countries.
Figure 2. The largest seat-share versus the product of population and district magnitude cubed.
Figure 3. Effective number of assembly parties versus the product of population and district magnitude cubed.
Figure 4. Deviation from theoretically predicted N versus N predicted by regression.
Figure 5. Mean duration of cabinets versus the product of population and district magnitude cubed.
Table 4. Predicted characteristics for countries with nationwide PR, and factor by which the actual value is off.
Figure 6. Effective number of parties for given population and district magnitude.
Figure 7. Residuals of largest seat-share versus population.

Last Paragraph:
While population has a definite overall effect on size characteristics of party systems, for individual countries this effect is heavily blurred by other factors. Only FPTP countries of less than one million are markedly hemmed in; they could expand their one-and-a-half party systems only by shifting to PR. Thus population is not destiny, as far as party system is concerned.

What are the practical implications? Countries are not prisoners of their population sizes. They are not locked in regarding their choice of party systems as the limits that population size puts on party systems are mediated by the choice of electoral rules.

Last updated June 2014