the strength of the rules

Image: Hamilton Grimaldi


The end of proportional coalitions will change the configuration of political parties in Brazil

The high number of parties in Brazil is an issue that is part of everyday conversations about politics. Recently, Congress approved the end of coalitions for proportional elections (Constitutional Amendment 97/2017), a measure that has the potential to reduce the number of parties over the years, as exposed by Jairo Nicolau in a text published in the Election Observatory/UOL.

Coalitions for proportional elections worked in such a way that the votes of coalition parties were added together. Thus, small parties could form alliances to survive in politics. With the end of this possibility, candidates for proportional positions – councilors, for example – tend to concentrate their forces in a few parties.

Graph 1. Average number of parties disputing councilor(s) by municipality

Source: Own elaboration based on data from the Superior Electoral Court (TSE).


Graph 1 shows the average number of parties disputing councilor positions in Brazilian municipalities in the years 2012, 2016 and 2020. This year, there is a drastic reduction in this average, which shows the strength of the new rule. However, there are differences between large and small municipalities. Applying the cut of 200 female voters (s) – the number used by the TSE to define which municipalities will have a second round in the race for mayor (o) –, it is possible to observe that the cities that exceed this mark have a higher average number of parties in the race .

Graph 2. Average number of parties disputing councilor(s) positions by municipality – Excerpt: 200 thousand female voters(s)

Source: Own elaboration based on data from the Superior Electoral Court (TSE).

In Graph 2, it can be seen that in larger municipalities the average number of parties disputing positions in City Councils also underwent a reduction in 2020, but is still higher than that observed in smaller municipalities. One of the explanations for this phenomenon is the limit of candidacies per slate: for the position of councilor (a), each party can only launch a number of candidacies that corresponds to 150% of the number of vacancies available (according to Law nº 9.504/1997 ). That is, in a city with ten seats in the Chamber, a party can only launch 15 candidacies. In larger municipalities, parties tend not to completely fill their slates. Therefore, there is a larger “market” of votes, opening up space for more subtitles to enter the dispute.

In larger municipalities, the average number of seats available in City Councils is 23,8, varying between 15,9 and 31,7. If we separate the municipalities that have 15 seats or less, as shown in Graph 3, the average number of parties disputing vacancies in municipal legislatures is even smaller, reaching 6,7 parties in 2020. In contrast, this year, municipalities with 16 seats or more have an average mark of 20,2 disputed captions.

Graph 3. Average number of parties disputing councilor(s) by municipality – Excerpt: 15 seats in the Chamber

Source: Own elaboration based on data from the Superior Electoral Court (TSE). The number of vacancies used refers to the year 2016, as it was not possible to capture the variations that occurred in the municipalities over the last legislature.


In terms of strategy, the parties are adapting to the new rule. The trends are twofold. First, that the number of subtitles in competitions reduces over time. Second, that these parties launch the largest possible number of candidacies, to reduce the effects of the end of the coalitions.

Finally, one cannot ignore the fact that Brazil has the most fragmented party system in the world. In this way, the end of proportional coalitions can undermine the strength of a party only in certain regions, and it can survive if it has strength in others. Thus, another possible development for the coming years is a process of regionalization of Brazilian political parties.

*Otávio Z. Catelano is a master's student in political science at Unicamp.

Originally published on 2020 Election Observatory of the Institute of Democracy and Democratization of Communication (INCT/IDDC).



See this link for all articles