by Robert Bond
I have been working on a project using legislative co-sponsorships to study polarization in Argentina’s Senate and Chamber of Deputies. Similar work on the U.S. Congress can be found here. The initial results are displayed in the two plots below.
The increase in modularity in the Chamber of Deputies suggests that communities within the chamber are becoming more closely knit, suggesting that polarization is increasing. However, the decrease in modularity when the partition is defined by legislative block indicates that the increase in polarization may not be due to polarization of parties (the pattern in similar when parties, rather than legislative block, define the partition). As modularity when the partition is defined by region is decreasing, region seems to be playing a decreasing role in the polarization of the Chamber of Deputies.
Argentina’s Senate paints somewhat of a different picture. Total modularity decreases initially, hitting a low point in 1993, before increasing over the present period, indicating that polarization decreased over the first half of the period of study, but may be increasing in the second half. Modularity when the partition is defined by legislative block initially seems to have defined community structure, before decreasing to near zero. The recent increase indicates that legislative blocks may once again be defining community structure and leading to party polarization in the Senate. The fact that modularity when the partition is defined by region remains near zero indicates that region plays little role in defining community structure, and thus polarization, in the Senate.