Apr
12
2009

Italian deputies’ ideal points

In previous posts I’ve tried to explain why some traditional methods for identifying legislators’ ideal points, such as optimal classification, don’t work in the Italian parliament. (A legislator’s ideal point is the point in some political space, such as the one-dimensional left-right political space, which they would prefer to any other point in that space).

Given this, it seems natural to look for alternate ways of estimating ideal points. One such way is to look at cosponsorship. The idea is this: although voting for or against a bill has consequences (and therefore strategic manipulation of the vote is worthwhile), that’s not the same for cosponsorship. It’s a low-cost signalling mechanism, which is not really subject to strategic manipulation.

This is doubly true for cosponsorship of motions.

So, using the same methodology as Aleman, Calvo, Jones and Kaplan [described here], I used cosponsorship of motions in the 15th legislature of the Italian Chamber of Deputies to work out legislators’ ideal points.

The graphic below shows the ideal points for the seven main parliamentary groups — Rifondazione Comunista, the PD, Italia dei Valori, the UDC, the Lega Nord, Forza Italia, and Alleanza Nazionale.

Two dimensions are plotted — the first dimension seems to be the standard left-right dimension. I’m less sure about the second dimension. In some legislatures it comes up as a north-south dimension, but that’s not the case here. Party labels are put at the centroid of each group (which, just to make things confusing, is not the same as the dimension by dimension median, which is more commonly used in political science).

What can we tell from this?

  • The parties run, left-right: RC > PD > IdV > UDC > LN > FI > AN
  • In terms of parliamentary behaviour, the LN deputies are contained within the union of UDC ^ AN
  • There are a number of Forzista deputies who are within the convex hull of the PD and IdV — Berlusconi, for example (though Berlusconi signed few motions, so take the position with a grain of salt), and Fabrizio Cicchitto (which makes sense for someone who came from the PSI).
  • Gianfranco Fini lies to the left of the AN convex hull, and is within the LN and UDC convex hulls…
  • Emilia de Biasi (PD) and everybody’s favourite transsexual parliamentarian Vladimir Luxuria are the two most left-wing deputies.
  • Marco Zacchera and Roberto Pedrizzi are the two most right-wing deputies. I know nothing about these two.

Any comments about the positioning of these deputies is more than welcome. Additionally, anyone who wants to see the R code, raw data, or raw coordinates used to plot these points, is welcome to contact me. Finally, any hints about how to display this information better would also be appreciated.

Powered by Zoomify

posted in italy, parliament by Chris

Follow comments via the RSS Feed | Leave a comment | Trackback URL

3 Comments to "Italian deputies’ ideal points"

  1. Caminadella wrote:

    How do you interpret the size of the areas? Is FI’s area so large because of the sheer number of MPs or do they cover a larger part of the political spectrum?

  2. Chris wrote:

    Thanks for the comment: Forza Italia deputies certainly cover a larger part of the political spectrum. But this might result from a very heterogenous party (and Forza Italia’s deputies are the most “spread out” of all parties — that’s too say, the have the highest variance on both dimensions, apart from the gruppo Misto), and it’s not certain by any means that the political spectrum which has been unpacked here has anything to do with the political spectrum as perceived by voters.

  3. Jim Newell wrote:

    Thanks for this. I am unsure about what is going on here, in particular about how deputies’ co-sponsorship behaviour translates into their positions on your diagram and about how one can tell whether the relative positions are accurate/seem appropriate/make sense or not without knowing what one of the dimensions is (as I understand it, one is the left-righ dimension but the other is unknown as yet). Perhaps it would help to turn the question on its head and to ask: Is there any reason for thinking that the positions might be wrong? — you mention that the parties run left-right RC > PD > IdV > UDC > LN > FI > AN — which does seem to fit, more or less, what we tend to think of as their relative placements in left-right terms. Sorry I cannot be of more help. I am sure that you don’t want to waste time having to try to explain it all to me over the internet but if at a future conference or similar occasion you care to have a chat about it, then I will be delighted to share my thoughts for what they might be worth.

Leave Your Comment

 
Powered by Wordpress and MySQL. Theme by openark.org