Saturday, April 11, 2015

Legislative Council Voting Patterns 2011-2015

Advance Summary

1. This article presents a revised analysis of voting patterns in the Legislative Council (the upper house of Tasmanian Parliament) based on contested divisions in the last four years.

2. Although there is a degree of independence in all Legislative Council voting, the Council continues to have a clearly defined "left wing" consisting of Craig Farrell (Labor), and independents Mike Gaffney, Ruth Forrest, Kerry Finch and Rob Valentine.

3. Excepting Adriana Taylor and increasingly Rosemary Armitage, the remaining MLCs can be considered to lean to the right to varying degrees.

4. In the past year several Legislative Councillors have displayed more polarised voting patterns.  Especially, Tony Mulder has more often voted with the endorsed Liberals than before, though this may well be a reflection of a changing mix of voting issues.

5.Despite this, there remains a diversity of positions on the "right" side of the Council and a lack of strong clustering in voting patterns there.

6. The revised analysis again shows most of the conservative MLCs to be between Liberal and Labor positions, albeit closer to the Liberals.

7. While the Liberals have had greater success on the floor of the Council since coming to office in the Lower House, they have still been defeated on several divisions.


Three Tasmanian Legislative Council seats go to the polls in May. Both of the independent incumbents seeking re-election are already facing accusations that they are too close to one political party or other.  The ALP campaign in Windermere (see Windermere preview page) has sought to link Ivan Dean with the Liberal Party. In the Mersey campaign (see Mersey and Derwent preview page) Mike Gaffney's opponent Vivienne Gale has wasted no time saying he is too close to the ALP.

With these claims being made I decided it was time for an update of my assessment of Legislative Council voting patterns.  The last instalment (2010-13) was released over a year ago, and since then we have seen a change of state government.  How much difference has that made?

As explained in previous editions, this analysis is about how Legislative Councillors vote on issues on which they do not all agree.  Many pieces of government legislation are uncontentious and passed unanimously; these are of no interest to me here.  I'm interested in looking at who tends to line up with who on those motions that are contentious enough to attract at least one vote on each side.

I've decided that each time I review this article, I will use just the last four years of data.  I think that is about the right balance between letting ancient data go and thus allowing changes in MLCs' behaviour to be reflected on the one hand, and not letting the issue mix and small sample of any given year dominate the findings too much.

Since the last review there have been 24 distinct contested divisions (I usually count two identical divisions on different parts of the same bill as the same thing) but by far the biggest contributor to the tally was the Protection from Protestors bill (eight divisions included).  The issue makeup since the last election has also included several environmental and law and order votes, but so far none of the social-issue conscience votes that cropped up now and then in the previous parliament.

I've again aimed to produce two different descriptions of the observed voting patterns.  One of these is two-dimensional and the other is a left-right sort.  Again this closely follows the methods spelt out in the attachment to an old TT piece for local councils, but with council-specific fiddles removed.  Again I had to use assumed agreement values at one point (this time for Wilkinson with Hiscutt and with Armstrong - I used 64% and 68% respectively).  There is also a minor judgement call in the ordering, mentioned later.

In Two Dimensions

The following is a two-dimensional representation of the voting patterns of the fifteen current MLCs since the 2011 LegCo election (or if elected since then, since they were elected).  Jim Wilkinson as chair no longer votes on the floor of the Council (except on tied motions on which he is typically bound by precedent) but is included anyway (it's possible he will return to voting on the floor in future).

For those unfamiliar with graphs of these sorts, a principal components analysis aims to represent patterns in 2D with as little distortion as possible.  Both the angle of different lines to each other and the distance of different data points from the centre are relevant here.  The angles indicate what kind of voting pattern is displayed and the distance indicates how strongly it is realised.  Even if two Legislative Councillors appear opposite each other, if one is close to the centre they will still agree reasonably often.  If two Legislative Councillors are at a similar angle and a similar distance from the centre then it is likely their political views are rather similar.

Key to abbreviations: AI: Rosemary Armitage, AS: Robert Armstrong, D: Ivan Dean, FA: Craig Farrell (ALP), FI: Kerry Finch, FO: Ruth Forrest, GA: Mike Gaffney, GO: Vanessa Goodwin (LIB), HA: Greg Hall, HI: Leonie Hiscutt (LIB), M: Tony Mulder, R: Tania Rattray, T: Adriana Taylor, W: Jim Wilkinson, V: Rob Valentine

(Although twelve MLCs are "independents" some of these have known past party connections, as discussed last time.  Tony Mulder stood for election as an "independent liberal", ie as a party member whose candidacy was not officially endorsed.  A claim connecting Ivan Dean with past Liberal Party preselections is discussed and found on current evidence to be misleading at best in the Windermere section.)

There is not much difference between this year's graph and the 2013 edition.  There is still a cluster on the left, a large cluster between that cluster and the Liberals, and Tony Mulder still has a vector all of his own. The main change worth noting is that Dean moves from a position between the two Liberals to a position closer to the main conservative cluster.  There is still one MLC between the two Liberals, but this time it is newcomer Robert Armstrong.

Left-right sort

The following is an agreement matrix, which shows the percentage of contested divisions on which each pair of MLCs, if both present, voted together.  For instance the table shows that Robert Armstrong and Rosemary Armitage voted together 50% of the time.

The highest agreement percentages were 95% for the two Liberals (Goodwin and Hiscutt) and 91% for Armstrong with each of the two Liberals.  The lowest agreement percentages were 14% for Valentine and Hiscutt,  18% for Valentine and Armstrong and 21% for Forrest and Hiscutt.

A figure of three-quarters (75%) is often used as a cutoff for identifying clusters of members with similar views in this sort of work.  I've highlighted percentages of 75% and above and also weakly highlighted those between 70 and 74.

I sorted the MLCs from left to right using the same comparative methods as detailed in my old HCC methods piece (warning: wonk factor 5, casual and even normal readers should avoid!) but with one exception.  When it came to the placement of Armstrong, the comparative method placed Hiscutt as the most right-wing member of the Council, followed by Goodwin and Armstrong.  But my older method of computing ratios of agreement with "left" and "right" (this time treating Taylor and Armitage as neither) placed Armstrong as the most right-wing, followed by Hiscutt and Goodwin.  The reason for this is that while Armstrong has a higher tendency to agree with the right of Council on average than Hiscutt or Goodwin, he is more likely to agree with a broad range of the right side, while the two Liberals are more likely to agree just with the hardliners.

I don't think either of these methods is perfect so I've split the difference between them and placed Armstrong between Goodwin and Hiscutt.  In any case it sometimes turns out that the limited data from a new MLC's first year in the job are unrepresentative.  This was certainly the case with Armitage, and the heavy focus on forestry issues in divisions in Armstrong's first year means his results here should be treated with some caution.

Highlighted ratios in red represent left of centre, and those in blue represent right of centre.

Once again the left-leaning cluster of five MLCs in the top left of the matrix is extremely clearly defined.  In the bottom right corner things are less clear-cut because of the variety of more or less right-wing positions in the Council but there is an obvious cluster of the two Liberals and Hiscutt, with Dean, Mulder, Hall and Rattray all having above 70% agreement with at least one of these.

Pattern changes

There have been some changes in the ordering since last time. Two minor changes in my left-right ordering are that Forrest and Gaffney have swapped places (again) and Rattray has again swapped places with Hall.

The largest change involves Mulder, who was between Taylor and Armitage in the previous chart but is now placed to the right of all of Armitage, Wilkinson, Rattray and Hall.  Since the election of the Hodgman government Mulder has voted with the (official) Liberals on 79% of contested divisions, whereas up to the end of 2013 he voted with Goodwin only 46% of the time.  This probably largely reflects the changing issues mix in the House, with forestry and law and order issues more to the fore, and social issues (on which Mulder often disagrees with the official Liberals) absent since the Hodgman government was elected.

My methods don't try to impose any judgements about which issues are most important, so from time to time some leanings of some MLCs will be brought out more strongly or disguised by the mix of issues being covered at the time.  Most likely the overall picture will be most accurate when the sample includes a mix of different Lower House governments.

The other major change is the increased stretching of the ratios of agreement since the Liberals came to power.  The voting patterns of Forrest, Gaffney and Finch are more strongly left-leaning in this year's analysis, and the patterns of Hall, Mulder and Goodwin have become more consistently right-leaning.  Most of the others have changed little, with Armitage moving slightly towards the centre.  The movements of Dean and Rattray on the 2D graph aren't because their own voting behaviour has changed greatly, but are more because the rest of the LegCo right has become more inclined to side with the Liberals.

This doesn't mean the Liberals have had things all their own way lately.  The following are the percentage of times each member has been on the winning side of distinct contested divisions specifically since the Liberals came to power:

Taylor 95, Rattray 82, Dean 81, Hall 79, Forrest 67, Finch 65, Armstrong 65, Gaffney 64, Farrell 62, Armitage 59, Hiscutt 57, Goodwin 55, Mulder 52, Valentine 48

(Wilkinson has not voted, except on casting votes, in this period.)

Most of the Liberals' defeats have been minor, with public service pay freezes one big exception, but also numerous losses on amendments to the Protection from Protestors bill, which ultimately led to it being heavily revised.  The closely contested votes frequently see the whole left group all vote against the Liberals, at which point the Liberals need the support of six of the seven non-left independents to carry the day (assuming everyone is present).  Cases in which they can rely on all of Rattray, Dean, Hall, Armstrong, Mulder and Armitage have become rare (Armitage especially has voted against them more often than not since the state election) and as a result Adriana Taylor has become the crucial vote, appearing on the winning side of almost everything.

OK, I think that's quite enough to get my analysis of LegCo voting up to date and to hopefully inform debates taking place about the records of particular incumbents.  Some more comments are posted in the relevant election pages:

Derwent and Mersey


I will have live coverage of all three seats (though I don't expect much life in the first two) on election night, Saturday May 2nd.

No comments:

Post a Comment

The comment system is unreliable. If you cannot submit comments you can email me a comment (via email link in profile) - email must be entitled: Comment for publication, followed by the name of the article you wish to comment on. Comments are accepted in full or not at all. Comments will be published under the name the email is sent from unless an alias is clearly requested and stated. If you submit a comment which is not accepted within a few days you can also email me and I will check if it has been received.