In this graph I’ve plotted estimates of Australia’s total resident population over time (mid 2002 to mid 2006, from ABS), as well as (near-monthy) enrolment numbers from the Australian Electoral Commission; vertical lines indicate the timing of state and federal elections. Total resident population is increasing linearly over the time period, but voter enrolment is more volatile, showing a big run up around the time of the 2004 Federal election, and then, curiously, a levelling off and even a decline. Indeed, as Stephen Luntz noted, if not for a boost in enrolment associated with the Victorian state election, the pattern of declining enrolments would be less ambiguous.

With the rolls closing the days writs are issued for the 2007 election, one doubts that there will be the big ramp-up in enrolment observed in 2004. The AEC says that in 2004, they got over 70,000 new applications for enrolment and over 400,000 enrolment cards in the week between the issuing of writs and the close of the rolls, so almost of all the run-up in enrolment ahead of the 2004 election came in one week. That “close-of-roll” week-long window for getting voter enrolments lodged/updated is gone now.

Methodological caveat: total resident population does not equal the population eligible/required to enrol to vote under the Commonwealth Electoral Act (18+ years of age, citizens, etc), but its a quick and dirty proxy and probably a pretty good one at that (i.e., over the 2-3 year window graphed here, simply take a constant proportion of the total resident population to get the eligible enrolee population). This means that the relationship between the two variables (population and enrolled population) as shown in the graph, and the disjuncture between enrolments and population growth after the 2004 election enrolment peak is probably real.

It would also be helpful to have enrolment data going back a little further in time, so as to see what the trajectory in enrolment numbers looked like immediately after the 2001 election? Maybe a slight decline after a big run-up in enrolments around a Federal election is par for the course?

This uncertainty aside, I suspect that what I said in a previous post is correct: its getting easier to get tossed off the roll and harder to get on it, and its almost certain that there will be no “surge” in enrolments in the weeks ahead of the next Federal election, given that the rolls the day that writs are issued.

Crikey – Politics Etc – Where have all the voters gone?

The Green’s Stephen Luntz writes for Crikey on an actual decline in the number of enrolled voters in Australia. The tag is misleading: we know pretty well for sure “where the voters have gone”. They are younger, more mobile people who are almost all renters. The numbers are somewhat surprising though — possibly half a million people who should otherwise be enrolled are expected to be missing by the time of the 2007 Federal election, on a base of about 13-14 million eligible voters — which is getting in the range of being politically consequential, if not seat-by-seat, then potentially in Senate election outcomes (where the states and territories are usefully thought of “at-large” multimember districts).

Section 101 of the Commonwealth Electoral Act makes it an offence for any eligible citizen not to enroll within 21 days of a change of address (or turning 18). I’m not sure of any prosecutions under this section, nor how vigorously the Australian Electoral Commission enforces compliance with this provision of the Act.

Is this a case of “benign neglect” from the AEC (taking their cues from the government that sees Labor and Green to be the net losers from people drifting off the roll)…? Note also that this is the government that just passed amendments to the Act that will close the electoral roll the day that writs are issued for a Federal election (see section 155 of the CEA).

So what should we infer? That turning up at a polling place is compulsory if you are on the roll, but they’re not really too serious about ensuring that people actually enroll…? At the very least the situation is somewhat odd, in that having made voter turnout compulsory, other provisions in the Act (and “on-the-ground” custom) do make it easy for enrolled voters to comply with this legal obligation: e.g., relatively many polling places, elections on a Saturday, widespread provision of absentee and pre-polling services, above-the-line voting in Senate elections, and the parties provide how-to-vote cards to help ease the cognitive burden posed by preferential voting in House elections (giving a complete rank ordering of candidates). That is, if you mandate that your population must do something, you make it easy. Is enrollment as “easy” as it ought be, as it should be, given that it is compulsory? Almost surely not. The AEC is certainly aware of the issue (see some interesting testimony to the Senate Finance and Public Administration Committee), but seemingly unable to do much about it: the 2006 amendments to the Electoral Act seem to make it easier to get tossed off the roll and harder to get on the roll.

Bottom line: voter enrollment in Australia is compulsory de jure, but not de facto, and in whose interests is that?

There has been some talk recently of how badly off the Liberals will be should they lose the next Australian Federal election. Every other (state and territory) government in Australia is a Labor government (at least for the time being), and so losing federally would probably see the Liberals in opposition throughout Australia. Kim Beazley and Norman Abjorensen (ANU, in The Age, and brought to my attention by Mumble) have made two recent interventions on this issue. Organizationally (they say), the Liberals will be up the proverbial creek in a post-Howard era, given the way Howard has centralized the party around himself and the federal secretariat etc.

I agree with Peter Brent at Mumble that these organizational explanations only carry so much weight. Both parties are enduring through a long-term, steady decline in branch memberships. And of course you’d rather have more of the spoils of incumbency than less. But there is a certain kind of “here we go again” tone to the current ruminations on party decline. Mumble has a killer grab from Howard himself on this, from 1995 (when the Libs were still in opposition, federally): “There’s nothing much wrong with the structure of a political party that a decent win wouldn’t cure.” My own take on this appeared in an essay I wrote for the Cambridge Handbook of the Social Sciences in Australia: “It would seem that party ‘crisis’ is the handmaiden of more than a parliamentary term or two of opposition in Canberra.”

There is a conventional wisdom of sorts that says the Liberal Party organizational structure (a formal divide between the organizational and parliamentary party; a federal model) leaves themselves especially vulnerable, particularly in the wake a strong, long-term, centralizing leader like Howard. From a 1996 essay by Dean Jaensch:

…the structure of the Liberal Party is sound, efficient and effective – for a party in government, with a strong and popular leader… For a party in opposition … without a clear, strong and popular leader, the structure is inappropriate – even a potential for disaster.

But once again the (unsurprising) implication is that winning an election cures what ails a party in “crisis”. The open question is just what role “good organization” of the party matters for winning elections, whatever that might mean (and by the way, contrast “good organization” of the campaign, a la the Liberals in 1996), and again, I’m not so sure its a pivotal role; i.e., compulsory voting delivers your punters to the polls, which means the party’s job is to help recruit candidates, get how-to-vote cards printed and handed out on election day, help get postals and absentees out etc.

And it also has to be restated that a lot of this speculation re the coming crisis in Australian conservative politics is premised on the idea that the Liberals will lose the next election. That surely has to be a big “if”, although Mumble reckons it could well be on (and he’s got runs on the board with his prescience re Latham).

I might just be the last person in Webdom to discover Khoi Vinh’s blog. Very nice layout. His blog says that since Jan 2006 he has been the Design Director for NYTimes.com. Sweet.

Rudd to cleric after rant: don’t come back – National – smh.com.au
Summer in Oz, not a lot of political news, and so not a bad get by Rudd. Mind you, the Sheik’s statements make for pretty easy politics.

Helvetica
The blog is worth a look too: apparently, there is a cafe in Melbourne called Helvetica, and this poster is also pretty awesome.

Also found this also on daringfireball.net.

Microsoft Windows 386 Promotional Video
The world has changed soooo much. Found on Gruber’s daringfireball.net.

I built JAGS a couple of years ago for my (old!) Mac G4/PPC laptop, but need a build for my office MacIntel. I’m yet to try this, but since John Brehm asked I’ll pass along the following instructions I got from Bill Northcott at unsw.edu.AU (yeah!).

Hi Simon

I am just in the process of rewriting the instructions to cover building with R-2.4.x on Tiger including Intel.

Roughly, but not yet thoroughly tested:
Install the 2.4.0 R binary distribution including compiler.
Set the F77 env var
export F77=/usr/local/gcc4.0/bin/gfortran

Make sure you have the latest Xcode 2.4.1 (this is vital on Intel 2.3 should do on PPC)
you must have gcc-4.0 as your default cc (sudo gcc_select 4)

Get the R 2.4.0 sources and expand somewhere in your home directory.

At the top level of the R sources run
./configure –with-blas=’-framework Accelerate’

Change to standalone directory:
cd src/nmath/standalone

make
sudo make install

This will put Rmath.h, libRmath.a and libRmath.dylib into the R framework.

Set two more env vars:
export CPPFLAGS=-I/Library/Frameworks/R.framework/Versions/2.4/Resources/include
export LDFLAGS=-L/Library/Frameworks/R.framework/Versions/2.4/Resources/lib

cd to the top of the JAGS source tree and build

./configure
make
sudo make install

This should work on JAGS up to v0.97. 0.98 will be very different.

Let me know how it goes.

Bill

Bobby Gillespie from Primal Scream complains about noise from a pub…? In the SMH.

I went to New Orleans last week for the annual meeting of the Southern Political Science Association. At the suggestion of former New Orleans resident Charles Barrilleaux (and inspired by sitting across from Harry Shearer on the plane ride to NOLA), I ventured away from the tourist-dense regions around the downtown hotels and the French quarter, although there are plenty of signs of loss and damage (and repair) still lingering there — but heading north and east out of the French quarter (roughly paralleling the river), you encounter neighborhoods (the start of the now internationally infamous 9th Ward) so thoroughly destroyed it is difficult to imagine them coming back (but people are trying). That near total devastation continues for block after block as you turn north towards the Lake out towards the University of New Orleans. Boarded up strip malls, churches, banks, gas stations, mud and dirt filling the car parks; parks and golf courses returning to their natural states; utility trucks doing the rounds, laying cable, just now getting services up; the fluorescent orange “TFW” (toxic flood water) spray-painted on house after house, still there 15 months later. The kicker is to hear about the murders there; like, how much more can a city take? Everyone should see it, and understand that it is far from over for so many people down there…

And yes, it would nice of the new Democratic majority in Congress to remind us about the ongoing disaster in New Orleans and kick some butt.

Last thing: I am hardly the first to notice the irony in the acronym for the New Orleans Police Department…