Saturday, January 28, 2017

Unicameral By Stealth: Tony Abbott's Senate Referendum Call

The Australian (Tony Abbott calls for Senate referendum, warns 'we are turning into Italy') reports that relevance-deprived ex-PM Tony Abbott has called for a referendum to reduce the power of the Senate to obstruct government legislation.

As outlined, the plan would allow for a deadlock between the Houses to be broken by a joint sitting of both Houses without the need for a prior double-dissolution election.  It appears that this follows one of two options outlined in this 2003 discussion paper on resolving deadlocks.  These options were:

1. A joint sitting can be convened after a bill has been rejected twice with three months between rejections
2. A joint sitting can be convened after every election for the full House and half the Senate, rather than requiring a double dissolution

It appears Abbott favours the first option, but this is not yet totally clear.  The 2003 proposals were killed off based on a finding that they would not pass a referendum.

Abbott's Argument

Abbott's argument, as paraphrased by Paul Kelly, is that the blocking of legislation by Senate crossbenchers and oppositions is producing a revolving door of Prime Minsters and is weakening the ability of governments to act.

Concerning the first, I don't agree.  The changes from Rudd to Gillard, from Gillard to Rudd and from Rudd to Abbott were all primarily caused by Labor infighting.  Senate obstruction did contribute to Rudd squibbing on "the greatest moral [..] challenge of our time", which helped kickstart his downfall, but Rudd was not obliged to squib.  Tony Abbott's own demise was largely caused by his own unpopularity and issues with his internal leadership style.  Far from Senate obstruction contributing, it probably protected his government from its own folly, especially by blocking nasty stuff in the 2014 Budget. In a similar way, the Howard government actually benefited from being unable to implement a more radical agenda in its early years, and helped snuff itself out with WorkChoices once it was no longer thus restrained.

Concerning the second part, there is a valid question whether the surge in minor party voting at the last two elections might produce a kind of vicious spiral.  Voters lose faith in the major parties, so they vote more and more for minor parties.  Minor parties fill the Senate crossbench and obstruct major party governments.  Those governments are unable to do anything and appear weak and beholden to a Senate circus.  So still more voters are driven to minor parties in protest at the weakness of the majors.  Rinse and repeat, presumably until you end up with a European situation with no party attracting more than about 20% of the vote, with an increased chance of hung parliaments in the House of Representatives and permanently huge crossbenches in the Senate.  At this stage though, this is all speculation, and even then, some will say bring it on.

Abbott quotes an argument from the 2003 paper that "In practice, the minority has assumed a permanent and absolute veto over the majority".  Now that one is just drivel.  Senate obstacles occur because the crossbench minor parties combine with the Opposition of the time to block things.  Frequently in such cases the parties voting against legislation represent the votes of more than 50% of voters after allowing for the preferences of parties that did not win a seat.  It's true these obstacles often arise from Oppositions blocking for strategic rather than policy reasons. It's also true that Opposition voters generally cheer such blocking and indeed anything that makes life hard for the government of the day.  When Oppositions vote with Governments, crossbench minorities cannot carry the day.

Another Abbott argument is that the 2013 and 2016 wins gave the Government a "double mandate" to pursue its ABCC plans.  This too is incorrect.  When a government chooses to go to a double dissolution the reward if successful is often a more easily delivered mandate.  The cost is that all previous mandates expire.  The fresh mandate achieved by the Coalition is a far from overwhelming one and should hardly stop the Senate pursuing strong amendments if the numbers exist there to do so.

Applying Abbott's Proposal To The Past

It is useful to see at what past elections the Abbott proposal would have given a government the ability to pass legislation through a joint sitting without having to go to a double dissolution.

Australia's Senate was entirely determined by proportional representation (apart from distortions caused by Group Ticket Voting from 1984-2013) from the 1951 election onwards.  The Coalition held a majority in both Houses from 1951-1955, from 1958-1961, from 1975-1980 and from 2004-2007.

Under the Abbott proposal the Coalition would have also held a majority in a joint sitting (assuming the same results) from 1955-1958, from 1961-1970, losing its joint sitting majority at the 1970 half-Senate election.  Labor would have come to office in 1972 with an initial dual-sitting majority of a single seat (see below for the special case surrounding the Dismissal).  The Coalition would have had a dual sitting majority from 1980-1983. The Hawke Labor government would have held a dual sitting majority until 1990. The Howard Coalition government would have held a dual sitting majority at all times, as would have the Rudd/Gillard Labor government from 2007-2010, and the Abbott/Turnbull Coalition government from 2013-2016.

The current government is well short of a joint sitting majority, but could pass legislation at a joint sitting with any eight votes out of 15 non-Greens crossbenchers, which is obviously better than trying to get nine out of 11.  (These figures apply after the High Court has finished cleaning up the messes left behind by Culleton and Day.)

The events leading up to the Dismissal are a special case.  Labor under Whitlam would have held 95 joint sitting seats out of 187 from the 1974 election (though that election was a double dissolution, which would not have been necessary under the Abbott proposal, so the result might have been slightly different.) Labor would have lost one seat upon the contentious appointment of Cleaver Bunton and another with the death of Bertie Milliner on 30 June 1975, leaving it with exactly 50% of seats until Milliner's vacancy was filled.  Assuming that Joh eventually got his way on the vacancy somehow or other, Labor would have been down to 93/187. It's doubtful the help of two crossbenchers in even version 1 of the proposed joint sitting model would have directly averted the Dismissal, but surrounding events would have played out very differently.

Thus only the Coalition from 1970-1972 and 2016-present, and Labor from 1990-1996 and for part of 1975, would have held a Reps majority but not a dual sitting majority, on the basis of the actual results of elections.

However had the reported Abbott proposal applied, especially version 1, many events would have panned out differently.  As noted above 1975 might have panned out very differently.  The Rudd government may have used a joint-sitting majority to pass its original climate-change plans, Rudd would not have been vulnerable to being rolled by Gillard (at least not then), the 2010 campaign would have been a different beast and Abbott himself might never have become Prime Minister.  Going back further, the Howard government might have successfully pursued a radical-right agenda from much earlier and been wiped out after one or two terms - the room for alternative timelines is endless.

It would seem from the above though that the Abbott proposal would effectively neuter the Senate except after very close House of Representatives elections, meaning that at most times a government winning a majority in the Reps could then pass legislation with only temporary obstruction from upstairs.

Perhaps, though, the Abbott system would have some unintended consequences.  In the case of Tasmania, the state's share of the power to block legislation would fall from 12/76 (15.8%) to 17/226 (7.5%).  It's at least possible that this would make major parties more inclined to disregard the state's interests, and lead to more interest in state-based parties that aimed to hold the country to ransom by capturing a joint-sitting balance of power.  It might also be that as it became easier for indifferent governments to win joint sitting majorities even when they didn't really deserve them, more voters would switch to minor parties in the Senate.

Would it be more democratic?

I used to be deeply suspicious of having two Houses of Parliament at all, having written stuff along the lines of that an obstructive Upper House insults the democratic ideal while a compliant one is just a parasite on the public purse.  However, that foolishness was partly a product of my environment - 1980s-90s Tasmania, with a 35-seat Hare-Clark lower house and an appalling, malapportioned, 19-single-seat Legislative Council full of  "independent" fringe-right homophobes who often hailed from small rural electorates.

These days (especially following the Tasmanian anti-discrimination debacle I was raving on about on here again just yesterday) I have a much more conservative view of these things.  I think it is most important that bad legislation has two chances to die, even if it means good legislation sometimes or even often perishes with it.

Under the current system, legislation has to pass two houses, neither of which is optimally democratic.  The House of Representatives embodies the principle of "one vote one value" well, but support in it is not proportional.  The government is formed by the major party that wins the 2PP vote in the most seats, but having a majority on the floor off around a low-40s primary is not a very reliable sign that every vote passed would be supported by parties that gather the majority of the primary vote.

The Senate on the other hand - especially now we've put that Group Ticket rubbish in the bin where it belongs - provides an excellent proportional representation system on a state by state basis, but is also severely malapportioned.  Voters in NSW and Tasmania elect the same number of seats per state, meaning that a Tasmanian voter's vote carries almost 14 times more power than that of one in NSW.  This distortion hasn't had the impact that might be feared, mainly because major party votes are reasonably even across states. Indeed, the 2016 result was highly proportional.  But some day major party votes might distribute more unevenly or small-state-based parties might flood the system, so the representativeness of the Senate is not that reliable.

To pass therefore, legislation must get through two chambers - the one where it has the blessing of one-vote-one-value, and the one where proportional representation (albeit potentially distorted by state factors) provides for political diversity.  The chance that a bill that is thus passed could claim broad indirect public support is therefore pretty good.

If a government cannot get good legislation through both houses, it has the option to force a double-dissolution.  If the government is much more popular than its opposition then it is likely to get a majority or nearly so - even these days - in a joint sitting after that election.  If it is not much more popular, and its measures are being opposed in the Senate by both the crossbench and the Opposition, then the claim that those measures deserve to be passed is pretty flimsy.

The Hodgepodge Of Joint Sittings

A joint sitting, in contrast to the two-house system as the dominant decision-maker, creates not two distinct mandates but one hodgepodge mandate.  The composition of a joint sitting is not consistent with one-vote-one-value, because Tasmania still sends nearly four times as many MPs per voter to a joint sitting as NSW.  But it also does not provide much scope for minority representation, since only one-third of the seats are PR-based.

It is frequently contentious (if overstatedly so) that a party can win a Lower House majority while losing the two-party preferred vote.  One of the good things about the current system is that such a winner is typically nowhere near gaining a Senate majority and therefore it is likely to have to play well with others to govern effectively.  Considering the 1998 federal election, the Howard government would have held 115 seats to 109 for all other parties combined in a joint sitting - this despite losing the 2PP by a substantial margin and polling below 40% of the primary vote in both houses.  That majority would have just been an artifact of sitting member personal vote effects left over from the 1996 win.  The advantage of the current system is that had 1998 been a double dissolution, the government would not have been able to carry its 1996 Senate result into the new chamber.  Indeed, the prospect of One Nation winning a bagful of seats in a 1998 DD would have been enough to scare off any idea of an undeserved Coalition joint-sitting majority.

Regular joint sittings per model 1 would effectively create a single-chamber parliament by stealth, albeit a clumsily designed one.  It would create a system in which ultimate power was eventually determined by an unprincipled mixture of elements stripped of the logic of the original system and there just because they were the bits we had before.

The whole effect of the Abbott proposal - whichever version it is - is to increase the ease with which governments can govern while making the mandate they require to govern easily more dubious.  It would be bad for our democracy.  Joint sittings should be safe, legal and rare.

It Aint Broke, Do Not Fix It

It is not as if the current environment makes passing legislation at a joint sitting impossible.  Even with a disappointing election performance and a barely scraped Lower House win, the Coalition would have only needed the support of 8/15 non-Greens crossbenchers to get bills through a joint sitting.  That's hardly unreasonable given that neither major party polled much to crow about in the 2016 poll.  A 52:48-ish result would have been good enough for another eight seats in the Reps alone.

The Coalition would have done much better had it not had such a mess of a first term.  But a mess it was, thanks in large part to, er, Tony Abbott!  It might also have done better at the DD that eventually occurred had it spent more of its first two years stashing double-dissolution triggers on a wide range of issues so that it would have a better story to tell at the 2016 poll, and also more chance of actually doing something with the joint sitting power at the end of it.  But somebody was mucking about with flags, shirtfronts, death cults and knights and dames when they should have been working on their party's future strategic preparedness.  Could that someone have been, um, Tony Abbott?

So what we have is Abbott creating a mess as PM and then blaming the system for the trouble that the rest of his party is having (with less than zero help from him) in cleaning it up.  It's a damaging piece of kite-flying by the former PM because it will make it look like the current Government is too weak to address its Senate problems.  It is yet more proof that the ex-PM is no team player and will never stop his meddling until the entire nation kneels around his feet convinced he was the best PM since Menzies.  Or at least til they give him a ministry.

Finally, even if this ever attracts enough support to seriously promote it, I think that this has Buckley's at a referendum.  Any government would be nuts to support it unless it thought that it would win the next election, and any opposition likewise.  But even with bipartisan support I expect it would cop massive heat from minor party voters and be greatly distrusted in the smaller-population states that stood to have their power watered down.  (A shame really, the latter is the only good bit).  You need to pass four states to win a referendum and Tasmania and South Australia wouldn't touch it with a barge-pole.  Good luck getting all the rest.


  1. Because it reduces the smaller states' power in a joint sitting, do you think the referendum would have to pass all states?

    1. The relevant provision in S128 is "No alteration diminishing the proportionate representation of any State in either House of the Parliament, or the minimum number of representatives of a State in the House of Representatives, or increasing, diminishing, or otherwise altering the limits of the State, or in any manner affecting the provisions of the Constitution in relation thereto, shall become law unless the majority of the electors voting in that State approve the proposed law."

      The proposed change doesn't do any of these things so no. Joint sittings would still have the same composition as they do now, except they would become fairly commonplace events. At present they're extremely rare, even after DDs most governments don't end up going to one.

  2. If Tony Abbott was capable of feeling embarrassment (which of course he is not) he would be embarrassed to refer to the 2003 "Discussion Paper" on resolving deadlocks. When the consultation group set up to hear public responses (Neil Brown, Michael Lavarch and Jack Richardson) brought its roadshow to Brisbane, I got the cheer of the night (from a crowd largely dominated by Aust Democrats, I must admit) by referring to it as the "Unsupported Assertions Paper". I'd like to think I killed the "reform" off single-handedly at that moment, but I had the impression that the consultation group had _already_ heard enough in other cities to realise that the "reform" would get no support from the voters and I was only reinforcing a conclusion that they had already come to. And now the Ab-bott wants to revive it - hah!

  3. Having joint sittings allowable after House-only elctions is a recipe for knocking the House and Senate terms out for (at least short-term) electoral and legislative advantage.

    Allowing joint sittings after combined House and half-Senate elections would either mean a 5th attempt at simutaneous elections or potential for opposite party state premiers, particularly conservative ones, to refuse to advise their states` governors to call their states` Senate elections at the same time as House elections.

    The High Court may decide that the proportionate representation in the Senate either includes or more likely is in relation thereto to the ability to hold joint sittings. A bet tenuous but possible. The Nexus Clause probably also requires majorities in all states because it is integral to the proportionate represenation of larger than average states in the House, as otherwise there would be a minimum House size of 5 seats for every state with states thus being equally represented.