Everyone assumes the election will be in May. The Constitution says that senators elected in 2016 were deemed to have been elected on 1 July 2016 and that they must face re-election sometime in the 12 months before 30 June 2019. Given all sorts of electoral requirements and unacceptable dates that means on or before 18 May.

Everyone assumes the House of Representatives election will be held at the same time. That may be the practical thing to do, but the Constitution does not demand it. The Constitutional requirement for the House of Representatives is that it cannot last after three years from the first sitting after the previous election. It must be dissolved. If you then add time for election writs to be issued and the campaign it means that the election can be postposed until before or on 2 November, nearly six months after the last possible date for the half-Senate election.

We have recently heard the intriguing metaphor that Prime Minister Scott Morrison needs more “runway”. He needs more time and runway length to get the momentum to enable takeoff.

So maybe the runway can be extended to November.

Morrison is in an unusual position. Unlike most inmates on death row, he gets to choose his execution date, at least up to a point. So why wouldn’t he postpone that for as long as possible in the hope that some Micawberian event turns up to give him a reprieve?

Morrison is as about delusional as Micawber. He seems to blame the Wentworth result almost exclusively on a Malcolm factor: that Wentworth voters were angry that their favourite son was bundled out of the prime ministership. But that can only be part of the story – only 12 percentage points of the story which represents the 12 percentage points of vote increase that Turnbull won from the time he won the seat with a 5 percent margin to the 2016 win on a 17 percent margin.

That leaves nearly 8 percentage points of swing unexplained by the Turnbull factor and only explained by the poisonous policy promotion being forced by the far right of the party. That 8-percentage-point swing (Turnbull and leadership change aside) is enough to send Morrison into oblivion, even if he clings, McMahon-like, to every possible day in office.

Indeed, the more runway he is given and the more momentum he gains, the greater the crash will be.

As to minority government, expect it more often. As I wrote a month before Wentworth, the minor party and independent vote is reaching critical mass and will sound in enough seats to make majority government the exception rather than the rule.

At present the 20 to 30 per cent of vote for the independents and minors has resulted (after Wentworth) in just 4.5 per cent of the House of Representative seats. But once elected, independents usually get re-elected.

Expect more Wentworths. Perhaps in Warringah. Perhaps as many as 10 per cent of the seats for the 20 per cent of the vote. Expect, therefore, that a hung Parliament is a very likely result in 2019 despite Labor thinking it has it in the bag.

And now to the Senate stitch-up. The stich up went as follows. After a double dissolution the Senate decides which of the 36 senators get six-year terms and which get only three-year terms.

There are two ways of doing that. The first is the fair and democratic way which is the count back method. That used is in electoral legislation and also determines who takes the seat if a senator is declared ineligible. It reflects what the voters wanted. The second way is to give the long terms according to the order in which senators are elected. The second method favours major parties – those that get more than two quotas (about 29 per cent of the vote).

So the major parties ganged up to give themselves the lion’s share of long-term seats by using the latter method.

Of 52 major-party senators elected in 2016, only 23 got short terms and face election before 30 June 2019. Whereas 13 of the 20 minor-party candidates got short terms and face re-election.

In past half-senate elections Labor had a sporting chance of getting three of six seats in Victoria and NSW and the Coalition had a sporting chance of getting three of six in WA and Queensland.

The practical effect of the stitch-up in normal times would have seen up to five of the 13 cross-bench senators lose their seats.

But these are not ordinary times.

For either major party to win three of the six Senate seats in a state it would need 43 per cent of the vote after preferences. Given that neither is getting many preferences, they need almost that much in primary vote. That is not happening in any state, according to most polls.

It means the minors and independents will win two seats in each state, resulting in 12 of them replacing the existing 13, not much of a loss and a complete failure of the 2016 major-party stitch up.

Labor will pick up the seat lost because it has only one senator up for re-election in both NSW and South Australia and will almost certainly get two in both those states. But it has three up in Tasmania and can only get two.

Its existing third seat will most likely go to Jackie Lambie or similar.

So the Senate will remain much as it is now. But come 2022, the Coaltion is in for a bath: 16 of its senators come up for election and at best it can expect only 12 of them to win.

The hoisting on the petard of the 2016 stitch up will take six years to happen. But happen it will. Unless, of course, the major parties try to avoid half-Senate elections every three years because in the long run it would likely to give each of them only four senators out of 12 in each state. Instead, they would go for double dissolutions whenever possible because in a double dissolution each major party can reasonably expect five out of 12 seats in each state.

The possibility of Morrison holding on until November and just a half-Senate election in May is a real mess and begs for constitutional change for fixed terms and simultaneous elections.

www.crispinhull.com.au