Four Year Terms? Fixed Terms? Fixed Four Year Terms? Part One - Polling

This week there was an article regarding a report released by a lobby group called 4 Year Terms Australia, which supports "fixed" four year terms for federal parliament.  The SMH report contained this very brief comment regarding polling:

"Polling commissioned from JWS Research found just one in five Australians supported the current system of three-year terms and the prime minister choosing the election date."

Well, from that it may sound like Australians would strongly support "fixed" four year terms, right?  In fact the polling cited tells a rather different story.

I've decided to write one article on this subject dealing mainly with the polling, to be probably (subject to time) followed by another in the near future dealing with the (de)merits of the issue.  That said there will be few surprises on the latter front to anyone who's seen my article re fixed terms in Tasmania or my 2025 JSCEM submission.   To summarise my views, I'm totally against "fixed" federal terms, whether by legislation or referendum.  I think that fixed term laws either fail to genuinely secure fixed dates or else prevent Prime Ministers from calling elections in circumstances where they should be able to.  I'm also against four year federal terms if it means eight year Senators.   

I think it's worth considering the option that was smashed 32.9-67.1 at the 1988 referendum (unfixed four year terms with simultaneous all-in-all-out Senate elections) but that even that proposal has some balance, stability and manageability risks that are not immediately apparent to people who think it is a good idea.  Does our current system have some drawbacks?  Yes - though some of those are being overstated by those pushing for fixed and/or four year terms.  Is it so bad that we should rush into alternatives with serious problems because somebody thinks it will save money?  Hell no!  

I'm concerned that after decades without regular referendums, lobby groups in the 21st century seem not to have absorbed the lessons of the past as to how hard they are to win and we seem to be beset with a plague of bad inner-city-focused referendum campaigns that just have no idea.  "Yes 23", the current Australian Republican Movement (many would say they were already hopeless back in 1999) and now this lot.  The report the group has put out is riddled with bad arguments and errors that I expect to deal with in the sequel.

For now my focus is on the polling - which I'm confining to the last ten years (and will add others if I become aware of them).  What do the polls on this issue say, and how do they stack up compared to the claims being made for them, and do they offer hope for one or more successful referendums?  I will start with the polls mentioned by 4YTA.

* JWS (2026).  "Support for a fixed election date is consistently strong, and above that of four-year terms.Research by JWS Research for this report found 61 per cent supported fixed terms, compared to 22 per cent for the status quo"  This isn't surprising given that voters aren't familiar yet with the nuances of the fixed terms debate and don't necessarily understand that fixing parliamentary terms is easier said than done and that no fixed term system will guarantee the date of any future election.  But what is interesting is the breakdown:

That's 35% for fixed three year terms, 30% for fixed four years, 11% for unfixed three years, 11% for unfixed four years and 13% unsure/no preference.  So in this poll three years is more popular than four! (I add also that JWS was once a major seat pollster but in recent years its accuracy, frequency and relevance as a voting intention pollster have declined.)

There's also this "31 per cent would prefer a four-year term if it was paired with a long weekend for the election. This lifts support for four-year terms to 59 per cent".  Well of course it does.  Everybody wants a pony.  At this stage, no question wordings from any of this are available.  

* Newspoll (March 2024). "In 2024 Newspoll found 51 per cent supported a fixed, four-year term with 37 per cent opposed."  The question in the Newspoll was "On balance, and while you may have no strong feelings either way, would you be more inclined to say you are in favour of or against changing the term of Australia’s House of Representatives from the current three-year terms to four-year fixed terms?" The responses were 17% Strongly in favour, 34% Partly in favour, 21% Partly against, 16% Strongly against, 12% Don't know.  With two-thirds of the respondents who are "in favour" picking the very ambiguous option "partly in favour", this cannot be treated as unqualified support.  Personally I am not \a fan of potential yeah-nah type response options like "partly in favour" and "somewhat support", especially not for a question that could go to a referendum where a high share of "partly in favour"s are likely to end up as "no". I would suspect in this case a lot of the partly-in-favours liked the fixed bit but weren't so keen on the four years.

* Australian Election Study (2025).  "In 2025 it found that 42 per cent of voters prefer four-year parliamentary terms, compared with 30 per cent for three-year terms (28 per cent no preference)." The question was "The House of Representatives currently has a maximum term of three years. Some people think four-year parliaments are better to give governments more time to govern. Other people think three-year parliaments are better to hold governments accountable."  In this case the question and answer designs are fine, but the issue is the timing.  AES is taken in the wake of elections at a time when voters are likely to be suffering from election fatigue and to be more receptive to the idea of letting the government get on with it.  This is especially the case when the government is enjoying a honeymoon period, and the Albanese government had an unusually long second honeymoon.

* Resolve (2024). "A 2024 Resolve Strategic poll found 42 per cent support for moving to a fixed four-year term, compared to 23 per cent opposed."  This is cited as "unpublished research" so it is impossible for me to scrutinise the poll's wording, other than to note that Resolve issue polls published by Nine media frequently have wordings that I disagree with, often because the wordings are too complicated and likely to lead respondents astray.

* McKinnon (2025) "In 2025 McKinnon found 43 per cent support a four-year term compared to 28 per cent who oppose it. 57 per cent believed a four-year term would help government make decisions for the long term." This is again cited as "unpublished research" making it impossible to scrutinise any aspect of the poll (?) including the question wordings and answer choices.  It's also not clear what platform it was conducted through as McKinnon are not themselves regular pollsters but tend to conduct their polling through other pollsters, recently including JWS and Roy Morgan.  However given that there is no mention of this polling in McKinnon's own March 2025 report on the same issue, and given that there were two reported waves of McKinnon polling in 2025 in the months after the election, I suspect this has the same timing issue as Australian Election Study. Furthermore, McKinnon are not a source external to the four year terms push but are themselves a supporter of it, so this is basically no more than movement in-gossip without further evidence.  

* JWS (apparently March 2017 but reported in July 2017) Details of this poll are elusive; what I have found is that 22% supported fixed four-year terms, 20% supported unfixed three year terms, and 16% had no preference, leaving a 42% black hole because poll reporting in this country was as hopeless then as now (the culprit was of course the Herald-Sun).  I cannot find any reporting of what proportion preferred four year terms overall but when asked if eight year Senate terms were a drawback, 10% apparently said change to a longer term (did they even read the question?), 69% still said four year terms were OK, 15% now said change to a shorter term, 2% didn't care and 4% didn't know.

Now for some polls not mentioned by 4YTA:

* Freshwater (April 2024) This one is one of the easiest to find so I suspect the reason it didn't get a mention is it doesn't support 4YTA's case.  When asked "Should the electoral term for the federal House of Representatives be extended from three years to four years?” 38% said Yes, 44% No and 18% were unsure.  This is only Freshwater, a poll that skewed to the Coalition at the 2025 election and was apparently doing something incorrect with Voice No support (but nobody seems to know what), but interestingly Greens voters were less supportive than Coalition supporters. 

* Essential (Aug 2017) Respondents were asked "In Australia, Federal elections are currently held every 3 years, although the Government can decide to hold an election earlier. Would you support or oppose changing to 4-year fixed terms so that elections are held every 4 years on the same date?"

Responses were Strongly support 18% Support 40% Oppose 15% Strongly Oppose 9%.  The question however is fool's gold since no fixed term system guarantees elections are held on the same date (or even, more accurately, the same weekend!) Also federal terms do sometimes run slightly longer than three years.  

What should we conclude from all this?

Polling on this issue is sparse and lacks consistency in question wording between polls.  Overall those polls that offer respondents a simple choice of fixed four year terms vs unfixed three year terms tend to find that respondents prefer the former.  But considering polls that offer options such as unfixed four vs unfixed three, or fixed four vs fixed three, it's clear that the superficial allure of fixed terms is doing the pulling here.  Respondents support fixed terms in isolation.  They probably don't support four year terms in isolation.  

A second thing about this is that the length and flexibility of parliamentary terms probably isn't something most voters think about often.  So the polls don't really capture the ongoing views of voters on the issue at any one time; rather they tend more to capture how voters respond when they are asked to think about these issues and give a first reaction.  And the third thing is that as we have seen before in referendum campaigns - most recently with the Voice - something that sounds nice in principle can start with high approval but the Yes vote can quickly decline as the issue is debated and complaints about the proposal build.  

Among the many drawbacks of the latest 4YTA report is they are confused about what their model for change actually is, in one paragraph talking about legislating fixed terms first then holding a referendum afterwards so that voters can see that one change has worked before they move on to the second.  That is optimistic since it would take at least a few election cycles for any sensible person to draw a conclusion about whether introducing fixed terms by legislation had "worked", especially given we have had few early elections recently.  Then in the next paragraph they are talking about having referendums on both fixed terms and four year terms and talking about whether they should be sequential or simultaneous!  This is all a major issue for a movement for fixed four year terms because if fixed terms were legislated then it looks like an uphill battle to get voters to accept four year terms.  On the other hand if both are taken to a referendum at once there is more to attack.

There is also a lot of discussion about bipartisanship as a road to success and trying to smuggle the referendum into the election season to maximise chances of success.  Hello, we are in an environment where the major parties are now frequently getting below 50% of the vote combined in national polls and we are not even sure if we actually have a two-party system anymore.  The way things are at present if a referendum was supported by the current major parties but opposed by One Nation I would not like its prospects of success, especially if it was on a theme that allowed One Nation to attack it as the major parties voting more priveleges and less scrutiny for themselves.  (On the other hand it is always possible that if One Nation feared losing support over an election cycle they too might get on board).  

None of this suggests that a referendum on four year terms (fixed or otherwise) has much hope of surviving the dynamics we have seen in other referendum campaigns.  "Politicians are bad at what they do and the answer is to give them more time in office" will always be a challenging sell.