Sunday, April 5, 2020

Morrison Breaks Two Newspoll Records Amid Coronavirus Crisis

The Newspoll just released deserves a special post in the absence of other polling, because of a couple of historic bounces for incumbent Prime Minister Scott Morrison.  It should be stated from the outset that these records have fallen partly because Morrison was coming off a low base.

To give the numbers first, the government now leads 51-49 two-party preferred, a gain of two points from three weeks ago.  The primaries are Coalition 42 (+2) Labor 34 (-2) Greens 13 (+1) One Nation 5 (+1) Others 6 (-2).  (I have a concern that the new Newspoll methods may be overestimating the Greens' vote by naming only them and the majors on the initial screen).  Scott Morrison leads Anthony Albanese 53-29 as "better Prime Minister", up from 42-38 last time, noting that Better PM is an indicator that tends to skew to incumbent PMs all else being equal (so 42-38 was actually a bad result for Morrison).  Morrison has a net satisfaction rating of +26 (61 satisfied 35 dissatisfied), up 38 points from -12 (41-53) last time.  Albanese has a net satisfaction rating of +9 (45-36), up 9 points.

What is notable overall here is that the government has only registered a modest bounce on voting intention but perceptions of Morrison's leadership have been changed dramatically by the crisis. This is indicative of a bipartisan mood where many voters are willing to say that although they support an opposition party, the government leader is doing a good job with this crisis.  (One can hear similar from Victorian Liberal voters regarding Daniel Andrews.)



Netsats

It is better to look at net satisfaction ratings when comparing the current Newspoll with historic results.  The changes made by YouGov late last year have dramatically reduced the "undecided" factor on the question of satisfaction with the PM's performance.  As a result, the fact that Morrison's 61% satisfaction beats Malcolm Turnbull's best of 60% and is one point behind Bob Hawke's best of 62% (Newspoll missed Hawke's stratospheric first few years) is pretty meaningless.  Morrison has 61% satisfied with 35% dissatisfied whereas Turnbull had 60-22 and Hawke had 62-32.  For the record, the last satisfaction rating higher than 61% was a 63 for Kevin Rudd in October 2009, but Rudd had a netsat of +35.  Morrison's netsat rating of +26 is still not exceptionally high.

However, the bounce from Morrison's previous netsat of -12 to his current +26 smashes the all time record for such bounces.  Here's the leaderboard and the causes of each bounce:

+38 Morrison (Coalition), 2-5 April 2020 (COVID-19)
+24 Howard (Coalition), 10-12 May 1996 (Port Arthur massacre)
+20 Howard (Coalition), 31 Aug-2 Sep 2001 (MV Tampa incident)
+19 Hawke (Labor), Feb 1991 (First Gulf War)
+18 Howard (Coalition), 21-3 Sep 2001 (September 11 attack in New York)
+18 Howard (Coalition), 18-20 October 2002 (Bali bombing)
+17 Howard (Coalition), 22-4 Sep 2000 (Sydney Olympics)

It can be seen that the biggest netsat gains by PMs have been driven by events.  The two terrorism-related events (9/11 and Bali) drove large 2PP voting intention gains for the Howard government, of six and five points respectively, but the others listed corresponded to smaller gains of 1-3 points.  So an incident that gives the PM a big boost does not always flow on in large degree to their party.

Morrison was not alone in coming off a low base in the above list - Hawke was coming off -31, and Howard before Tampa was coming off -10.  In the context of results for leaders elsewhere, it would have been disappointing for the government if Morrison did not break this record.

Better PM

Morrison's 20-point improvement in his Better PM margin from 4 points to 24 points is also an all-time record for a PM against the same opponent.  In this case the leaderboard looks like this:

+20 Morrison (Coalition) vs Albanese, 2-5 April 2020 (COVID-19)
+18 Keating (Labor) vs Downer, 5-7 Aug 1994 (Downer leadership implosion)
+17 Howard (Coalition) vs Beazley, 10-12 Mar 2006 (ALP infighting)
+15 Keating (Labor) vs Hewson, 19-21 Mar 1993 (election victory)
+15 Howard (Coalition) vs Beazley, 16-18 Oct 1998 (election victory)
+15 Howard (Coalition) vs Latham, 19-21 Nov 2004 (following election victory)
+15 Rudd (Labor) vs Turnbull, 26-8 June 2009 (Ozcar affair)

Previously, this record has been driven by a combination of election wins and Opposition debacles, and not so much by extreme events.  Again here, Morrison's result is only extraordinary in terms of poll to poll gain.  His 24-point lead, while good, is unexceptional - Turnbull led Bill Shorten by as much as 49 points in November 2015.

What does it mean?

As noted, these figures do not indicate that Morrison is stratospherically popular just yet, but they do show he has turned around perceptions of his performance very quickly, with his tardy response to the bushfires now a distant memory for most.

Nothing should be read into these numbers in terms of the next election, other than that people who a few months ago thought Morrison was doomed by the bushfires and some Coalition scandal that even I can no longer clearly remember off the top of my head should now realise their view was extremely premature.  It would have been so even had none of this happened.  Those won't be major issues when (at least I hope it's when) Australia next votes.  Even when major issues do appear much closer to an election, as was the case in 2001, event bounces can very quickly disappear once a poll is called.  And war-like situations are no guarantee incumbent governments anywhere in the world will be re-elected - just ask Robert Menzies who lost his majority early in WW2, or Winston Churchill who was voted out at the end of it.

In terms of reactions to Morrison's bounces, there will be three that should be dismissed.  The first is that some people will say the poll is rigged.  There is no reason for such partisan stupidity; polls worldwide have been showing large bounces for other leaders (except for Donald Trump who has so far only had a small bounce.)  The federal government's performance has come in for some criticism from political opponents on social media but that does not mean attempts to pin, say, the Ruby Princess debacle on it have yet resonated with the wider public.  Nor have the more alarmist projections of the rate of spread of the virus that informed (or misinformed) calls for harsher and faster lockdowns covered themselves with glory.  It is not the same as the situation with Trump where his failures are already being used in partisan advertising against him.

The second is that some people will say the bounce is simply because of the event, and that people blindly rally to the leader in a crisis, that this is all about the virus and not the PM's handling of the matter.  I think that is actually quite wrong too.  Many voters might be responding like that but the icing on that cake must be that Australia is currently doing very well at keeping both the growth of the virus (given our high testing rates) and the death rate down, compared to other countries of similar size and especially similar political culture.

To what extent this is down to the effectiveness of Australia's responses, and to what extent to whatever else might be limiting the virus's spread in Australia - and hence whether Morrison has or has not earned these results - is something I leave to the experts.  But the public perception will be that Australia is responding well at this stage and that this is something to be proud of.  Indeed the backlash against Morrison's response to the bushfires earlier this year has already showed us that voters do not always reward the leader in a crisis, so it hardly makes sense now to suddenly assert that they do.

The third dismissable response is that Newspoll failed at the last election so whatever it finds should be ignored.  That it did, but only modestly in the context of routine polling failures overseas, and it has since brought in new methods that, whether they work or not, are at least targeted at likely causes of the 2019 failure.  Moreover, poll to poll changes are not affected by any systematic effects in a poll (if there still are any).  And the changes seen in Morrison's ratings are so large that a few points of voting intention error would not explain them away.  Furthermore, if someone thinks Newspoll is still failing in the same way as in 2019, then the result of that would be that Morrison is now more popular than it shows him to be.  However, those using the 2019 stick to beat current Newspolls are almost always opposed to the Government.

Finally, Morrison's large Better Prime Minister gain is no reflection on Anthony Albanese, whose own net rating is up nine points.  It just means that at a time like this Opposition Leaders struggle for relevance unless governments perform very badly.  If they fall in behind the government they are superfluous and if they criticise it they are seen as undermining solidarity in a crisis.  Indeed as the graph below shows, Albanese is actually doing well to not be further behind as Better PM given the PM's high net satisfaction rating - the current poll (red dot) is at the low side of Better PM leads in that circumstance.


I add further that of eleven cases where an incumbent PM gained 15 or more points in a single poll, Albanese's 9 points is the equal largest gain.  It is the equal second largest among 41 cases where an incumbent PM gained 10 or more points.

I'll be following future Newspolls that show dramatic changes (if any) - after all it is not like there is much else to do on a pseph website at the moment!

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.