Poll Reporting Unfair To Hockey

A few knives are out for Treasurer Joe Hockey, who is repeatedly rated one of the government's poorest performers, and whose political fate seems very much tied to a Budget that received the second-worst ratings in the past 29 years. Hockey this week announced revised forecasts with deficits over the next four years nearly two and a half times what was expected little over half a year ago.  Not just his economic competence and consistency (given that in Opposition he railed against Labor's supposed debt addiction but seems powerless to escape it himself any time soon) but also perceptions of the Treasurer as a surly communicator are coming into the spotlight.  Admittedly, he has had a difficult time with many savings measures being blocked by the random rabble of crossbench Senators, but most Senates of times past would have done the same thing.

It may seem difficult to argue then that Hockey deserves much sympathy at all, or that any pain he experiences in the role is anything but self-inflicted.  But today I've got to do it.  This report in the Australian by Troy Bramston declares Hockey is regarded by the public as "the worst Treasurer of the last 40 years".  The poll on which the article is based shows absolutely no such finding.  The sensationalised and fallacious reporting of the poll is quite surprising - Bramston, unlike some at said paper, is far from a serial offender for this sort of thing. It is in no way the pollster's fault - rather it is just a case where a journalist uses a poll to say something it doesn't actually say. 



The poll, by Essential Research (see link above for tables), asked voters who they would rate as Australia's best treasurer of the past 40 years.  Options provided were Peter Costello, Paul Keating, John Howard, Wayne Swan, Joe Hockey and Don't Know.  Costello romped in with 30%, with Don't Know second with 28 (high undecided rates are a hallmark of Essential issues polling*), and then came Keating with 18, Howard 12, Swan 8 and last was Hockey with 4.  There were the usual partisan effects: 78% of Coalition supporters went for a Liberal treasurer and 11% for a Labor one, while Labor supporters split 48-23 for one of their own.  (Greens supporters were 39-21 for Labor treasurers and Others supporters were 50-18 for Liberal ones).  A similar Essential poll in August had figures of 30 for Costello, 23 for Keating, 8 for Swan and 5 for Hockey, with Howard not included.

(* I had some feedback on this, so to clarify: I believe the online format - which requires a pollster to offer "don't know" on the same level as other choices or not at all -  leads to higher don't-know rates compared to live interviewing. Live interviewing has the option of not offering "don't know" as an answer, but accepting it if the respondent is clearly undecided.  Tentative respondents who will offer views in phone polling are more likely to say "don't know" to a computer.  I suspect it is an issue endemic to the online method rather than necessarily a particular pollster using it.) 

Poll does not support conclusions drawn

The first problem in drawing conclusions about who is considered the worst treasurer is that the poll did not ask that.  It asked who voters considered the best.   You cannot conclude from a treasurer being rarely rated the best that he is commonly considered the worst - it might just be that he is generally considered mediocre or that there are few strong opinions on him. 

In Hockey's case this isn't terribly likely, and in a poll that asked voters to pick the worst he might well score very highly indeed.  I'd expect voters to show both a strong partisan bias and a strong recency bias on such a question, so that Hockey and Swan might be the top scorers.  I'd be surprised if Hockey didn't top such a poll and poll at least 30% in so doing.  However, that was not the question that was asked.

Imagine if voters were instead asked who they thought was the best performed member of the current Cabinet.  Probably someone harmless, low-profile and competent would score the lowest (and score rather lower than Hockey), but would that make them the "worst" member of Cabinet?

The second major problem with this poll - as an indicator of who is the worst Treasurer - is that it does not include the full set of Treasurers of the past 40 years. Not even close! In the last 40 years, all the following have served as Treasurer of Australia: Jim Cairns, Bill Hayden, Phillip Lynch, John Howard, Paul Keating, Bob Hawke (for 1 day), John Kerin, Ralph Willis (twice), John Dawkins, Peter Costello, Wayne Swan, Chris Bowen, Joe Hockey.  Thirteen in all (the tenure of Frank Crean ended 40 years and two days before the poll started being taken, so I shouldn't strictly count him.)

Of these, it makes sense to exclude Hawke, who held the post for a day as a pure formality, and probably also Hayden and Bowen, whose governments were defeated before they could do all that much.  But the remaining five Treasurers excluded from the poll (Cairns, Kerin, Lynch, Willis and Dawkins) are a different story.  Dawkins, for instance, was the Treasurer who brought down the classic "horror budget" of 1993, which received a worse poll reception than even this year's.  (See graph.)

Kerin's stay as treasurer lasted only six months (mainly because he wasn't up to it), Lynch had to resign following a tax-minimisation conflict of interest, and Cairns' tenure covered the turbulent period of the Khemlani and George Harris loans scandals.  Ralph Willis may not have met such a colourful end as these, but he was treasurer for over two years, and much more recently so than Howard.

In any reasonable assessment of the worst Treasurers of Australia, a political historian would look especially closely at those whose tenures had ended in quick and self-inflicted disgrace.  Yet they're excluded from this Essential poll, precisely because the public would have forgotten about them, and that in turn is often precisely because they were so bad. 

So even if we grant that Hockey has been shown by the poll to be a worse Treasurer than Costello, Howard, Keating and Swan (which he hasn't), there are still eight other Treasurers to whom the poll has not even compared him.  Those to whom he's being compared were the long-term survivors, who were in the role for almost 31 years between the four of them.  No matter how good Hockey was (assuming he actually was good) he would struggle to scrub up after just a year and a bit.

The third problem with this poll is that some of its respondents are confused.  John Howard rates fairly well in the poll, but his reign as Treasurer ended with the economy in a parlous state with double-digit unemployment and inflation and high taxes.  However, this was much more a result of Malcolm Fraser's economic views than his own - Howard claims he considered resigning in 1982.

Whatever the merits of Howard's treasury, the strangest result of this poll is that 13% of respondents aged 18-34 rated Howard the best Treasurer.  The very oldest respondents in that age bracket would have been three and a bit years old at the end of Howard's tenure, and most of them wouldn't have been alive.  Very few of them would have made a detailed study of Australian political history - it's clear that many of those picking Howard as the best Treasurer are doing so not on account of their memories of his performance as Treasurer, but on account of his years as Prime Minister. (I also suspect some rosy nostalgia for Howard-as-PM among the 16% of 55+s who opted for him.)

Hockey vs Bowen

The other - and more useful - finding of this poll is a matchup between Joe Hockey and Chris Bowen on the question of who voters would trust to handle the economy.  It shows Hockey trailing 27-29 compared with leading 34-23 in August and 33-27 in May.  (Note that the May sample was pre-Budget.)  The don't-know rate is very high, so it's hard to read a great deal into it, but at best it's severely unflattering.  That said, in the context of the government's current polling position, it's also more or less to be expected, and doesn't show Hockey to be any deeper in the mire than Tony Abbott.  A Minister would have to be doing a rather good job to have any kind of lead over their opposite number while their Government is trailing as heavily as this one.