Making sense of the polls

Making sense of the polls

Conservative landslide likely at 8 June UK general election

Last night, UK Prime Minister Theresa May announced that a general election would be held 8 June. Under the Fixed Term Parliament Act, Parliaments are supposed to run for five years, but a 2/3 majority of the House of Commons can vote for an early election. Labour is supporting the early election, and the Conservatives and Labour easily have more than 2/3 of the seats.

This election will be held just over two years after the Conservatives won a majority at the May 2015 election. No sane government would cut short its term by almost 60% unless they were very confident that victory was assured.

Current polls give the Conservatives every reason for confidence; they lead with 43%, to Labour’s dismal 26%, 11% for the UK Independence Party and 10% for the Liberal Democrats. Three of the most recent five polls have given the Conservatives 21-point leads, with one 9-point lead poll looking very much an outlier. Under the UK’s First Past the Post system, these polls would result in a Conservative landslide if replicated at a general election.

There are two key reasons for Labour’s woes: Jeremy Corbyn and confusion over Brexit. Since Corbyn became Labour leader in September 2015, I have thought that Labour was headed for a large loss at the next UK general election. Better PM polling has May leading Corbyn by 50-14, and some polls have shown May leading even with 2015 Labour voters.

The Brexit referendum in June 2016 has left Labour’s Brexit policy in a muddle that pleases neither side of that referendum. Labour is trying to support Brexit, which won the referendum, but with conditions. As a result, neither Leavers nor Remainers are attracted to Labour.

The only significant UK-wide party that explicitly opposes Brexit now is the Liberal Democrats. Local council by-elections have had large swings to the Lib Dems. If any party is likely to gain ground during the campaign, it could well be the Lib Dems, who will appeal to the 48% who voted Remain at the Brexit referendum.

Local council elections will be held on 4 May, and these should give an indication of what is likely to happen in the general election. UK polling has not been accurate in the past, but misses have underestimated the Conservatives.

French Presidential election: 23 April and 7 May

The first round of the French Presidential election will be held this Sunday. Polls close at 4am Monday Melbourne time. A runoff between the top two candidates will be held on 7 May.

There has been little change since last week in the standings of the top four candidates. Centrist Emmanuel Macron leads with 24%, followed by the far right Marine Le Pen on 23% and both the conservative Francois Fillon and hard left Jean-Luc Melenchon on 20%.

The differences between one poll and another are very low, and this implies that the French polls are being “herded”. Herding happens when pollsters suppress their raw data to conform to a trend. If the trend is right, herding is not a problem, but sometimes the trend pollsters conform to is wrong. If there is a major polling miss next Sunday, it will likely be due to herding.

If Macron makes the runoff, polls show him thrashing any of his potential challengers. He leads Le Pen by over 60-40, Fillon by 65-35 and Melenchon by 58-42. Current polls show Melenchon easily defeating Fillon and Le Pen, while Fillon defeats Le Pen.

By-election in US House seat goes to runoff

A by-election in Georgia’s sixth Congressional District (CD) was held today. This was a “jungle primary”, where many candidates from the two major parties run on the one ballot paper. Unless one candidate wins a vote majority, the top two candidates proceed to a runoff.

At this by-election, the lead Democrat, Jon Ossoff, won 48.1%, and the lead Republican, Karen Handel, won 19.8%. Virtually all remaining votes were for other Republican candidates. A runoff election on 20 June will be contested by Ossoff and Handel. Polls suggest this contest is close.

In 2016, this CD voted for Trump over Clinton by 48.3-46.8. However, this represented a large swing to Clinton from the 2012 Obama vs Romney contest, when this CD voted for Romney by 61-38. A Democratic win in a naturally Republican CD would be a boost for them.

Lucy Gichuhi to replace Bob Day in Senate; Essential at 54-46 to Labor

Two weeks ago, the High Court declared Bob Day invalidly elected, and ordered a recount of the SA Senate votes to decide his replacement. As expected, that recount elected Kenyan-born Lucy Gichuhi, the No. 2 on Family First’s ticket. Today, the High Court dismissed a last minute Labor attempt to challenge Gichuhi based on her citizenship. Gichuhi will be sworn in when the Senate resumes on 9 May.

Labor led by 54-46 in this week’s Essential, gaining a point from last fortnight. Primary votes were Labor 37%, Coalition 36%, Greens 10%, One Nation 8% and Nick Xenophon Team 3%.

41% approved of the US bombing of Syria, and 36% disapproved. 39% would disapprove of sending US troops into Syria, and 37% would approve. 50% would disapprove of Australian military support in Syria, and 31% would approve. 49% thought terrorism was the biggest threat to global security, with 15% for US aggression and 11% for climate change.

36% thought Barnaby Joyce’s suggestion that people who cannot afford houses in Sydney and Melbourne should consider moving to the country was a good idea, and 24% a bad idea. 50% thought superannuation should be preserved for retirement, while 38% thought people should be allowed to use it to buy homes.

Approval ratings of the crossbench Senators were assessed. Nick Xenophon had a net approval of +10, followed by Derryn Hinch at +8, Jacqui Lambie at +2, Pauline Hanson at -16, David Leyonhjelm at -19 and Cory Bernardi at -24. Hanson’s net approval is much better at +27 among Other voters

In last week’s Essential, Turnbull’s net approval was -12, up five points since March. Shorten’s net approval was -13, up six points.

50% disapproved of the $24 billion in tax cuts to businesses with annual turnovers up to $50,000, and 31% approved. Among Other voters, whose Senators voted for the tax cuts, disapproval was 52-34. 60% would disapprove of a similar cut for businesses with annual turnover over $50,000, with just 20% approving. 57% thought the more likely result of the tax cuts was that businesses would make bigger profits, while 26% thought businesses would employ more workers.

Respondents were generally pessimistic about whether various items were getting better or worse for them, but the most pessimistic assessments were on housing affordability (9% better, 62% worse), cost of electricity and gas (6-74) and quality of political representation (7-60).