Tomorrow’s Sun has some YouGov polling around David Cameron’s first hundred(ish) days in power. Topline voting intention is CON 41%, LAB 37%, LDEM 15%, a four percent lead for the Conservatives. Over the last week or so YouGov seem to have started to show leads of around 4 or 5 points for the Tories, compared to 5 or 6 a couple of weeks ago. Ever so slowly the polls are narrowing.

Before looking at the 100 days questions, the fortnightly tracker on the alternative vote referendum is worth noting – 37% would vote yes, 38% would vote no. This question has bounced about a bit from poll to poll, so I wouldn’t be surprised if we got a plurality for yes again in a fortnight’s time, but nevertheless, it’s the first poll to show the NO camp in the lead and suggests there is a downwards trend underneath the noise. Labour voters, who polls initially showed supporting AV, in this poll back FPTP by 46% to 34%.

Moving on to the additional questions, looking back at the first 100 days Cameron (+20), Clegg (+5) and Osborne (+7) all enjoy good approval ratings; all significantly more positive than that of the government as a whole, today’s net approval there is once again +1.

The majority of the public have confidence in the government’s ability to run the economy (55%) and there is widespread confidence in their ability to cut the deficit (62%). However, the public are generally pessimistic about their ability to deliver in other areas – less than half have confidence in their ability to improve the NHS (32%)and schools (32%), cut crime (33%) or immigration (39%) and to protect the country against terrorism (42%).

Asked about some of the specific policies the government have introduced or announced during their time in office, the most popular were aiming to withdraw troops from Afghanistan by 2015 (80%), limiting housing benefit (72%), cutting the number of MPs (77%) and protecting NHS spending (79%). The least popular were the increase in VAT (supported by only 22%) and the free schools policy (supported by 37%).

Moving onto attitudes to the coalition, 44% think that the Conservatives got the better deal, 29% that the Lib Dems did (amusingly enough, Conservatives are more likely to think that Lib Dems got the better deal, Lib Dem supporters that the Conservatives did). 34% of people think that the government would have been worse had the Conservatives been in power alone, compared to 23% who think it would have been better (including only 55% of Conservative supporters, 31% of Tories think it would have made no differece and 8% think it would have been worse).

Opinions on the Lib Dems are rather mixed. 55% think they have moderated the government and made it more centrist, and 60% think going into the coalition was the responsible thing for the Lib Dems to do given the economic crisis. However, 59% think it involved selling out their principles, and 62% agree with the statement that it is no longer clear what the Liberal Democrats stand for (including 38% of their own supporters).

Finally, the public still don’t expect the coalition to last the distance. Only 14% expect it to last more than four years, 47% expect it to last less than two years. Asked what they expect the result of the next election to be, 40% expect the Conservatives to retain power (27% outright and 13% in a coalition), compared to 28% who think Labour will win (21% outright, and 7% in coalition).

There was a Harris poll in the Daily Mail yesterday. Voting intention figures are incomprehensible. To quote from the Mail “The Tories are down from 36 per cent at the election to just 29 per cent, while Lib Dem support has collapsed from 23 per cent to just 12. Labour are on 28 per cent, down from 29 at the election. But a huge 17 per cent of people said they were undecided.”

Where to start? Firstly the Harris poll was of Great Britain, not the United Kingdom, so they should be comparing it to the GB result (CON 37%, LAB 30%, LDEM 24%). Secondly, 17% for don’t know isn’t huge, it’s comparable to other polls. Most importantly, they haven’t repercentaged to exclude don’t knows, so obviously all the parties are down. It is unclear whether or not they have also excluded won’t votes, so it’s not even possible to repercentage yourself. If they have excluded won’t votes, it implies 17% support for other parties, which seems unfeasible (though the newer online pollsters did tend to produce some very high scores for others before the election). If they didn’t exclude won’t votes either then it implies shares somewhere in the region of CON 36%, LAB 35%, LDEM 15%, but we can’t be sure.

On other questions, Harris found 26% thought Cameron had done better than they expected, 22% worse. 42% said he had been in line with expectations, though obviously we don’t know if those people’s expectations were positive or negative! For Osborne 12% thought he had exceeded expectations, 20% that he had done worse, Clegg was 19% and 29% respectively and Cable 13% better and 21% worse.

On opinions of the government, Harris asked people which words they’d use to describe it, with particularly unenlightning answers! All the words quoted in the paper were agreed with by about 52%-59% of people, included positive and negative ones – so 59% thought they were honest, 52% effective, 59% united… but 57% thought they are disappointing, 59% unpopular and 52% unbalanced. At least, I suppose there is an answer that everyone liked.

The only poll I’m aware of from the Sunday Papers is YouGov’s regular tracker survey – voting intention stands at CON 42%, LAB 37%, LDEM 13%.

UPDATE: Tabs for the Harris poll are here now. Had the poll been repercentaged it would have shown voting intentions of CON 38%, LAB 36%, LDEM 16%, Others 10%.


600 seats again

On Newsnight yesterday they had a price about what the political impact might be of the reduction in seat numbers to 600, based on a projection by Simon Wilks-Heeg and Stephen Crone. Mark Pack has kindly put Wilks-Heeg and Crone’s actual paper up on scribd here.

The paper is based on the electorate from December 2009, rather than the electorate at the General Election in 2010. In actual fact, the boundary review will be based upon the electorate in December 2010, but for obvious reasons those figures don’t exist yet. Wilks-Heeg and Crone have worked out the number of seats each country will get, and what that should mean for each region, and then made some educated guesses about how that might pan out in terms of partisan effect. Their estimates are that the Conservatives would lose 13 seats, Labour 25 and the Liberal Democrats 7 – meaning the Conservatives end up gaining 12 seats relative to Labour.

At this stage it is very difficult to make firm predictions about what the partisan effect will be from boundary changes. We don’t know the final electorate figures, nor what county boundaries the Boundary Commissions will choose to cross, nor indeed whether there will be any more changes to the proposed legislation as it goes through the Commons (campaigns like OneWight may, for example, manage to get extra exemptions). Even once those things are decided, there will be various different ways that boundaries could pan out. You can work out complex projections of what might happen, as Lewis Baston of the ERS did for Scotland and Wales a few months back, but they can only project one possible solution out of many alternatives. At this stage, Wilks-Heeg and Crone’s projections are probably as good as it is worth doing. For what it’s worth, my “back of an envelope” calculations are much the same as Wilks-Heeg and Crone’s – based on a crude look at where seats are likely to go and what knock on effects might be, I’d expect the Conservatives to lose somewhere around 10 to 12 less than Labour.

Government Approval

YouGov’s daily voting intention figures today are CON 41%, LAB 37%, LDEM 15%. Government approval is at 41%, disapproval at 40%, giving a net value of just +1, the lowest it’s been since the general election. There is a slow downwards trend in the net score, so I’d expect to see a negative rating sooner rather than later.

To look a bit closer at the approval ratings, below is a graph of approval ratings since the election. The grey line is the net figure for all voters, the blue amongst Tory voters, red Labour voters and gold Lib Dem voters. You can see approval amongst Tories is (unsurprisingly) high and steady. Approval amongst Lib Dem voters is less strong (it averages around 53% approval and 18% disapproval) and underneath all that bouncing about (that’s because there are fewer Lib Dems, and a smaller sample is more volatile) is slowly declining. Finally Labour voters were initially only slightly negative as they gave the government the benefit of the doubt, but quickly became strongly negative – again, as one would expect.


So the main cause of the big decline from the high government approval ratings in the early days of the government is just Labour voters going from “don’t know” to disapprove – a rather inevitable change. More significant is the slow decline since then. This has mostly been amongst Lib Dem voters – both the very gradual decline on the graph, but more important something that isn’t shown on the graph – the change in the size of three groups that those lines represent. Lib Dem voters may be continuing to support the government, but at the same time the number of Lib Dems voters is in decline and that positive yellow line is representing fewer people.

Yesterday ConservativeHome and Guido both had stories about a possible return to the frontbench for David Laws, saying the government were testing the water on how it would be recieved.

YouGov ran a question for the Sun yesterday asking if people thought David Laws should be able to return to the government – at present, the answer appears to be no.

23% of respondents thought that Laws had paid the price for his earlier errors and should be able to return to government. A further 20% said he should be able to return to government eventually, but it was still too soon. 39% said it would never be appropriate for him to return to government.