Political Post-Mortems: of pollsters and parties.

I haven’t written -shock horror- about the political scene for a while, so I thought I ought to remedy this. Tuesday provided me with the required inspiration and story.

The 19th January saw not one, but two important reports of political significance published. Yes, Tuesday marked the release of the two post-mortems into the 2015 General Election: the pollsters’ examination of how they could have gotten the election so wrong, and the Labour party’s internal inquiry into how exactly they lost said election. Because you know, nothing helps boost a party fraught with internal struggles, tension and splits like a) reminding them their party lost at the General Election, and b) reminding them one of the reasons their party lost was due to the public’s perception of the party leader. (Not that this is an issue of topical relevance, of course.) But I digress.

The first report, which is the preliminary findings of an independent inquiry established by the British Polling Council and the Market Research Society, seemingly concludes that the pollsters appeared to have succumbed to ‘herding’. This essentially occurs when individual companies deliberately alter their sampling formulae to ensure they are producing results not dissimilar to those published by their competitors. Moreover, the report states that sample failures are to blame for the polling mistakes; pollsters generally underestimated the support for the Conservatives ahead of the general election because of a failure to include sufficient numbers of Tory-leaning voters in their survey samples. As the report puts it, there had been a ‘systematic over-representation of Labour voters and under-representation of Conservative voters’. Judging by the sharp contrast between the polling predictions and the actual result, that is quite an understatement.

Now, most surveys during the campaign last year consistently placed Labour and the Conservatives on an equal footing or within the margin of error, especially in the lead up to May. I remember being in America and following the news, and was led to believe I would return home to vote in an election which would produce a hung Parliament. I had to explain what that term meant, and what the implications of such a Parliament would be to some of my American friends. So imagine the thrill of surprise I felt on results night, when the now-infamous exit poll flashed on the television screen, revealing the Conservatives to be on course for an outright majority win. From the polls proclaiming them to be neck-and-neck, the Tories went on to defeat Labour by some 6.5% nationally. That is some difference to be sure, and thus the question was raised: how could the pollsters have called so incorrectly? Cue the announcement there would be an inquest of sorts into the conducted polling.

According to this report, it was determined the evident survey imbalance was a result of older voters being harder to find, and as traditionally the older voter is Conservative, this led to a disparity of views being surveyed. In addition, young voters were easier to reach, and in particular the younger voters polled were far more likely to vote than their peers, being more engaged in politics than their peers. This category arguably led to the over-sampling and representation of Labour-leaning voters.

It was also concluded that there was a category of ‘busier voters’, who were more difficult to reach and so their view not considered in the polling. As such, it appears this category were also likely to be Conservative-leaning, given the disparity in sampling and recorded representation across collective polling.

The report also stated any evidence of a late swing from Labour to the Conservatives because of public fears of an SNP-Labour pact were in fact ‘inconsistent’. And speaking of Labour…

The report assessing Labour’s performance at the General Election, seeking to determine where the party went wrong, appears to conclude  Labour lost because voters did not trust the party on the economy, leadership, or immigration. Intriguingly, the Beckett Report was keen to reiterate that many of the popular explanations pertaining to Labour’s loss in 2015 ‘should be treated with caution‘, and may not actually explain  why the party lost. Some of these popular and oft-quoted explanations include branding Labour as being perceived as ‘too left wing’ under Ed Miliband’s leadership, as well as fears that Labour was considered ‘anti-business’ and ‘anti-aspiration’. (Goodness me, I had thought that word had been consigned to the past. I thought wrong; it has come back with a vengeance to haunt me.)

Instead of the above explanations being deemed as valid, the report submitted the weight of research evidence undertaken since the General Election argues a number of other so-called ‘contributory’ reasons are more significant. Such contributory reasons, the report states, include the electorate’s perceived weakness of Ed Miliband, a particular fear amongst English voters that Labour would work with, and be controlled by the SNP, and a belief that the economic crisis was ill-managed under the last Labour government. The report also finds that Labour failed to convince voters of its welfare and immigration policies.

Another contributory reason, which may be overlooked but I believe is important considering the current political landscape, is considering the political ‘revolution’ of sorts which was, and arguably still is, ongoing. I am referring to the utter dominance displayed by the SNP, which resulted in the virtual extinction of all other political parties in Scotland. This would have been a factor beyond the control of Labour during the election campaign, but arguably Labour did not take the SNP threat seriously. The result? Scottish Labour fell, even in their historic strongholds. The report itself states:

…We were badly beaten. The collapse in Scotland made it impossible for us to be the biggest party and the Liberal Democrat collapse enabled the Tories to gain an overall majority and keep us out of power…

The report also noted – and this may make for painful reading for Labour –  it would be difficult for the party to win in the 2020 General Election. This essentially comes down to changes to constituency boundaries, voter registration changes, and restrictions on trade union funding of parties: all due to policies adopted by the current Conservative government. Labour’s loss in 2015 was the Conservatives’ gain, and may continue to be so in 2020.

These two reports are intriguingly-interlinked, hence their sharing a blog post. Why? I hear you ask. Well, the Labour party would have carefully gathered and scrutinised the polls which, declaring Labour to be on equal footing with the Conservatives, ultimately suggested a hung Parliament. Labour therefore would have assumed it was on track to enter government, as part of a coalition or if it pushed hard enough in the final days, a minority government. Remember, whilst we the electorate, the Press and political analysts were sidelined by the polls, so too were the political parties themselves. Polls dictate the actions of campaigning politicians, so it is fair to assume that Labour acted according to the polls during their 2015 campaign.

Which brings me to the next point. If the findings of the independent inquiry established to examine the failings of the pollsters are correct, that the sampling was scattered and imbalanced, what did the advisers in the Labour camp think when scrutinising the polls? What did they say to the Labour leadership? Most especially the poll findings regarding the electorate’s views in relation to leadership, immigration, and the economy? Remember, Miliband was still viewed as the weaker leader compared to Cameron, and Labour was taking hits pertaining to its handling of the economic crisis when it was in government. This is despite, as the independent report suggests, these polling results being over-representative of Labour-leaning voters, and under-representative of Conservative-leaning voters. So goodness knows what the results would have been had the polling surveys been conducted on an equal footing. On the basis of both discussed reports’ findings, the surveys published in 2015 painted a sympathetic picture, but an unrealistic one.

Considering the establishment of an independent inquiry, polling companies do appear sincere to ensure such mistakes as witnessed in 2015 are not repeated come 2020. After all, there is money at stake: to appear complacent and unwilling to admit to mistakes in their survey process could undermine these companies commercially. Yet, it could be argued this is the same for the Labour party, but on an electoral basis. Labour have been presented with the post-mortem results pertaining to their electoral defeat in 2015, and cannot afford to seem complacent and unwilling to change.

We shall have to wait and see whether the current leadership will consider the findings of the Beckett Report with the same resolve to change the party as the polling companies’ resolve to rectify their mistakes. The difference is one aspires to govern, whilst the other aspires to predict who will govern. So, no pressure.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s