My Lords, many Members of the House will recall that I introduced this Bill in the last Session. It only got to First Reading, but when I introduced it in this Session it got a much more animated welcome, for reasons that I will go into.
First, I will go back a few years to when I was a student—well, that is many, many years. I did some work on polling then, and I learnt some of the techniques of polling, such as random sampling and the importance of finding the people who were chosen for the random sample and going back to them until you actually get their views; you do not take any substitutes. I learnt about quota sampling and the importance of getting class, sex and age in the right numbers and the right groupings to represent a cross-section of society: the make-up of the whole population.
I also learnt about the inexorable margin of error, however good your polling is. Most important of all, I was told about what makes a good poll and what does not. A poll should be run in a scientific, politically neutral manner, with no influence from those who have commissioned and indeed paid for it. That means, to take just one example, avoiding leading questions and ensuring that the wording of questions is fair and unbiased.
For some time now I have been concerned at the direction in which polling in this country has been moving. Rigour and accuracy seem to be subordinated to the demands of speed in getting the poll out and keeping the cost down: instant polling and cheaper polling. The media expect polls to be completed in an extraordinarily short time, often to be ready for publication the day after the sample has been taken. That means that there has been a preference for a predominance of
internet and telephone polling, often using predetermined panels which stay the same throughout the whole of the polling period.
My concern about the dangers of this corner cutting were reinforced when I attended a seminar chaired by my noble friend Lord Lipsey, who I am glad to see is going to speak in the debate. Polling experts John Curtice and Peter Kellner were in attendance and they confirmed, in answer to a question I put, the paramountcy of speed over accuracy because of the incessant demands of the media. That is the background which gave rise to my strong determination to introduce this Bill.
What reinforced for me the point that accurate polling is an important issue for the future of our democracy was the one rogue YouGov poll held on 7 September 2014 that seemed to indicate for the first time in the referendum that the Yes campaign was ahead, by 51% to 49%. This caused a widespread panic among politicians in the Better Together campaign. It resulted in a vow set out on the front page of the Daily Record to go for even greater devolution of powers to Scotland and led to the creation of the Smith commission. In light of the actual result of the referendum, it is clear that the fears of a Yes victory were unfounded and that the nationalists had directly benefited from just one highly inaccurate poll. It is not right that the real issues of democratic politics should have been so materially affected by a statistical prediction that turned out to be so wrong. Indeed, the course of history was changed by that one inaccurate poll.
The general election gave further evidence of the direct and highly undesirable impact of polling on politics and on events. The constant polling—and it was constant—and the constant media coverage which resulted from that made it seem beyond doubt to the media and to all of us that the contest for the general election was going to be neck and neck. Some polls suggested that Labour would be the largest party. All of that polling shaped the nature of the debate. Members of this House will recall that throughout the election, the main topics of debate were not important policy issues such as defence, foreign affairs and the health service, it was the consequences of a Labour minority Government, with the SNP set to hold the balance of power.
Noble Lords will remember the posters of Ed Miliband in Alex Salmond’s top pocket and of Nicola Sturgeon pulling the strings. All those resulted from the polls, which predicted that the election was going to be neck and neck. Consequently, major policy issues were absent from the campaign and the result of the election could well have been different if we had focused on those major policy issues. Inaccurate polls, as they turned out to be, again appear to have changed the course of history.
As a result of those manifest polling errors, I have found that when I talk to—I was going to say “comrades”—colleagues and others, there is now far greater support for the creation of some kind of regulatory body overseeing political opinion polling than ever before. That is why I have reintroduced this Bill in this Session. The political will is there to ensure that similarly damaging mistakes are not made in the future and that our democratic process is not undermined.
Even the British Polling Council realised that something went wrong and admitted it. It has set up an inquiry into why the polls in the run-up to the election were so consistently inaccurate. But the British Polling Council is a self-regulating body and will be so in carrying out the inquiry. It proposes merely a one-off investigation. Understandably, the council claims credit when the polls are correct but it needs to accept some blame now that its methods have been shown to be ineffective. To ensure that a more rigorous and accurate system is introduced, we need an independent and permanent regulator.
Contrary to what has been said in some media headlines and comment, the Bill does not legislate to ban polls but would allow the proposed regulatory authority to impose limits on their publication if it thought fit. That is already the case in Spain, France, India and Italy where, if it is thought that that could be helpful, the decision could be taken. The Bill would replace the self-regulation of the British Polling Council with an independent body which would have responsibility for issuing regulation and guidance on things such as sampling methods, the wording of questions and arrangements concerning publication, including how close to election day polls could be published.
I should answer some of the criticism made by Professor Ron Johnston of Bristol University in a letter to some noble Lords—interestingly, he did not send a copy to me—and by the Political Studies Association on Twitter. I wish that both had approached me directly. They have expressed concern that the Bill might infringe academic freedom to undertake polling on political attitudes and behaviour for the purposes of independent research. That certainly is not my intention. Clause 1(8) makes clear that the authority’s regulatory powers will be restricted specifically to polling concerning voting intentions in, first, local authority elections; secondly, in parliamentary elections, including the Welsh, Northern Irish and Scottish Parliaments; and, thirdly, referenda. Academic research on other political behaviour will therefore not be affected. If there is any worry that wording changes might be needed, I would be happy to consider any amendment necessary.
It is also important that all those with an interest in polling are represented on the board of the authority. The Bill proposes that we should have representatives nominated by the British Polling Council, which would represent the industry, as well as representatives from all the political parties and the media. Indeed, I am open to other suggestions as well. Transparency is also important, and this Bill provides that the authority would publish its rules within six months of its establishment and consider amendments at least annually.
I welcome the fact that the noble Lord, Lord Cooper of Windrush, will speak in the debate—in the absence, sadly, of the super-pollster Lord Ashcroft from our midst. Are we not less spectacular and exciting in his absence? Incidentally, he is not really a pollster. He contracted organisations to carry out his polling during the election. He decided where it would be, the questions and the publication—he is a multimillionaire, of course, so he can pay for it—but he would not tell us which organisations carried out his polling.
So we have the noble Lord, Lord Cooper, from Populus, who will no doubt give us a view from the industry. I will be interested to hear what he has to say. I feel that the industry has been unduly defensive about my Bill. Given that the BBC is regulated, that we have regulators such as Ofcom, and that other organisations and industries are regulated, it surely makes sense that the multimillion-pound political opinion polling industry is brought into line with the others. I look forward to the noble Lord’s comments.
In conclusion, polling has grown exponentially in recent years.