Two Leaks

Two leaked memos, one by Tony Blair and one by Labour's Political Consultant and pollster Philip Gould, have stirred up the news agenda this week, but beyond the embarrassment of the leaks themselves it is hard to see what much of the fuss is about. They tell us little we didn't know or guess already.

Consider the second of the leaks, the Gould memo, first. It is not the first of Mr Gould's memos to have been leaked, and in any case it is public knowledge that Mr Gould conducts regular focus groups and polling to test public opinion for the Labour Party and reports on the results to the party leadership. That is the job of a party's private pollster, and is what he is paid for.

The consensus opinion of many of the commentators is "Sack Gould". Why? Two, to some extent contrasting, reasons - either because his conclusions are negative ("Mr Gould should never have committed this bleak view to paper, still less circulated it. His memo contains ... enough pessimism to suck the life out of a demoralised cabinet" The Sun argued), or because they oppose the idea of the governing party monitoring public opinion altogether. The first of these, the "shoot the messenger" argument, is plainly misconceived. Are we really expected to prefer the Prime Minister to be entirely surrounded by fawning yes men? If there is bad news (of whatever sort) to be delivered, somebody should deliver it. Members of the Cabinet would end up a great deal more than demoralised if they were allowed to continue in blissful ignorance of their failures until unexpectedly ejected from office by the electorate. The job of any party's private pollster is to make objective assessment of the state of public opinion and its political implications, and ensure that it is delivered to the party leader or other relevant officials without being distorted or diverted by his own or anybody else's political agenda. If he doesn't his work is worse than useless.

Of course, it is possible to challenge not the propriety of Mr Gould's having written the memo, but the accuracy of his assessment. One or two writers have confined themselves to complaining that focus groups are not a reliable means of measurement, only a means of exploring the thought processes of small groups of electors in depth, which can throw up hypotheses that should then be tested quantitatively by conventional opinion polls. True enough, but beside the point in this case. It is not clear from Philip Gould's memo whether his conclusions are based entirely on focus group work or have also been tested with representative samples, but in any case the published polls by MORI and the other companies back up almost all of what he said.

"We are outflanked on patriotism and crime", he wrote in early May. MORI's leader image poll for The Times in April found, as had the previous one in October 1999, that more people chose patriotic as a description of William Hague than Tony Blair - the only one of 14 measures on which Hague beats or has ever beaten Blair. On crime, the position is less clear cut. A MORI survey for the Mail on Sunday at the end of April, just after the Tony Martin case and when the Tories were making crime a particular issue for targeting the government, found that in England Labour led the Conservatives by 34% to 23% as the party with the best policies on law and order; but Mr Gould's concern was perhaps justified by the 37% who thought no party had a good policy on law and order or didn't know which party was best.

"We have been assailed for broken promises" wrote Mr Gould. At the start of June, only 25% thought the government had kept its promises, while 60% thought it had not. "We quickly seem to have grown out of touch"; 51% of the public think Mr Blair is out of touch, and 44% that he is in touch. (MORI/Mail on Sunday poll, 8-9 June).

And, just, as Mr Gould's assessment seems justified in its details so we can surely agree that there was reasonable cause for the concern expressed both in his memo and Mr Blair's. By the end of April, satisfaction with the government had dropped 8 points in four months (from 45% to 37%) and with Mr Blair personally 5 points (57% to 52%); certainly the honeymoon period when 57% were satisfied with the government and 75% with the Prime Minister was well and truly over. Furthermore, although the voting intention polls remained healthy, primarily because of continuing public distaste for the Conservatives, Labour suffered dramatic losses in the council elections across England and humiliation in Ken Livingstone's sweeping victory in London.

The second complaint against Mr Gould, that the Prime Minister shouldn't be taking any notice of public opinion in the first place, raises different questions. Stated in these bald terms it sounds, and is, ridiculous; but the real objection is to "government by opinion poll", and some people take this as a reason to object to opinion polls as such. The arguments of former Clinton pollster Dick Morris, mentioned by Mr Gould, that endorsement by opinion poll confers some sort of democratic legitimacy on a policy, is one with which most of us would disagree. Democratic legitimacy is conferred by popular election to office, sometimes by referendum, arguably perhaps even by "citizen's jury", depending on how that is construed. But governments have always tried to take public opinion into account, even long before there were scientifically-conducted polls to use as measurements, and have sometimes let it dictate their governmental or electoral strategy. (One remembers in particular Stanley Baldwin's "appalling frankness" in explaining how he refused to advocate rearmament in the 1930s because it would have lost him the election.) In exercising the decision-making power that is democratically his by reason of his general election victory, a Prime Minister is perfectly entitled to consider public opinion as one of the factors he ought to take into account. The real point is that Prime Ministers should not be isolated from public opinion, but that they should be capable of putting it in its proper place, weighing the thoughts of the electorate against the demands of good government.

For here we come to the earlier of the two leaks, Tony Blair's memo. This is far more revealing, though it only confirms what everybody had already assumed, that the government's reaction to the situation outlined in Mr Gould's memo has been panicky, fatally concerned with a quick fix of appearance rather than substance, and especially centred on maintaining a special aura of popularity round Mr Blair himself. The outcome, a disastrous series of initiatives and statements, has much more sharply undermined the government's position than anything that had gone before. In April, the poll figures on almost every front were still at a level most previous Prime Ministers could only have dreamed of. Remember that the 42% to 37% lead of which Mr Gould talks with such dread would, on uniform swing, deliver a Labour overall majority of 90 or more. But since April, satisfaction with the government has dropped by a further 12 points (to just 28%), with Mr Blair by 13 points (he is now for the first time scoring negative net ratings), and the voting intentions though still showing a Labour lead have also taken a perceptible knock.

On individual issues the same is true. Look at the deterioration on law and order, a particular focus of recent policy panic. At the end of April, as we have seen, 34% thought Labour were the best party on law and order, and they led the Tories comfortably. (That poll was only in England, which should slightly disadvantage Labour). Last weekend our poll for the Mail on Sunday found only 20% picking Labour, while ICM's in the Guardian put it at 22%; both polls found the Tories almost neck and neck.

In other words, everything Labour in general and Mr Blair in particular has done since these memo were written has been counter-productive. A leader should be reading and responding to the poll data, but also acting with judgement, not simply allowing himself to blown about by the winds of transient public opinion. It has always been an accusation of New Labour's opponents that this is one of Mr Blair's faults, but the impression over the last couple of months - backed up by the Blair memo - has been much stronger.

Paradoxically, the deterioration in the government's position since April exemplifies the reason why the accusations against Mr Gould, and the rest of us in the polling industry, that our work brings an undemocratic threat of permanent "government by opinion poll", is misplaced. Except in the very short term, government by opinion poll doesn't work. The sooner Mr Blair remembers this, the better; and, if he does, he is still well placed to win the next election.