AshcroftBy Lord Ashcroft KCMG PC.

It is a frequently heard
complaint in Westminster that political reporters and commentators do the
public a disservice by obsessing over trivial “process stories” at the expense
of things that actually matter.

Certainly it can be
exasperating when, as has happened recently, journalists write unhelpful
process stories and the following day criticise the government media operation
for the proliferation of – guess what – unhelpful process stories. Of course,
complaining about the media is not likely to get you very far in politics. And
the best way to stop them reporting trivia is to make sure they have something significant
to report instead.

That said, how much of this
political froth do people notice? Last week I conducted an experiment to find
out which recent political stories had got through to people, and whether they
were more likely to recall the substantive or the trifling. Half the sample in
my online poll were asked to write a brief description of any stories about
political parties or politicians they had heard about in the last few weeks or
months, whether they thought the story was important or not.

It is worth noting at the
outset that 41% of respondents could not recall a single thing. The two stories
most likely to be recalled spontaneously were perhaps at the trivial end of the
spectrum, though the partisan would claim they highlighted important truths.
Far ahead of the pack was Andrew Mitchell’s contretemps with a Downing Street
policeman, recalled by 33%. Next, on 13%, was the misunderstanding over George
Osborne’s train ticket. Eight per cent remembered a story about the Scottish
referendum, and 7% referred to Child Benefit cuts. MPs’ expenses, the Jimmy
Savile affair, David Cameron’s announcement on energy prices, and other welfare
reforms were mentioned by 6%.

One in twenty mentioned the
GDP increase and the end of the recession; the same proportion recalled stories
about prisoner voting. Fewer than one in thirty three mentioned pressure for an
EU referendum or speculation about Boris Johnson wanting to lead the
Conservatives; the postponed badger cull, the Police Commissioner elections and
Ed Miliband’s conference speech were mentioned by one in fifty.

We asked the other half of
the sample whether they recalled particular stories (while warning that some
false ones had been added to the list, to reduce the risk of people claiming to
know more than they did). Again, “plebgate” was top, with 88% saying they had
heard about the story, including 70% who said they had heard a great deal. In
this question, though, we also asked people to rate the importance of the story
in question. Strikingly, the two stories with the highest importance ratings
(the benefit cap and raising the tax threshold) came well down the league table
for recall. The unfortunate matter of the Chief Whip and the policeman rated
eleventh for importance out of eighteen stories tested.

Of the stories we prompted
for, the most widely recalled were “plebgate”, the pasty tax, the Virgin Trains
West Coast franchise debacle, David Cameron accidentally leaving one of his
children in a pub, and new rights for householders to tackle intruders. This
last was the only one to number among those regarded as most important – the
others being raising the tax threshold, child benefit changes, and the council
tax freeze.

Overall, men were more likely
to recall news items, whether spontaneously or prompted, than women. Older
people were more likely to do so than younger respondents (who were also less
likely than others to say that any given story sounded important). Liberal
Democrat voters were the most likely of all to have noticed particular stories.

Whatever the apparent disparity between what people notice and what they
think matters, it would be a mistake to assume that the stories people
regard as unimportant have no impact on them politically. The findings,
especially of the unprompted question, show that people do register the petty
things. They may protest that these incidents are irrelevant, but their view of
a government or party can only be shaped by the things they hear about. In an
ideal world, the media would pay more attention to the big things and devote
less time to the Chancellor’s railway mishaps and other ephemera. But
frustratingly, it is true both that most people do not hear a political message
until well past the point at which politicians are sick of repeating it, and
that they are more likely to notice small things than big speeches or policy
announcements. It means the parties need to work all the harder to shift
attention to what they want the story to be about.

Comments are closed.