How Labour lost (and how we might lose again)

By Declan Gaffney

The idea that Labour’s election defeat was due to the sudden desertion of the party by skilled working class voters – known as ‘C2’s in the jargon of market research – has gained widespread currency, reflected in the convergence of most of the candidates for the leadership around certain themes: notably immigration and welfare reform.

As Liam Byrne put it in a Progress pamphlet published on Friday:

“These voters are the bedrock of our coalition. But their support for Labour has fallen off a cliff. In 2005, 43% of C2’s were Labour. Now Mori say it’s down 20%, to just 23%.”

How accurate is this picture? Estimates released by Mori on Friday allow us to compare Labour’s vote share among different social groups since 1997 on a consistent basis. While this sort of analysis is not an exact science, and different methodologies will yield different results, these are the best estimates we have to go on at the moment, and they tell a more complex story.

On these figures, Labour’s vote share among the ‘C2’ demographic did fall in 2010, but by 11 percentage points, rather than 20 (see chart, below).*

Secondly, on Mori’s figures, the fall in the C2 vote share in 2010 was nothing new. In 2005, Labour’s share in this group fell by 9 percentage points. This suggests that Labour’s problem with the C2 demographic is longer term in nature than the narrative offered by Liam Byrne indicates.

Thirdly, Labour lost vote share across all these demographics in 2005 and 2010. While the fall among C2’s is proportionally greatest, it looks like part of a broader pattern.

Fourthly, consider what has happened to Labour’s share of the lower income DE demographic since 1997. In 1997, Labour had 59% of votes in this group; by 2010, this was down to 40%. Liam Byrne’s pamphlet virtually ignores this group, which is no less important in terms of Labour’s share of the national popular vote than the C2’s. (Lower turnout among DE’s is offset by the greater size of the group.)

These figures do not, on the face of it, support the sort of post-election narrative offered by Progress or the leadership candidates so far. They suggest that Labour’s electoral problems are longer term in nature and more widely spread across social groups.

Much of the post-election debate has been couched in terms of how to win back the C2 vote. But there is little reason to think that task can be accomplished by focussing on this demographic in isolation. Meanwhile, there is another question which nobody seems to be asking. How did Labour, with its highly redistributionary policies, manage to lose nearly a third of its vote share among the lowest income demographic while in government?

There will be no simple answer to that question, but the very invisibility of this group in Labour’s post-election narrative points to one line of explanation. Labour’s political strategy has long been concentrated on maintaining what was seen as the winning coalition of C2’s and more affluent voters of 1997. For many in the party, the most persuasive argument for this approach lay in the progressive policies Labour pursued in power, which directly benefited lower income groups. But the political strategy meant that voters in these groups were not an important target of Labour’s political messaging. Could it be that many lower income voters stopped supporting a party which often spoke about them but rarely addressed itself to them?

*The difference is easily explained. The figure cited in the Progress pamphlet used an earlier estimate by Mori produced just after the election, which gave Labour a lower share of the C2 vote for 2010, and combined it with an estimate of Labour’s 2005 share from a different source, which gave Labour a higher share than Mori.

Source for graph: Ipsos-Mori ‘How Britain voted’ 1997, 2001, 2005, 2010′