Skip to main content

BLOG UPDATE: is Labour really too left-wing to win an election?

Ed Fieldhouse

Few could have anticipated the surge in support for Jeremy Corbyn, who after getting on to the Labour leadership ballot by the skin of his teeth has become remarkably popular among his party’s grassroots . Most Labour people have something to say on the Corbyn phenomenon, and opinions range from doom-laden scenarios to a welcome break from the New Labour past. Former PM Tony Blair, unsurprisingly, speaks for the former. Earlier in the year, he gave a controversial interview to the Economist suggesting Ed was a little too left-wing . And last week, he struck again, arguing Labour could not win from a “traditional leftist platform”. Blair urged against a lurch to the Left and argued people who said their heart was with Jeremy Corbyn should “get a transplant”. According to Blair, then, not only is Labour in danger of turning back the clock to the bad old unelectable days of the 1980s, it already went too far in that direction under Ed Miliband. Is this a reasonable claim? In January in this blog I suggested Labour had more pressing concerns than whether their position was too far to the left (specifically the perception of economic competence and Miliband’s leadership). Here I revisit the same questions using data from the post-election wave of the BES (wave 6). Although Miliband had already resigned at the time of the fieldwork, the findings were almost identical.

First, I argued that what is important is not whether Labour are slightly more to the left of centre, but how this compares with to where voters’ preferences lie, and whether rival parties take a more or less centrist position. In very simple terms, this is at the heart of the ‘Economic Theory of Democracy’ in which Anthony Downs made famous the ‘median voter theorem’, which stated that a majority rule voting system will result in the outcome most preferred by the median voter (as defined on a one dimensional spectrum like left-right)[1]. As I suggested above, elections are about much more than this, and plenty of political scientists have critiqued, modified or abandoned Down’s spatial model. Nevertheless, it is enlightening to take this idea as a starting point in order to assess Tony Blair’s claim and to examine data from the British Election Study which, in recent times, has asked about respondents on left-right placement[2] and also where respondents view the main political parties[3].

Figure 1 shows the responses to these two questions on an 11 point scale (zero to ten) for post-election BES in-person surveys since 1997[4] and for the post –election wave of the internet panel (carried out in May 2015). A low score represents a more left-wing position and a high score a more right wing position. The grey line shows that the mean position of the electorate was approximately 5 (i.e. in the centre) at each time point (the median position is 5 in each of the datasets). The red line indicates the mean position of the Labour Party as perceived by respondents. As Blair suggested, Labour is now slightly further away from the centre ground with a mean perceived position of 3.2 (and median 3) than during his time in charge, including 1997 when he was first elected PM.

Figure 1

Party and respondents left right placemenrt since 1997

It is important to see this in the context of where the other parties stand. Figure 2 shows the average placement of parties by BES respondents, and by political scientists who responded to the BES expert survey. We see that whilst respondents see Labour rather left-of-centre, the Conservatives are seen as much more right of centre and only slightly more moderate than UKIP. Whilst the Lib Dems are in the middle ground, this could not save them from electoral disaster. Our sample of experts corroborates the perceptions of voters: there are no substantial differences between the average expert assessments and the average voter perceptions, though there is a suggestion that compared to the experts, voters are more inclined to agree with Blair and see Labour as left wing.

Figure 2

Expert and Citizens W 6

 The preceding analysis tells us about average position but nothing about the distribution of voters’ preferences. Figure 3 shows the distribution of respondents’ placements of both Labour and the Conservative on the 11-point scale. We see that there is a fairly wide distribution of perceptions of where Labour stands, with a fair number (16%) putting Labour on the extreme left, but a significant minority seeing them as right of centre. The most common positions are on the centre left (positions 2 to 4) broadly in line with the expert placements. What is more interesting, however, is that whilst Labour might be seen as a little off the centre-ground, in May 2015 they were still not regarded as less far to the left as the Conservatives were to the right. Over 20% of voters regarded the Tories on the extreme right-hand end of the scale, and the modal position is at point 8, further from the median voter (5) than Labour (3).

Figure 3

Self and Party Placement Wave 6

The last and perhaps most important piece of the jigsaw is to see how far away parties are from their own voters (who they want to keep) and from their rivals whose votes they want to steal. Using the 2 questions analysed above, I subtracted the party’s perceived position for each respondent from his or her own position, so that if those intending to vote for a particular party are, on average, to the left of Labour the score is negative; and if they are to the right the score is positive. The resulting statistic is broken down by 2015 vote in Figure 4. What we see is that Labour are actually a tiny bit to the right of its own supporters which is the natural place to be for a party on the left that wishes to attract new support (a difference of -0.6). This again compares favourably with the Conservatives (not shown here), who are regarded, on average, as slightly to the right of their own supporters, and to UKIP voters (-0.9). Moreover, Labour is also to the right of some of its’ other key rivals in the shape of the SNP, Plaid Cymru and the Greens. On the other hand it is moderately to the left of Liberal Democrats and non-voters. The position in relation to the Lib Dems is particularly interesting with the election of Tim Farron, who will undoubtedly shift the centre of gravity of the Lib Dems to the left and appeal to the many voters lost to Labour in 2015. For Labour to hold on to these new supporters a shift to the left could be risky. However, since the average position of SNP voters is roughly the same distance to the left of Labour as the Lib Dems are to the right, the dangers of a rightward shift are just as real.

Figure 4

Own LR versus Labour LR W6

Labour’s perceived position on the left going into the 2015 General Election was more one of image than one of policy. Arguably, there was little between the major parties on major issues of policy, with Labour desperate to appear as fiscally responsible. Moving forward, Labour faces a war on many fronts. Following the 2014 Independence Referendum, it will be a challenge to win back seats in Scotland from the SNP without a powerful anti-austerity and anti-Tory message. At the same time, it will be just as hard to win over voters in middle England without persuading voters that it is fiscally responsible. And as argued above, elections are not just about left and right: an analysis of which party voters are closest to shows that only 12% placed themselves closer to Labour than to any other party. In other words, a simple spatial model could not explain why Labour (or any other party for that matter) would get as much as 30% of the vote. If we adjust this by the ‘propensity to vote’ score in the BES (effectively weighting according to whether a respondent would ever consider voting for each party) the proportion increase to 18% (82% of whom did actually vote Labour). Using the same procedure we can observe how many would be predicted to vote Labour at different left-right positions. We find that it makes only a small difference, the predicted share changing from 16.5% at point 2 on the left right scale, to 18.0% at point 3 and 18.6% at point 4.

In reality, there are many more important factors that, whilst affected by the left-right position of the party, cannot be reduced to a question of ideological positioning (not least economic competence). Ultimately Labour’s fortunes are unlikely to rest on whether Labour choose a leader to the right or to left, but rather on choosing a leader who can perform the seemingly impossible task of simultaneously restoring the party’s economic credibility (broadening its support base) and at the same time appealing to its traditional supporters. What Labour needs is to avoid the ‘left-wing’ label but not necessarily the policies. This requires a leader with popular appeal and a program of progressive polices which can unify opposition to the Conservative government. Of course, that is easier said than done.

[1] Downs, Anthony (1957). An Economic Theory of Democracy. New York: Harper.

[2] In politics people sometimes talk of left and right. Where would you place yourself on the following scale?

[3] In politics people sometimes talk of left and right. Where would you place the following parties on this scale?

[4] The questions were not asked in 2010.