A New Study on BNP Support: Why IPPR Got (Some of) it Wrong

"Extremism of various forms and its support have much greater social and policy relevance than previous years..."

With each election comes a new wave of panic about the possibility of a BNP breakthrough. One problem with debate about the far right is that often remains completely detached from the rapidly-growing evidence base on what drives support for these parties.

We are not where we were in the general elections of 2001 or 2005: we now know a great deal about what motivates some Britons to vote BNP and the characteristics of communities where the far right tends to poll well.

A report released this week by the Institute for Public Policy Research (IPPR) attempts to add to this evidence base. First, even though the report completely glosses the academic literature on this topic we should take a moment to applaud IPPR for trying to add to our understanding of this important issue. Clap clap.

The think tank community has something which some academics often crave: immediate impact and an ability to bend the ears of journalists and politicians who desire a quick digestible soundbite based on ‘the evidence’. If they play it right, think tanks can help shape agendas.

It is precisely for these reasons that think tanks need to remain as close as possible to the underlying evidence base which informs our understanding of questions like who votes BNP, or how can we explain the rise of the far right?

Unfortunately – as is the case with IPPR’s latest report on BNP support– sometimes this does not happen.

There are many things which are correct in this report and corroborate our existing work on these questions. Crime and unemployment have little effect on BNP voting. True. The party performs less well in areas where education levels are low.

True. We agree with these. But then there are other things in this report which are seriously problematic and need to be flagged.

We will highlight these flaws against the backdrop of the press release which accompanied the report. This release tells interested journalists and policy wonks that it is ‘not immigration but alienation and an ability to overcome social challenges such as isolation and low skills which are the main drivers for BNP support’.

It continues, ‘this important finding contradicts the argument that immigration is to “blame” for pushing voters into the arms of the BNP’. Enter the co-director of IPPR Carey Oppenheim who concludes: ‘What our findings can finally lay to rest is the mistaken popular belief that it is the experiences of immigration which leads to people voting for the BNP’.

Ok, so here are the problems as we see them.

Problem #1: The report tells us nothing about actual BNP voters or their motives because it is focused at the level of local authorities. This is a common approach in many early studies of the far right. It has potential to tell us something about the types of areas where parties like the BNP prosper, but tells us next to nothing about individual BNP voters and – crucially – their motivations.

By looking only at local authority level the report glosses over an enormous amount of heterogeneity and is susceptible to what academics call the ‘ecological fallacy’ whereby people start making inferences about the behaviour of individuals based on the characteristics of the communities in which they are based.

We are told, for example, that support for the BNP is less about immigration than political and social exclusion. Why then do BNP voters rank immigration the most important issue facing both their families and the country?

It might not be the direct experience of immigration which is fuelling BNP voting, but anxiety about immigration is a major driver of the party’s support, like it or not.

Problem #2: The methodology of the report is flawed, in several respects. First, the report has not corrected BNP support to account for ethnic diversity which means that they look at support for the party by looking at all voters rather than the ones who really matter here: white British voters.

The report compares support levels for the BNP among the entire electorate instead of support levels for the BNP among its target electorate of white voters .Unsurprisingly, this leads the researchers to find that the BNP does less well in ethnically diverse areas. Of course they do! Such areas include a large population that the BNP doesn’t target and which give it no support.

Problem #3: The report adopts a blanket approach to ethnic diversity. The claim is that ‘areas with larger numbers of non-white people are less likely to vote for the BNP’. Yet the report does not break ethnicity down by different groups, choosing instead to lump everyone into the category of ‘non-white’. As our research has shown, taking the time to make this distinction makes all the difference.

We (and other academics if the authors had read the existing literature) find a strong relationship between higher levels of BNP support and large Muslim communities. We also find that non-Muslim Asians have no effect on BNP support and that BNP votes are actually lower in areas with large numbers of Black voters (read my earlier blog).

Simply looking for a correlation between the proportion of non-white and BNP voting misses the point: support for the far right is far more complicated.

Problem #4: The report measures turnout by looking only at turnout in one general election (2005). This glosses the critical importance of local elections to explaining the BNP’s rise, where turnout has not always been low.

Initial local BNP gains in one ward in Burnley in 2002 took place against the backdrop of a rise in turnout, likewise in local elections in Barking and Dagenham over the years 2002-06.

Looking only at one general election and drawing conclusions about turnout and party support is risky business.

Problem #5: The report puts most of its chips onto the argument that there is ‘strong evidence that recent immigration is not driving people to vote for the BNP’, and that ‘immigration to an area appears, on the whole, to make people less likely to vote for the far right’.

The authors allude to the fact that it is less objective experiences of immigration which motivate far right support than subjective experiences of immigration, i.e. I might not have seen or interacted with immigrants on my street, but I’m concerned about the effects of immigration on my society.

This is a crucial point but is downplayed throughout the report. Instead, the ‘finding’ that objective experiences appear less important (which is itself based on limited evidence) is taken as evidence that immigration per se is not a major driver of support for the BNP – which run counter to all the evidence we have from actual BNP voters.

With these problems in mind, the point about all of this is that future research – whether from think tanks, academics or others – needs to think more seriously about ways of building on what we already know, rather than spinning out a convenient headline finding to journalists and politicians.

Despite the claims of IPPR’s co-director, no popular beliefs have been laid to rest here. We already knew most of this, while the ‘new’ claims lack the robust evidence which is required to turn them into facts.

Extremism of various forms and its support have much greater social and policy relevance than previous years. It is for these reasons that we need to think and research together in order to answer the challenges thrown up by these phenomena.

Matthew Goodwin and Robert Ford