Labour and Muslim voters

A big drop in support will worry the party, but the electoral risks are limited

This week Survation - working with the Labour Muslim Network - published the first reputable poll showing a large drop in support for Labour among Muslim voters since the beginning of the current Israel-Gaza conflict last October. Survation’s figures show Labour winning 60% of Muslim voters expressing a preference, down from 86% support recorded in an earlier 20201 Survation poll where Muslim voters were asked to recall their 2019 vote preferences.¹ As the chart below illustrates, Labour are holding on to only 72% of their 2019 Muslim supporters (light bars), with substantial numbers of Muslim votes now going to the Greens (14% overall, 12% of 2019 Labour voters) and the Lib Dems (9% overall and with 2019 Labour voters). The Conservatives, however, remain very weak with Muslim voters - winning 8% overall and just 3% among 2019 Labour voters.

Muslim vote intentions, all Muslim voters (dark bars), 2019 Labour voting Muslim voters (light bars)

Source: Survation/Labour Muslim Network

This poll has added to a growing debate about Labour’s possible troubles with Muslim voters, and the electoral risks posed by a substantial decline in Muslim. The Labour Muslim network, who commissioned the poll, described the current situation as “a crisis point for the future of the relationship between the British Muslim community and the Labour Party”, while last week the Guardian reported that the Labour leadership were now commissioning regular polling of Muslim voters, quoting one senior Labour MP as saying “Muslims are not only predominantly Labour supporters but they are also geographically important. There are many in a range of key target seats…and we need to pay attention to that.”

How big an electoral risk do Labour face from a large drop in Muslim support? I will try and answer this question in three was. First, could a large drop in support from Muslim voters put Labour at risk in the seats with the largest Muslim populations? Second, could a big drop in Muslim support make life harder for Labour, and potentially save Conservative incumbents, in marginal seats with large Muslim populations? Third, could a big drop in Muslim support put seats out of reach that Labour would otherwise expect to win on current polling?

The seats with the largest Muslim populations

Nearly all of the seats with the largest Muslim populations are seats Labour won in 2019. There are 38 constituencies in England and Wales where more than one in five local residents are Muslim. The recently released notional election results for the 2019 election on new constituency boundaries show Labour winning 37 of these 38 seats.² Given the concentration of Muslim voters in Labour seats, it seems logical that Labour might face serious problems from a big drop in Muslim support. This is not, however, the case in practice. Labour are so locally dominant in most of the seats with large Muslim populations that even a slump in Muslim support akin to that found in the Survation poll would not put the seat much at risk.

A little bit of modelling can show this systematically. If we make the simplifying assumption that the Muslim share of the electorate is the same as the Muslim share of residents captured in the census³, and further assume that Muslim voters in each seat voted in 2019 in accordance with Survation’s recalled vote figures (i.e. 86% voting Labour), then we can estimate the “Muslim Labour vote” in each seat. We can then apply the change reported in the Survation poll to this Muslim Labour vote to estimate of what might happen to Labour’s 2019 support if the 26 point drop in Muslim Labour voting reported by Survation happened in every seat.

Estimated impact of a 26 point drop in Muslim support in the ten seats with the largest Muslim populations

Sources: Notional election results calculated by Colin Rallings and Michael Thrasher, Census data from Office for National Statistics and House of Commons Library

The impact of lost Muslim votes in the most heavily Muslim seats seats is, in fact, pretty minimal, because nearly all the seats with large Muslim votes are seats where Labour is politically dominant across the board, and hence not dependent on Muslim support. The chart above illustrates this. Here we look at the ten seats with the largest Muslim populations. Labour has notional 2019 majorities of 30 points or more in every one of these seats. Every one still has a 2019 majority of 24 points plus even if Muslim support dropped as sharply as Survation’s poll reports. The pattern continues as we work down the list - Labour comfortably hold every one of the thirty seats with the largest Muslim populations even with a Survation level slump in Muslim support, and in only two of these thirty seats would the notional Labour majority fall below ten points. Even if Labour did as badly overall in the next election as it did in 2019, a large loss of Muslim votes won’t put any of these seats in jeopardy. And Labour is very unlikely to do as badly as in 2019 with non-Muslim voters, given all the polling has for several years now pointed to a very substantial Conservative to Labour swing.

Marginal seats with large Muslim populations

Given that most Muslim voters cluster in very safe Labour seats, I need to use a fairly generous definition of marginal seat to produce a substantial number where Muslim populations are large enough to be influential. I’ll define two sets of seats. Firstly, Labour seats with 2019 majorities under 15%. Secondly, Conservative seats with majorities of any size.

Labour do not face many risks from lost Muslim support in the marginal seats it currently holds either. There are only four Labour marginal seats with Muslim populations above 20%, so even large changes in Muslim voting produce only small impacts in the overall outcome, unless the seat is on an absolute knife edge. A 26 point drop in Muslim support change the notional 2019 outcome in just one of the ten Labour marginals with the largest Muslim populations - flipping Kensington & Bayswater from a Labour ultra-marginal to a Conservative ultra-marginal. Dagenham and Rainham would also a coin toss and there are wo other ultra-marginal Labour seats with small Muslim communities where the notional result might flip if Muslim support slumped. However, in all of these seats there is a large non-Muslim electorate which would swing to Labour with the national tide. A three point Tory to Labour swing, miles below current polling, would be enough to save every Labour marginal seat potentially imperilled by lost Muslim support.

Estimated impact of a 26 point drop in Muslim support in the ten Labour marginal seats with the largest Muslim populations

Sources: Notional election results calculated by Colin Rallings and Michael Thrasher, Census data from Office for National Statistics and House of Commons Library

Labour face little electoral threat from a slump in Muslim support in the seats they currently hold. Muslim seats are not marginal enough, and marginal seats are not Muslim enough, for Muslim votes to change the outcome in an election where the tide is rising for Labour. But are there Conservative held seats with large Muslim populations, where a slump in Muslim could help save the incumbent Tory MP from that rising Labour tide?

Not really. There are very few Conservative held seats with large Muslim populations - only 17 Tory seats have Muslim populations above 10%.⁴ And most of these are so marginal they will likely fall anyway even if the swing to Labour is much smaller than we see in current polls. The chart below shows the ten Conservative seats with the largest Muslim populations. The Conservatives gain add about one or two points to their majority if Labour lose a quarter of their Muslim support in these seat, and eight out of ten seats would still have Tory majorities of 10 points or less, so Labour would take them on a modest five point swing.

There are two seats in this top ten where Tory majorities are large enough that a loss of Muslim votes might plausibly help them hold off a Labour challenge - Pendle and Clitheroe and Harrow East. There is only one other seat outside this top ten where the Tories start with a substantial majority and stand to add at least one percentage point to that majority from a slump in Muslim support for Labour - Boris Johnson’s old seat of Uxbridge and South Ruislip. Every other Tory seat is either too marginal, or has too few Muslim voters, to be materially impacted by a big Muslim swing away from Labour

Estimated impact of a 26 point drop in Muslim support for Labour in the ten Conservative held seats with the largest Muslim populations

Sources: Notional election results calculated by Colin Rallings and Michael Thrasher, Census data from Office for National Statistics and House of Commons Library

How likely is a 26 point drop in Muslim support for Labour, anyway?

So far we have seen that a very large drop in Muslim support for Labour is unlikely to materially impact the next general election. Labour are too dominant in the most Muslim seats, and Muslim voters are too scarce in marginal seats, to move the dial much practically anywhere even with a big drop in Muslim support for Labour.

But there are also other indications in the Survation poll which suggest such a large swing away from Labour may not end up happening anyway come election day. If Muslim voters are going to abandon Labour in droves we would expect the following things to be true: Muslims should prioritise the Israel-Palestine conflict, where they are at odds with Labour, over domestic issues where they are not; they should have a negative view of Labour and its leader; and they should report becoming more negative about Labour in response to recent events. Yet on all these indicators, the evidence of widespread Muslim opposition to Labour is pretty limited.

“What would you say are the most important issues when it comes to deciding how you will vote at the next general election? Please name your top three issues”

Source: Survation/Labour Muslim Network poll

The Survation poll shows Muslim voters, like voters in general, are prioritising domestic isues - the cost of living, the economy and the NHS. While foreign policy in general and Israel-Gaza in particular do attract more attention from Muslims than voters in general, they are still second rank issues with this electorate. The top tier of issues are ones where Muslim voters, like everyone else, are likely to judge the Conservatives negatively and to hold a more positive view of the Labour opposition.

“To what extent do you think favourably or unfavourably towards the following political parties?”

Source: Survation/Labour Muslim Network poll

Nor is there much evidence for high levels of hostility to Labour or Keir Starmer among Muslim voters. Muslims have a strongly net positive view of the Labour party - nearly half give it a favourable rating and only around a quarter view it unfavourably. Keir Starmer’s net ratings are rather worse - around three in ten Muslims view him favourably, while four in ten give him a negative rating, for a net rating of -11. But this is not much different to ratings of Starmer among voters in general - for example the most recent IPSOS-MORI poll gave Starmer a rating of -13. This doesn’t suggest Muslims are unusually hostile to Starmer, and they remain strongly positive about Labour.

The Conservatives and Rishi Sunak are, by contrast, viewed very negatively by Muslim voters - just over one in ten view them positively, while around two thirds give unfavourable ratings, resulting in toxic net ratings of -50 for the Conservatives and -58 for Rishi Sunak.

“In the past 12 months, has your view of the following political parties become more favourable or more unfavourable?”

Source: Survation/Labour Muslim Network poll

Nor is there much evidence of a substantial deterioration in Muslim views of Labour in response to recent events. Survation did find that 38 per cent of Muslims reported their view of Labour had become more negative in the last 12 months. But this was balanced out by a nearly equal number of Muslim voters - 34 per cent - who reported becoming more favourable to Labour in the same period. And while there is in fact little evidence of a net decline in Labour’s image among Muslims, the same can’t be said for the Conservatives - over six in ten say their view of the party has become more unfavourable in the last year, while only one in ten say they have a more positive view of the Tories.⁵

Taken together, these measures give us substantial reason to doubt that the headline decline in Muslims’ current Labour support will hold up in a general election. While many Muslim voters are clearly deeply distressed by the situation in Gaza, and unhappy with Labour’s response to the Gaza crisis, they continue to prioritise domestic electoral issues where Labour are favoured, to hold positive views of Labour as a party, and to strongly prefer Labour to the Conservatives and Starmer to Sunak. It seems pretty plausible that many of those currently backing the Greens and Liberal Democrats as a way to express discontent over foreign policy will end up back in the Labour column in the general election to come, particularly in seats where a vote for smaller parties of the left will risk helping a Conservative party Muslim voters overwhelmingly reject.

Muslim voters in Rochdale and beyond

One major test of the fraying ties between Muslim voters and Labour will come in a few weeks, when voters in Rochdale elect a new MP in a by-election to replace veteran Labour MP Sir Tony Lloyd, who passed away in January.⁶ About one voter in three in Rochdale is Muslim, and Labour won the seat in 2019 with a twenty point majority. Based on the modelling above, Labour would be expected to win again with a substantial majority even if Muslim support fell sharply. But by-elections are tricky contests, where the rules of general elections sometimes don’t apply. Turnout is low, and the voters who do show up often treat them as a law stakes opportunity to make a point. By-elections can feature major shocks and massive swings, including the 2012 upset in Bradford West, when George Galloway took the seat for the Respect party on a massive 36 point swing. Galloway had bested Labour before, winning another heavily Muslim seat - Bethnal Green and Bow - for the Respect party at the 2005 general election. Could he score a hat trick in Rochdale, and how worried should Labour be if he does?

There are some reasons to doubt it. Rochdale has a smaller Muslim population than Bradford West or Bethnal Green, and Labour’s overall polling position today is stronger than it was in 2012 or in 2005. Galloway’s success in Bradford reflected skilful mobilisation of local discontents and internal tensions within the Muslim community. His Respect party then was a more effective campaigning organisation than the Workers Party he is standing for in Rochdale. Galloway’s most recently by-election outing was in Batley and Spen in 2021, where Labour prevailed despite a large Galloway vote, at a time when they were trailing the Conservatives nationally in the polls.

Galloway cannot be entirely counted out. He is a powerful orator and the current crisis in Gaza gives him a powerful cause to campaign on. But even if he wins in Rochdale, this does not necessarily presage a major Muslim swing away from Labour in the general election contest to come. The stakes in a general election will be higher, the focus on domestic issues will be stronger, and Galloway has a unique personal appeal which can’t be replicated across multiple seats. That appeal may not transfer to general elections either - Galloway lost Bradford West in the 2015 general election contest, and has flopped in his last two general election campaigns, losing his deposit in 2019 when he won just 1.4 per cent of the vote in West Bromwich East.

Whatever the overall outcome, a substantial vote for Galloway in Rochdale will add to Labour’s concerns about Muslim voters. The party is right to worry about falling popularity in this group, not because it will produces any immediate general election headaches, for as we have seen the electoral risk is minimal. But the job of political parties is not only to win elections. They exist to voice, represent and respond to the concerns of voters. Muslim voters have been strongly attached to the Labour party for generations, and representing minority groups who often face prejudice and hostility has been a core Labour value for just as long. If Muslim communities no longer feel Labour listens to their concerns, that is a failure of representation the party will need to address regardless of the next general election outcome.

1. Recalled vote measures like this are imperfect and subject to a number of biases. This measure may somewhat inflate Muslim support for Labour in 2019.

2. These notional results estimates are produced by Professors Colin Rallings and Michael Thrasher, using ward level local election results to estimate what would have happened if the 2019 general election had been contested on the new constituency boundaries. We cannot use general election results data to estimate outcomes like this because it is not reported at ward level.

3. This is a generous assumption, as many Muslim residents will either be too young to vote and some will migrants who lack voting rights in general elections (not that many though, as adult migrants from Pakistan, Bangladesh and other Commonwealth countries have full voting rights from the moment they settle in Britain).

4. This shouldn’t really surprise us - if Muslim voters backed Labour at rates of 80% plus then the Conservatives can win or build majorities in seats with large Muslim communities only if they manage outstanding performances among non-Muslim voters.

5. These figures do not in fact look massively different from the parties’ overall favourability ratings, so its not really clear these “change in favourability” questions are picking up anything distinct. At the very least, though, we can say there isn’t any evidence here of Muslims reporting large scale alienation from Labour in response to recent events.

6. The by-election will take place on 29th February, two weeks after by-elections in two Conservative held seats - Wellingborough and Kingswood.