Sunder Katwala is the Director of British Future.

Immigration was central to the Brexit referendum debate of 2016. Without the loss of public confidence in the handling of migration from Europe over the previous decade, there would have been less pressure on David Cameron to hold a referendum – and it would have been less likely that the knife-edge campaign would result in a narrow victory for the Leave campaign.

The opposing sides saw the issue of EU free movement differently, yet the lessons of the result for immigration were contested within as well as across the rival campaigns.

Many on the losing side saw the result as evidence of an irrational upsurge of nativist sentiment, while Leave advocates took different views. For some, the referendum was a mandate to cut immigration sharply, to demonstrate that the message had been heard. Others saw the referendum outcome as a rational assertion of control that might come to rebuild public confidence in the contribution that immigration can make to Britain.

Five years on, there have been significant shifts in attitudes to migration. The latest 2021 wave of an in-depth tracker project from Ipsos MORI, published by British Future today, shows that a long-term softening of public attitudes has continued during the pandemic.

This detailed evidence of how public views changed supports the argument of those who argued that control was not simply a question of how low the immigration numbers could go. The public are more likely to see the contribution of immigration as positive (46 per cent) than negative (28 per cent), overall. This is a direct reversal of the position in 2015-16.

Asked to prioritise, control (44 per cent) is chosen over reducing numbers (24 per cent), with another fifth of respondents choosing neither of these as a priority. The proportion of the public wanting to see immigration reduced overall is now at 45 per cent, its lowest level in this series, or in British Social Attitudes surveys over a much longer period.

A similar proportion is content for numbers to remain at current levels (29 per cent) or increase (17 per cent). That 46 per cent of people would like to reduce migration shows that numbers will continue to be part of the debate. Conservative voters are more likely to prefer overall reductions. But many reducers are selective balancers: 17 per cent would like to reduce immigration “a little” while 28 per cent would hope to see larger reductions.

Because these are not ‘one size fits all’ views, there are much broader public majorities for choices that would increase migration, with two-thirds support for the government’s offer of a new visa route for people from Hong Kong. Having ended free movement, Boris Johnson’s Conservative government has tended to make liberal choices on student and post-study visas, and on non-EU migration for work.

The attitudes evidence does not tell ministers what the right policy mix is to deal with labour and skills shortages – whether of lorry drivers, construction workers or fruit-pickers – in the short, medium and longer-term. What it does seem to suggest is that a government which chose to blend domestic training with flexibility in the points system where there are key gaps would be able to secure pragmatic permission from the British public.

So the immigration policy debate is not primarily about numbers, but about the choices that Britain makes, and what we do to make them work. If the post-Brexit debate has primarily been about who gets a visa to work in Britain, policy needs to focus more on what happens next. The Government has taken more proactive initial steps on Hong Kong than any previous wave of migration, which could be foundation for a more positive approach to citizenship and integration more broadly.

But if shifting attitudes create an opportunity for more light and less heat when we talk about immigration, significant challenges remain. There is still low trust in government on immigration – a perspective shared by broad majorities of those with liberal, restrictionist and balancer views, almost certainly for a range of contrasting reasons.

The two major parties both need to engage the balancer middle, but will often strike those balances differently, reflecting distinct electoral coalitions. Attitudes towards asylum are more polarised, though again the balance of attitudes has shifted. By a small margin there is public sympathy, rather than no sympathy, for those crossing the Channel in small boats – though nobody on any side of the debate would see the images of dangerous crossings as exemplifying a well-managed migration or asylum system.

The findings suggest that a debate about “control” versus “compassion” will produce a deadlocked stand-off, with a quarter to a third of the public on each side of a polarised argument. The key to securing the balancer majority on refugee issues is not to increase the temperature of the debate, especially if headlines over-promise and under-deliver – but to marry control, compassion and competence.

That means investing in an effective asylum system at home, cooperating with France over Channel crossings, and forging a multilateral response to those fleeing Afghanistan. Civic society critics of this government face a parallel challenge – to engage both liberals and balancers to unlock broad public support for a managed system of asylum that is effective, fair and humane. That could involve entrenching the majority support for Britain’s contribution to Afghan resettlement, and making the argument that all asylum seekers should have a fair hearing for their case, however they arrived in the UK.

How far these long-term shifts in immigration attitudes are now reflected in a new political and policy debate will depend on how the public debate is led. But politicians may need to steer a course that runs with the current of public opinion now, in 2021 – not that of a decade before.