The prevailing view about immigration among coalition policymakers goes something like this: Britain has far too many immigrants, and we must take drastic measures to reduce immigration, even if these hurt the economy, because this is what the public demands. British voters are fearful of the economic competition from migration, and will only be satisfied by drastic cuts in net migration levels.  Those holding to the conventional wisdom point to polling showing that 70 per cent or more Britons believe there are too many immigrants in Britain, and that immigration remains one of the most important problems in the minds of many voters.

Yet this view is dangerously dependent on a very simplistic reading of public opinion. Take the belief that Britain has too many migrants. It is true that MORI's most recent poll shows 70 per cent of Britons hold to this view, but 63 per cent also agreed with this claim in 1989, at a time when net migration was negative and Britain's migrant population was orders of magnitude lower than it is today. Research by the Migration Observatory shows that public views about migration levels have been stable over twenty years, despite massive shifts in migration levels over this period.

We are told by those holding to the conventional wisdom that British voters worry about competition from migrants for jobs, even more so in the current difficult economic circumstances, and that they will only be satisfied by sharp cuts in net migration. Yet despite a severe recession and rising unemployment, the proportion of voters naming immigration as an important problem facing the nation has fallen continuously for five years, from 38 per cent in 2007 to 22 per cent in 2012. The coalition cannot claim any credit for this: the trend predates their policy of a "cap" on migration, which has in any event completely failed to reduce net migration levels.

This shows how misleading a narrow reading of public opinion can be: the British agree there are too many immigrants, but they have always agreed with this regardless of how many migrants are actually coming in. They say they worry about competition from migrants, yet they are less worried about immigration now than they were at the height of the last boom. They say they want migration levels reduced sharply, yet their concerns about migration have declined in the last five years even as migration levels have remained consistently high.

In a new Transatlantic Trends analysis, I build up a broader and more nuanced portrait of British opinion on migration, based on four years of detailed survey data. It contains many important lessons for policymakers. Firstly, while concern about migration is high in Britain, this concern is not evenly spread across the population. There is a large generation gap, which is reinforced by differences in the education levels and social diversity between age groups. Older Britons, who are on average also less educated and less likely to have any migrant heritage, are much more negative about migration than the young, who are more educated on average and more diverse.

The evidence here speaks against a simple fear of migrant competition driving negative reactions. The most negative views are held by pensioners with little formal schooling, who face no competition from migrants for jobs, and as are indeed most likely to benefit from migrants' contributions of tax to support the pension system and labour to support public services such as the NHS. Here is an interesting paradox: those who have the most to gain economically from migration are the most opposed to it. This shows that these concerns are most likely not driven by everyday personal experience, but by a general perception that British migration is not being well-managed.

Current government policy, which calls for a reduction in overall net migration, rests on the assumption that the public is only concerned about migrants as an undifferentiated mass. When your explicit goal is to get the numbers down, turning away a cancer specialist recruited to fill a senior NHS post is just the same as turning away an unskilled labourer coming to look for work. Yet when we break down public attitudes more carefully, we see that voters do not see this issue this way at all. The Transatlantic Trends data show that voters are highly sensitive to differences between migrants. Majorities of voters support bringing in highly skilled migrants, and large majorities support the recruitment of migrants to support the NHS and the provision of elderly care. On the other hand, British voters are very concerned about illegal migration and support draconian measures to ensure it is controlled.

A closer look at public opinion on migration shows the current regime needs to change. The logic of the "cap" in overall migration puts pressure on policymakers to reduce forms of migration they can most easily control, such as highly skilled migrants and students from outside the EU. Yet the evidence shows that voters recognise the contribution such migrants make to Britain, and support admitting them as a result. We need to move beyond a simplistic fixation with overall numbers, which is based in turn on a simplistic stereotype of public opinion as monolithically opposed to all migrants. Policy should focus on reassuring voters that their real concerns, for example about illegal migration and pressures on public services, are being addressed, while taking a more supportive stance towards forms of migration the public support. Demonstrating that the immigration and border control system is competently run - something the present government has dismally failed to do - would probably do much more to assuage public concern than turning away thousands of professionals and students, or booting out settled workers whose incomes fall below £35,000. Voters in Britain recognise that migration is complex, and comes with many benefits as well as costs. It is about time politicians did the same.