Chris Murray is a research fellow on migration at IPPR. 

This week, the Prime Minister is touring India to drum up trade and open new markets. As she does, she is coming under pressure to relax restrictions on Indian students coming to study in Britain, whose numbers have plummeted from 68,000 in 2010 to 11,000 in 2015. Yet she seems not to budge. She has long been consistent that migration must come down, and cites numbers that tell her student migration must come down, too.

Even May’s critics concede that consistency is one of her strengths. Since 2010, she has had a target of reducing net migration to the tens of thousands. Over six years, despite struggling to meet it and ample opportunities to change course, she has stuck to that numerical objective. She set out her reasoning to the Conservative Party conference last year, stating that “when immigration is too high, and the pace of change is too fast, it’s impossible to build a cohesive society”. You may take issue with the analysis. But you cannot fault the Prime Minister for consistency in insisting that the numbers matter.

The Office for National Statistics (ONS) estimates that international students contribute some 90,000 to the net migration figures every year, as students come to study in Britain and then don’t leave at the end of their course. New evidence suggests that number is way off.

It was always doubtful that tens of thousands of highly qualified graduates who had fully complied with tough immigration rules throughout their courses were vanishing into the black economy every year. New analysis by IPPR compared the ONS estimates with other sources, such as estimates of former students in the general population and surveys of where students end up. We found the ONS statistics to be significant outliers, and discovered problems with the ONS’s methods.

Our conclusions were backed up a few weeks ago by an exclusive story in The Times. This suggested the Home Office’s internal data showed a massive disparity with the ONS data. When May was Home Secretary, she ordered the Home Office to start collecting ‘exit check’ data on when people left the country, as well as when they arrived. According to the Times story, these numbers suggest only around one per cent of international students overstay.

Last weekend, the former home affairs adviser to David Cameron revealed in the Telegraph that Number 10 have long been aware that the statistics are dubious, and that they are a poor foundation for government policy. He suggested that Cameron was planning to remove students from the net migration target if he had stayed as Prime Minister.

That puts May and her successor at Marsham Street, Amber Rudd, in a bind. Taking students out the target might be politically difficult to do. There may be accusations of moving the goalposts. After all, some may argue that if the Government announced that net migration was actually much lower than we thought, the British public wouldn’t suddenly feel much more comfortable with immigration.

But it is for precisely this reason that chasing phantom students is poor policy. Ultimately, it will not resolve the real causes of people’s concerns over immigration. It is a bad use of political capital and taxpayers’ money. And as the evidence mounts, it is an increasingly untenable position.

This Government says it takes immigration numbers seriously – so it needs to show equal discipline about getting immigration numbers right.

Firstly, the Home Secretary should institute a wide-ranging review into the exit check data, to see if it tallies with the ONS estimates. Secondly, she should refer the issue to the Migration Advisory Committee, the independent body of experts who advise the government on migration policy. Thirdly, the ONS data should be scrutinised by the UK Statistics Authority, who should be directed to produce a report on this issue. Fourthly, the chair of the House of Commons’ Public Administration Committee, Bernard Jenkin , should summon the ONS and the UK Statistics Authority to explain why the government’s statistics are at such variance with everyone else’s.

Before any of these steps are taken, the Home Secretary should be clear that she will accept the outcome whatever it may be. Whether the net migration figures go up or down, the Government must pledge to accept them if the evidence points that way. Committing to the results of such a review allows the Government to get ahead of any accusations of moving the goalposts.

It will also help draw the distinction between international students and other types of immigration. Surveys consistently show that public attitudes to immigration are not motivated by international students. There must be a clear explanation that it is this specific category that triggers any revision, to show that government policy remains focused on the issues that tally with people’s experience.

Public concerns on migration will only subside when the cause of people’s concern is tackled. May has been consistent that she thinks numbers are the cause. The Government must therefore ensure that it is getting the numbers right. Because chasing erroneous groups of migrants who don’t exist will self-evidently do nothing to allay the public’s concerns.