I welcome our witnesses from the Migration Advisory Committee to Home Affairs Committee. We are very pleased that you have come for your first visit to the Committee in this Parliament. We have lots of questions for you, both to help with the work we are doing now on our inquiry on settled status and our future work on legal migration. Would the witnesses like to introduce themselves?

I am Brian Bell, chair of the Migration Advisory Committee and professor of economics at King's College London.

I am Madeleine Sumption, deputy chair of the Migration Advisory Committee. I am also the director of the Migration Observatory at the University of Oxford.

Thank you very much. We will start with big picture questions on net migration.

Thank you for coming today, we really appreciate your time. I want to begin by focusing on the headline net migration numbers. Is it important to make distinctions between different migration routes—skilled work or students—rather than treating them as an overall net migration figure?

Let me begin with the headline: net migration has fallen from its peak in June 2023 at 906,000 to the most recent figure of 431,000 in December 2024. It has essentially fallen as fast as it rose, and my expectation is that it will fall when the next numbers are published at the end of this month. We expect it to gradually tail off from here. That fall has been almost entirely because we have reduced immigration. There is some increase in the number of people emigrating which helps, but mostly the reduction has been because of immigration changes—mainly policy changes by the previous Government, which were extended by this Government. The reduction is across the board—in work, student and humanitarian routes. The only two routes that have not really changed much are the asylum and family routes, which have stayed broadly constant. In terms of the importance of the distinction there are two things to say. First, one reason it is not important to look across different routes is that they are all people. If your concern is the number of people in the country and the size of the population, it does not matter whether someone is coming through a student, worker or humanitarian route—they are all people and they all count. Things such as congestion and pressure on housing may differ between those routes, but there will be those pressures regardless, because there are just more people, and more people put pressure on services. However, there is a clear difference if you are thinking more about economic and fiscal contribution, where there will be a very significant difference between reducing work migration to get net migration down or, for example, reducing asylum. You will get very different outcomes.

I agree with that. I would add that the impacts you are looking at will tell you more about whether you want to think about the overall numbers or the specific routes. In many cases it is really the specific routes that are important. If you look at the fiscal impact of migration, for example, those impacts are dramatically different. You have some people, such as skilled workers, who are very much fiscally positive, then you have other groups, such as family members and refugees, who are fiscally negative. It does not really make sense to talk about the fiscal impact of migration as if it is one monolithic category. The decisions are also different. If you say you want to bring down net migration from x to y, there is no single policy lever that says you can bring down net migration; you have to make decisions about family members, students, workers and so on.

If the public debate focuses on the headline net migration, do you think there is a risk in terms of misinformation and how the debate is handled in the press and the media?

I do not think there is anything wrong with looking at the overall figures, because that does have some relevance. As Brian said, there are particular areas like housing where it is a question of the number of people, more or less. I think it is important to make sure that we also discuss the specific routes and the impacts of the different routes, which can be dramatically different. It is important that people understand that when you are talking about reducing migration, we often see that the groups that people want to reduce are not necessarily the biggest ones, so there is sometimes a trade-off between the idea of reducing overall migration and potentially a desire to maintain similar levels of more economically beneficial migration, for example. You have to be able to separate the routes to be able to have those conversations.

You mentioned that the policies of the previous Government, continued by this Government, are what have driven the declining numbers. Do you think that you can predict quite well the impact of particular Government policies on the overall numbers, or do you think that is quite a tricky exercise?

You can do a reasonable job at predicting, on the route that you are changing, what the impact of the changes will be. For example, perhaps the easiest one to do was when the previous Government restricted the ability of the students who are doing taught master’s programmes to bring dependants. That was a fairly easy calculation—we just looked at how many people came the previous year with dependants on those courses. You have to assume how many people would still apply, given the changes that occurred, but you can make sensible predictions on that. Those kinds of predictions, which are often in impact assessment studies by the Home Office, are broadly right. What you cannot do is then aggregate up and say, “And that will be the effect on overall net migration”, because net migration is not just driven by policy; it is driven by behavioural changes, which are sometimes policy-driven, but sometimes not. For example, in the last year and a half, there has been a 74,000 reduction in immigration on the humanitarian routes. That is really nothing to do with Government policy, because the routes have not really changed; it is that we have already seen the big numbers coming on the Hong Kong route and now we are getting into much smaller numbers on an annual basis. Again, most people arrived from Ukraine a number of years ago. There are those kinds of macro effects going at the same time. Another example is students. Student numbers have come down—some of that is definitely because of Government policy, but it is also because Chinese graduate unemployment rates are high, so coming and doing a degree is less attractive now. Nigeria went through a currency crisis, so students were less able to come. You cannot isolate the two effects easily.

It is also particularly difficult to predict the impact of liberalisations. If you have a restriction, you can look at who was previously coming and say, “Okay, these people look like they would no longer be eligible.” If you have a new policy that has opened up—as we saw with the care sector in 2022-23, for example—it is impossible to predict in advance how many people are going to use a route that did not previously exist.

Turning now to the regional elements of it, how well do you think that we can align the different needs of different parts of the UK on migration policy? I am thinking particularly from a Scottish perspective, where the demographic pressures are more acute—we have a more acute skills shortage. Do you think a one-size-fits-all migration policy might not adequately support different regions of the UK? If not, why not? What would your suggestion be?

As you say, most immigration policy is essentially UK-wide as a reserved matter. One interesting thing to note is that it is often thought that the work route—one of the main routes into the UK—is heavily London-centric, because something like 60% of all workers who come on the skilled worker route work in London. It is important to remember than when freedom of movement existed, and EU citizens could go where they want, about 60% of them went to London. It is wrong to think of the system as deliberately pushing people to one place; often it is the choice they make as to where they want to work. In Scotland in particular, one of the things that we often hear about is the depopulating of more remote areas. In some senses, one needs to answer the question, why are they depopulating now? Just sending migrants there will not necessarily be a solution, because as soon as they get indefinite leave to remain, they will do the same thing that Scottish people do. You are right, though—we have highlighted this on a number of occasions—that Scotland in particular has a different demographic profile, although probably less than 15 or 20 years ago, when the differences were quite dramatic. Actually, if we do not have net migration going forward, the population of the entire United Kingdom will fall over time. That demographic issue is occurring throughout the UK going forward. There will probably be more focus broadly about whether migration should focus on the demographic challenges that are faced. Of course, it is up to the Government to decide whether they want a regional migration policy. Lots of countries have rules that say, “You get extra points in a points-based system if you go to certain areas.” In Australia, you get more points if you do not go to Syndey. You could have that in the UK if the Government wanted to.

Do you have any thoughts on whether that would work for the UK?

There is no reason that it would not work. I think some of the evidence is that, again, it works for the period in which you are restricted by your visa to be in those areas, until you loosen that. It should be pointed out that Scotland, as I understand, has been gaining population from the rest of the UK in the past few years, so internal migration is going in the other direction. It may not be as dramatic as we think.

We often see that the differences in the economic profiles in different parts of the UK are more significant within regions than between them—for example, between rural areas and cities in Scotland, but also in the rest of the UK, which makes it quite difficult to tailor immigration policy. There is a cost of having more complicated immigration policies, particularly if you are looking at work visas and you have an employer with an office in different parts of the UK. If they hire someone on a work visa in Scotland, do they then have to start making sure that that person does not go to too many meetings in London? There is that complexity, which means that in the business community not all but some companies giving evidence to the MAC have said that they do not really want regional variation. There are other companies who do, particularly on salary thresholds—they would like to see lower salary thresholds in particular areas. But I would say that there are mixed views on the benefits to business of having variation for different parts of the UK.

As the representative of central Edinburgh, I can tell you that we have a growing population from bits of Scotland, the UK and internationally—that bears out in my experience. I want to pick up on Joani’s point about the different levers that Government have to reduce migration. You said that net migration should probably taper off in the next couple of years. If the Government wanted to continue the drop in net migration and not let it taper off, what would your advice be on which routes should be the ones to focus on? What kind of restrictions or policy changes would you recommend putting in place to keep driving that down? Finally, what would the wider implications of that be for the economy?

This is really a political question in terms of where the priorities of the Government lie. There are different constraints and considerations, and this is one area where it is really important to look at the individual routes and what they are doing. Successive Governments have wanted to reduce the number of people coming through the asylum system. That would be economically beneficial if it happened, but it is very difficult to do, for reasons that we do not need to go into here. Similarly, for the family route it seems like there would not be particularly significant economic consequences. There could probably even be fiscal benefits if we do see family migration of people joining British citizens in the UK, but again, people have expectations that they will be able to live with their partners, so there is a trade-off there. On the work and study routes, I think with work visas we are now more or less at a point where your average person coming in on the work route is bringing significant economic benefits, although I should not really talk about the average person. In the past, there was a tail of lower-earning people whose access to visas could be restricted with very minimal economic consequences. That is less the case now that we have more restrictive middle-skilled work visas and now that the care route is closed. I would say that it is more challenging there, but if the Government want a category of migration that is very easy to restrict—you change the policy and the numbers do go down—the route is obviously one option that is open.

That is the basic problem that the Government face: the routes that they can more easily control are the ones that are more economically beneficial. Essentially, they can do whatever they want on work and students—there are no legal restrictions really on what they can do—whereas on family, asylum and, to an extent, humanitarian, although that is obviously a separate decision, they are restricted in what they can do. You would probably like to reduce the routes that are more costly for the UK, which are those ones, but that is hard to do, so most of the levers that have been pulled—certainly by the previous Government—were on the work and student routes. I think you could probably do more on the student route. There is still a question in my mind as to what the graduate route’s purpose is. If it is a route to allow graduates of UK universities from abroad to get into graduate-level jobs, it is not well designed for that purpose. If it is a route to allow people to spend 18 months in the UK doing whatever they fancy, it is a very well designed route. I am not sure that is necessarily the best use of the immigration system. On the work visa side, fundamentally we are at a point where further restrictions would really be about restricting graduate-level jobs, all of which are essentially fiscally positive. Probably the focus there, and the focus of this Government, is primarily about trying to increase the training of the domestic workforce. Probably the national health service is the area where it is most pressing. If we can train more doctors and nurses, you will get net migration down. That is a double win in some sense for the UK.

I am Peter Prinsley, the MP for Bury St Edmunds and Stowmarket. I am also a surgeon from Norfolk, and I am very interested in aspects of migration as it affects the health service, particularly medical training. Perhaps we can come on to that, but first I have a question about the relationship between a skills policy and immigration. Do you think that if it really were the case that we could increase investment in skills in this country, we would in fact bring immigration down?

There is no simple one-to-one relationship; that is absolutely the case. Do I think that part of our problem is that, particularly for mid-skill level jobs, we do not train enough people in this country to do them? Absolutely. The example that everyone uses at the moment is welders, just because it is a particularly extreme example. We train almost no welders in this country, given the demand. We are trying to build Hinkley Point and Sizewell. You really want good welders there, and you can earn £70,000 as a welder. We have a million kids who are not in employment, education and training. It should not be beyond our abilities to think of a system of training that gets those people to be trained to be welders, and into great paying jobs. I do not think that I have met an employer who says, “I desperately want a foreign welder.” They have to get them to do the job. At the higher end of the skills distribution, if you go to Goldman Sachs, you probably get a slightly different answer. They are saying, “We are actually looking for global talent, and that will not necessarily be in the UK.” Therefore there is a difference there. The focus of the Government is the mid-skill level jobs, and that is exactly the right focus. The one thing that has always been quite depressing in the MAC is the number of times we write reports about shortage, and it is always the same occupations that are in shortage, where we do not train enough. Up until recently, that never went anywhere in Government. It was almost like a separate discussion. “Why should we link training to that? Why should we link skills?” The Government are at least trying to link those two things.

I am sceptical that investing in training on its own would naturally lead to a reduction in migration. It might in some circumstances, but because there is no fixed number of people that we need in all these jobs, it might be that if we trained more engineers, we would still be recruiting engineers from overseas, and we would just do more engineering—which is not necessarily a bad thing. If the Government want immigration to fall, they have to restrict it rather than assuming that skills policy is going to be able to do that work. In some ways, I would think of the relationship the other way round. It is useful to think about training and immigration together in the sense that, if you are going to restrict immigration, having more trained people in the UK would mitigate any negative impacts of that, because there would be other people willing to do those jobs, rather than saying that it is going to induce a natural decline in immigration without having to make any restrictions.

Let me describe the situation in the hospitals. We have about 9,000 graduates of the UK medical schools, and about 9,000 higher professional training slots for people who wish to be surgeons, physicians, general practitioners or whatever. But we recruit about 13,000 to 14,000 international medical graduates into the second and third years, in order to fill the rotas within the hospitals, which means we have now created a massive bottleneck. We cannot get our own graduates naturally into the higher professional training programmes, and large numbers of them go to Australia. We have a situation where the unregulated immigration of international medical graduates has distorted the whole system of postgraduate training for doctors in this country, and that is a big part of the dispute that is going on right now. What do you think about specifically how we might control the numbers of doctors arriving in this country? Is that something that could be done?

It can be done. As I understood it, the Health Secretary said that he is in favour of doing something along those lines and trying to create a presumption that graduates from British universities will be first in line for those training posts. That asks a broader question, though, which a number of employers have spoken to us about. Quite a lot of employers are unclear—and I am completely unclear, because I am not a lawyer—what the employment law implications are if you are a sponsor of workers. Are you allowed to refuse to sponsor a foreign worker who is either the best candidate, or one of the best, because you have decided you would prefer to hire a British worker? It is unclear to me whether you are. Some clarity on that would be good, and personally I do not see any objection to having a rule that says that firms are allowed to discriminate in favour of domestic workers, since they do not have to pay fees for them and the like. The other thing I would say is that of course it will not solve all the problems. One of the other reasons why people go to Australia to be doctors is that Australia pays people a lot more to be doctors than in the UK. That is something we cannot solve—well, we can solve it, but it would be somewhat expensive for the Chancellor.

Moving on to the situation with care workers, I am hearing that, because of the fall in the number of care workers coming in from other countries, the care homes are struggling to continue their activities. What is your view on that? Was the instruction to forbid the recruitment of overseas care workers the wrong thing to do?

The first thing to say is that the decline in care workers that we have seen is not the result of the Government’s decision, because the Government have only just legislated for that decision. More important was the decision by the previous Government to restrict care workers from bringing dependants. That has clearly had a big effect on the numbers. There is a real challenge here, because the problem with letting care workers use the skilled worker route—that was the only way it could have been implemented fast after the pandemic; it is not as though there were a bright alternative that could have been quickly implemented—is that it has led to exploitation of care workers. We have seen a lot of stories about that exploitation. There is the added problem that the care sector operates a very odd contracting model that means they will often essentially recruit workers before they know where is work for them. You are not allowed to do that under the immigration rules, but people were doing it, so that has now been dealt with to a large extent. Fundamentally, you have to answer the following question: if you do not want to bring in international workers to work on minimum wage work in care, you have to find domestic workers to do it. If you do not, all you will do is ration care. That is the model of the previous 20 or 30 years of Government, whether they have admitted it or not—they just want to ration care. They will spend a certain amount, and some people just will not get care. Bringing in international care workers relieved that constraint and allowed some firms to expand and provide more care. If we cut that off, we will be back to rationing care. That is inevitable, unless we move to a situation where care workers are paid a more attractive wage. The problem with that is that paying them even £1 over minimum wage per hour, which would be a significant increase for most care workers, pays them about the same that Aldi is paying people to stack shelves. That is just a fundamental problem with the model. Unless we are willing to spend a lot more on care and to invest in care workers, we will have the fundamental challenge of either restricting care for a lot of people, or relying on international recruitment.

There is a sort of political mood that immigration is terrible and something must be done to reduce it. I do not think that we are sufficiently clear with the public about the trade-offs. A trade-off that occurs to me is that your elderly relatives might have nobody to look after them. I think that is a story that we ought to be able to tell.

Earlier this year the Government announced the youth mobility scheme jointly with the EU. The cynic in me sees this as a way of filling our seasonal vacancies on the farms and in our tourist hotspots. The scheme is meant to be an exchange, but how can we ensure that it is a true exchange programme so that it does not impact on our net migration figures?

Part of it comes down to the negotiation with the EU, what the structure of the scheme is and what the numbers are in each direction. If you had a scheme where 200,000 people came this way and only 50,000 went the other way, in the short run it would push net migration up. If you have something that has more balanced numbers, especially if the overall numbers are not very high—obviously there is a limit; our population is much smaller than the population of young people in the EU. The lower the cap, assuming we are talking about a capped scheme, the more likely it is that British people would fill those places. There is a separate techie issue about whether you are interested in the impact on actual net migration, which might end up being zero if you had 50,000 people going that way and 50,000 people coming in the other direction for a couple of years, and most of them then go back home afterwards. Some will switch to family routes or get other kinds of work visas. In theory, you could imagine a system with no impact on net migration. It might not look that way in the net migration statistics, because we are not very good at measuring migration of British people. This is a big problem that the ONS faces currently. The confidence in the measurement of British nationals is much lower than in the other. So it might be that we have more measurable people coming in and unmeasurable people leaving. It therefore looks like you have a positive impact on net migration, even though you do not.

Is that going to be fixed by e-gates, ID and things like that?

Not in the short run. It is very difficult. There are so many British citizens coming in and out. They do not have visas. The way that we measure migration of non-citizens is effectively based on their visa status. You have lots of issues around dual citizens. Once people naturalise, they may leave and arrive on different passports and that creates a lot of error in the system. The Office for National Statistics is going to publish some revised statistics in November, and we will propose a new methodology for doing that. But I think the British side of things is going to be tricky for a while. We may end up, at least in the medium term, being reliant on looking at what happens in the census and then having to revise the back series if it turned out that the methods are not working for the British.

On the youth mobility scheme, the other reason you will see some impact is that there is no guarantee people will go to the same places. You would expect, I think reasonably, that the Europeans who come here will primarily go to London, whereas you might expect a broader flow in the other direction. You would probably get that concentration. The sector that really wants the youth mobility scheme is the hospitality sector, for obvious reasons. Seasonal work in agriculture will be less because there is a seasonal worker scheme, anyway.

On social care and the costs of it, is there any evidence around the impact on the cost of goods and services of using migrant workers as opposed to domestic workers?

In the UK I am not really aware of any. We did a 2018 report with a small section on prices where we said it was very hard to estimate. There is some evidence in the US where they look across areas. They have good price data across cities in the US. They look at where cities got more low-skilled migration and what happened to prices. And there is some evidence there that service prices go down. So there is some benefit in terms of that. The other paper that I quite like from the US looked at what happens when you get more low-skilled migration in an area. One of the impacts is that higher-skilled women do more work. The reason for that is that childcare costs go down, so they are able to work longer.

My question is about temporary shortage lists and skill shortages. What do you think about the Government avoiding temporary migration routes? Is there a danger that they get used indefinitely?

We are currently in the middle of reviewing the temporary shortage list for the Government. We will give a final report in July next year. It depends what you mean by the word “temporary” and what the Government mean by the word “temporary”. The one thing they are very clear on—this is for mid-skill level jobs only—is that any occupation that gets on the list and gets access to the skilled worker route will be temporary in the sense that they do not expect it to stay on in the long run. I am not revealing anything by saying that welders have a pretty good chance of being on the temporary shortage list when we publish next year. The idea is that we do not want to see welders on the list in 10 years’ time. If we do, it means that we failed to train enough welders. That is the sense in which the Government mean “temporary”. That makes enormous sense. They try to put some pressure on the sector to do the right thing in terms of training and improving terms and conditions. However, “temporary” could also be meant in the sense that the visa is for only a short period of time. The Government have not been clear about that yet. In the interim report that we gave to the Home Secretary at the end of last month, we said that you could see an argument for these sorts of jobs to be more short term—perhaps a three to five-year visa without a route to settlement. The Government may want to do that, but they may want to make it part of the normal skilled worker route that would allow you to settle. One reason that I suspect the Government may go down the route of not allowing settlement and making it a fixed-term visa is that they have already decided that people on that route will not be able to bring dependants. It would be slightly odd to say that you cannot bring dependants, but you can stay here. Obviously, eventually you could bring your dependants, but my suspicion is that we will move to a model for those middle-skill level jobs where the jobs will be on only for a period to get the training up to standard in the UK and the workers who come will come for three to five years and not get entitlement to any further leave.

Following on from your point earlier, businesses in my seat of Dagenham and Rainham would want the welders from around the area. It is about the skills piece, isn’t it?

I have to say that employers have to take some responsibility here. The spending on training per worker has gone down and down in the UK over the last 15 years. It has not done that in Europe. There is a basic problem here about our long-term investment. That is true of physical capital as well. Firms just are not investing enough in the skills of the country.

To add to the debate about whether people should be on temporary visas and have to leave, I think there are pros and cons of both options. The fiscal impact of people on temporary visas, especially if they are lower or middle-paid workers, will be much more positive if they are temporary, because a lot of the costs come later in people’s lives. On the other hand, there is more churn in communities, which may have a social impact, and employers lose the investment that they make in people. Some of those things really matter if it is a really big route. If you are worried about the impacts on communities, 10,000 people coming in every year is not going to make that much difference, but if you had a really massive temporary programme, you might start to see some of the social impacts of people not being integrated because everyone has been here for only a year or two.

How do you think you are going to guide the Government on assessing which occupations should go on the lists?

We have a list of 82 occupations. About half the mid-skill level occupations have passed our stage 1 test and are now being considered for inclusion. The stage 1 test was essentially, “Are those occupations important for the industrial strategy or building critical infrastructure?” That was how the White Paper defined it. To be clear, there is a collection of occupations that have no chance of ever being on the temporary shortage list because they are not important to the industrial strategy.

Could you give us an example?

Chefs. Hospitality does not really get a look in. That is deliberate. There is some credit to give here in that the Government are at least saying, “We are putting our money on these sectors. These are the ones we want to grow. If we want to grow the economy, let’s invest in the high-productivity, high-wage and high-skill sectors.” That is very sensible. The MAC is currently chairing the labour market evidence group, which is made up of ourselves, Skills England, equivalent skills bodies from the devolved nations, the Department for Work and Pensions and the Industrial Strategy Advisory Council. We all get together and think about these issues. One of the key roles of that group, in the first half of next year, will be to assess submissions from Government Departments about the occupations that they think should be on the list. Those submissions will be called jobs plans, and they will have to be provided to us. They will have to provide evidence of their plans for skills training, on how many more apprenticeships they are planning, on whether that will fill the gap—and if not, why it will not—and on where the problems are. We will use that group to assess how convincing those submissions are. My suspicion is that quite a lot of occupations will fail that test, but many will get through. Those will be the ones put on the list. The key thing from the Government’s perspective is that they do not want occupations that say, “We’re not going to put any effort in.” They will not get on the list and then they will be forced to put the effort in because they will not be able to recruit internationally.

Are the Scottish Government involved in the group that you mentioned, as skills is devolved?

Yes, we have representatives of Scotland, Wales and Northern Ireland on the skills side.

Does that work successfully?

The Scottish Government tell me so.

I want to ask some questions about the role of the MAC but, to stick on the topic of the labour market evidence group, can you talk me through how your teeth are going to work? Will it just be that sectors that do not submit a good enough plan do not get to recruit, so they must improve their plan before they can recruit? I ask because we talked about the social care visa. Do you agree with my analysis of that visa? After the pandemic, the social care sector said to the Conservative Government, “We need more staff. We will do whatever it takes. We will train domestic workers and do everything we promise.” The Government said, “Okay, we will open this visa route for you.” The sector took the international recruitment but did nothing domestically. Is that a fair analysis of what happened? How are you going to stop that happening with your jobs plans?

I am not 100% sure that the social care sector ever said anything about what it was going to do. One of the problems with the social care sector is that because it is such a diverse group of employers, it is hard to get them together. That would have been ideal; I am not sure that is how it was. To be clear, the jobs plans are going to be produced by the sector-owning Departments of Government. For example, the DBT will produce the plan for the advanced manufacturing industrial strategy sector, with the involvement of sector representatives and employers. The key thing is that we are not expecting some miracle plan that will solve all problems. What we are asking for is quite straightforward. We are saying, “We want you to tell us, using published data that is already available, how many welders you think the UK will need over the next five to 10 years. How many are you training? If there is a big gap between those two figures, what are you doing about it?” If the answer is, “We’re doing nothing about it,” that might still be okay if the reason they are doing nothing is that there is a problem in further education—as there probably is, in welding—that means you cannot put on more courses. One of the problems for welding is that you get paid £70,000 to be a welder; you get paid about £30,000 to be an FE teacher. It is quite hard to get people to agree to teach welding in further education colleges, and the fees that the colleges are allowed to charge do not cover the costs; it is a bit like the university sector. It may well be that the Department comes back to us and says, “We are not going to be able to train enough welders over the next five years, and unless there is a change to the funding model of further education, that is not going to change.” In that circumstance, our view would probably be, “Yes, you get on the list, because it’s not your problem; it’s the Government’s problem.” We would put it back to the Government, being very clear about that, and say, “Unless you solve that problem, you will have to have immigration, and that’s your choice.” Whereas there will be some sectors where that is not a problem, and it is just that they do not put in the effort. That is what we are going to look for. Of course, there are lots of occupations in the list that will not be in shortage. For example, one occupation that has got through to the next round is human resources officers. I do not feel that there is a desperate shortage of human resources officers in the UK; there might be, and we will find out when the evidence is presented, but I think a lot of occupations will fail just because you do not get on the list if there is no shortage—you do not need immigration.

That is super interesting. I get that this is a redesigning of the architecture of Government for how these decisions are arrived at, but can you say with any certainty, at this stage, whether it will have a meaningful output? If enough sectors come to you with credible reasons why they cannot train domestically and you say, “Cool, you go through to the temporary shortage list,” and you turn to Government and say, “These ones you will have to recruit internationally,” it might end up in exactly the same place. Is this just architecture?

It may well do. In some sense, the Government—possibly inadvertently—are going to make themselves more transparent on that. Because they will have to deal with the fact that we are going to say, “These are some occupations where the real problem is you. You have set up a training system that isn’t working. Unless you solve that problem, don’t be surprised.” It is like nurses in the NHS. We all know that we do not train enough nurses in the NHS, which is why we recruit internationally. They have been on the shortage occupation list for 15 to 20 years. Every year, the Department of Health and Social Care said, “We will train more nurses,” and then never did. That really is a challenge for Government. The good thing is that it is a challenge that they should want to meet, because it fits exactly their agenda for growth. I do not see how we are going to grow the economy if we cannot grow the skills of the UK population.

More broadly, I wanted to ask you about the balance of the relationship between Home Office Ministers and the MAC. Thinking of the advice you are asked for, how political do you think your decisions are? How do you balance that relationship when pushing back on being asked to make political decisions?

They are inevitably political just because immigration is one of the one or two top topics in all opinion polls. There is an inevitable political element to that. I think it is fair to say that we rarely get requests from the Government that are overly political in terms of what they are asking us to do. If I am honest, probably half of the work we do for the Government is very technical. For example, we are currently reviewing salary thresholds for the worker routes, which we will report on next month. That is broadly a technical exercise looking at data. We are talking to firms, but thinking about where the right balance is. Even there, though, there is a political dimension. Where you put the salary threshold depends on how you balance out fiscal considerations that might encourage you to set a slightly lower threshold, because these people are positive, that fit the fiscal bands, against the net migration target or objective, which might push you in another direction. The way we try and solve that is often by not presenting a single option to the Government, but saying, “Here is a range that would make sense, and these are the things you might want to consider along those dimensions.” My sense is that broadly, our independence is definitely respected by Ministers. I have never felt that we have been given a question that is intensely political.

No, I absolutely agree with that. Nor have we ever been leant on in any way to come to a particular finding. I have been on the MAC for nine years and there has never been a case of that. The report that we did on the minimum income requirement for family members of British people is a good example, because that is very much a political question. The MAC could not possibly say, “The right answer is x” because it depends on how much you care about balancing economic goals versus family life ones. There, we saw our job as basically just to explain, “This is what the trade-off looks like,” so that politicians can make the decisions that are rightly for Government rather than for us.

Do you feel the Home Office responds constructively when you either make a recommendation or set out the parameters? Is the interpretation of the data from you fair? Do they make decisions that you think are robust and can be defended, or do you not agree on that?

They do not always accept our recommendations, which is obviously fine. That is how the system is designed to work. Sometimes there is not a response at all, so we are not quite sure how they are responding to the recommendations. There is no obligation for the Government to respond in any particular time. Sometimes we do not get a response from the Home Office. I think on social care we never got a response from the DHSC when we had recommendations.

The one area where Governments of all persuasions have consistently failed to listen to the MAC is on data infrastructure and investment. The data in the Home Office is absolutely awful—spreadsheets here, spreadsheets there that do not link, and people have lost identifiers that would have allowed you to match datasets across. Also just not being innovative enough in thinking about how to link datasets across Departments. We are doing some of that now. There is an excellent project that illustrates how powerful it can be. We have now linked all the visa records with HMRC pay records, so we can actually see how much people on the skilled worker route are actually earning, rather than what they said on their form—it is actually more than they said on the form, so it is good news for the Government. There, the Home Office often says yes and then does nothing about it. That is pretty consistent.

Do you think that has long-term implications for policy?

It is extraordinarily harmful for the UK. It means that the evidence base is much weaker. Looking at research into immigration globally, the countries where the most research is done in the world—that is published in good research outlets—are Denmark, Norway and Sweden. That is not because any of us—I am not being too rude—do not care about those countries; it is because those three countries have registered data, which allows you to match people with a digital ID across every single dataset you can imagine, so you can follow people over their entire lives and see how much they use the health service and how much education they get. We cannot do any of that. They get better advice as a result.

To give a concrete example of that with something that you are interested in currently—indefinite leave to remain—the linking exercise that Brian mentioned has been useful in reports that we have used when looking at family members or people on the skilled worker route. It only goes back a few years. There are lots of people who have been granted indefinite leave to remain over the last 15 or 20 years, but we have no idea how much they earn; how that varies between different groups; how many people are claiming benefits; or what the overall fiscal impact is of people with indefinite leave to remain—the kinds of things that would be really useful to know right now. We just do not have that data, because there has never been the investment.

That is interesting. My final question was going to be: how supported do you feel by the Home Office in doing your job effectively, particularly given your expanded remits? I am wondering whether you are saying that you want to be supported with better data.

Better data would be our No. 1 ask. To be fair to the Government, in terms of the additional resources we have required—my position is full-time at the moment, Madeleine has been appointed deputy chair, and we have increased the secretariat—that was a manifesto commitment that they have delivered on. We are perfectly happy with the resourcing we are being given by the Home Office. The problem with the data thing is that it is out of our control, because we are not the data owners, so we require the Home Office and other Government Departments to work together to do that.

The Home Office is fine, but what about other Government Departments?

It is a mixed bag, like all cross-Government working. We have good working relationships across Departments, and the labour market evidence group is a good example of that. There is a challenge, and there is probably nothing you can do about it. The graduate review from last year is perhaps a good example of it. The Department for Education did not really want us to do a graduate review, because the risk was that we would say something like, “Close down the review,” so I think it is fair to say there was more of a challenge there. There is always that worry from other Departments that our inquiries will harm what they see as their interests, but at the working level it is a good relationship.

I want to ask a bit about the story of the welders. You said that a welder earns £70,000 a year and a teacher of a welder earns £30,000 a year, so there is no way that we could teach people to be welders in further education colleges. What we need, surely, is a proper apprenticeship system in the welding businesses to encourage people—either our own citizens or overseas citizens—to come and be apprentice welders. Does such a system for migrating apprentices exist, or is there only one for migrating students in our universities?

We do not currently allow migrating apprenticeships. In some sense, it would feel slightly bizarre in a world in which we have 1 million people aged 18 to 24 not in education, employment and training. There does not seem to be a shortage of people who could do these training programmes. Further education colleges are key to providing the apprenticeships, because they have to do some of the day release training. That is where the bottleneck is. The one interesting example I can give you is that a company building one of the nuclear power stations—I cannot remember which one—has essentially set up its own further education apprenticeship college. Welding is a good example of where we need to think of better and more innovative models, because it will be difficult to get people to do one job that is half the pay of the other—some people will do it because they love that job, but most people will not. The firms that are employing people as welders can say, “One day a week, you won’t be welding; you’ll be training the next welders. You’ll still be getting £70,000 a year, and it is our investment in the future.” That is the kind of public-private partnership that is more needed in some of these occupations.

I think that is exactly what is happening at Hinkley Point, and it is coming now at Sizewell in Suffolk, where we will do exactly that.

I want to go back to the earlier comments on exploitation. You mentioned care workers as being a predominant group that is exploited. How effective are the Government at designing an immigration policy that minimises the risk of exploitation?

That is really challenging because the exploitation does not take place exclusively in the UK. In the care sector, a lot of the problems that we were seeing involved people who had paid illegal recruitment fees overseas and then were in debt and very vulnerable by the time that they came here. That then exacerbated the risks of further exploitation within the UK. Obviously, there is limited capacity. As Brian mentioned, this was a system designed with a very light-touch, self-enforcement model. It works fine for graphic designers, but was then applied to care work, which is a much higher-risk industry for exploitation. I don’t believe it is possible to completely eliminate those risks. The choice is partly between having a visa programme where we know that these risks exist and we cannot totally stamp out exploitation; or taking the decision that the Government took and deciding that we are not going to have overseas recruitment of care workers—partly because it is such a mess trying to enforce terms and conditions. I do think there is more that can be done within this country. For example, we do not have systematic checking of whether the wages people have been promised when they are sponsored for the job are actually what they get paid. That is probably something that could be done more on an automated basis—although there are all sorts of complications about precisely how you do it and making sure it is accurate. There is probably also a conversation to be had about the exploitative sponsors. My understanding of the system is that currently they get struck off for 12 months and can waltz back in and continue to employ workers. Does that system make sense? It would not necessarily be the case with many other things that people require a licence for. In terms of the design of the immigration system, one question is about switching—could it be made easier for people to switch away from an exploitative employer? If you look at the care issue—where we obviously do not have new overseas recruitment, but we still have quite a lot of people displaced within the country—it is striking to me that there is no visibility as to where the jobs are in the care sector. In theory, you could design systems that required online posting of jobs. Then the people who had lost their job because of their employer’s licence being revoked and those with an exploitative employer who wanted to leave could work out who are the sponsors still recruiting people within the country. That is an area where things are currently not working so well.

To follow up, do visas tied to specific employers inherently increase the risk associated with exploitation?

Yes. Although there are also reasons that we tie people to employers. The visa system is designed for people to come into particular qualifying jobs. If we want to make sure that the job meets the criteria, there must be at least some link between that job and the visa.

Are the Government or Home Office focusing enough on reducing exploitation when designing the roll-out of the visa schemes. Is it something that they are providing enough weight to in designing that?

It is difficult for me to say how much weight it gets internally having not been part of those discussions. There are probably things that could get more attention in terms of making it easier to switch and how to do more automated enforcement. As far as I am aware, those things are not really happening currently. As an outsider, it is difficult for me to truly say what the right amount of focus is for them to put on it.

There is also a bigger fundamental problem, which is that Parliament keeps on passing legislation about workers’ rights and then does not enforce any of it. We have a system in place in this country that says that once we have passed a law, for some reason, all employers are just going to obey. It turns out that the bad employers are not. I do not know if it is still true, but one of my predecessors had this statistic that your likelihood of being visited by a national minimum wage compliance team was once every 500 years or something. Once the worker is here, there are immigration issues, but fundamentally it is a problem of your employer being exploitative. That applies to British and migrant workers. Whether the new Fair Work Agency can fulfil its promise is a real test. If it does, that will help migrants as much as it will help British workers.

May I ask what work you do on the overseas domestic worker visa route?

That is a very good question, because we are publishing an annual report next month, and there will be an entire chapter on the overseas domestic worker route. We are not commissioned by the Government at the moment to do anything on that. The White Paper had a fairly fleeting reference to it, in which it said it was concerned about exploitation on that route. What we are doing for the annual report is essentially almost like a primer on what the overseas domestic worker route is, why we have it, and what the pros and cons are. My reading of that—we have not quite finished it yet—is that it might sound superficially attractive simply to abolish the route and say it is bizarre to allow people to bring in their domestic staff for up to six months when they are coming on holiday to the UK, but I think there are two arguments for keeping it. One is that—let us be honest—they are very rich people by and large, mainly from middle eastern countries, and there would be a cost to the UK economy if they chose not to come as a result of not being able to bring their staff. The other thing is that what you would find is not that they would not come, but that they would bring their staff on visit visas and the like. In other words, you would just drive it underground rather than making it any better for those workers. The focus should probably be on whether there are things you can do—probably at the margins, to be honest—to reduce the risk of exploitation while they are in the UK. To be realistic, if they are being exploited in the middle east, they are probably going to be exploited in the UK. You could argue, therefore, that we are not making it any worse for them, but we may be more uncomfortable about it happening here. The question then is whether you can do things like improve the national referral mechanism if they think they are victims of modern slavery, and that kind of thing. So there is some work to be done by the Home Office on that, but at the moment we are just laying out the groundwork.

I want to follow up on Robbie’s earlier point. Dr Sumption mentioned the balance between giving employees the ability to potentially transfer to another employer, and giving them the advantages of being sponsored by a specific employer. In a past life, I have seen clients suffer from that restriction with one employer, and in some cases not bring a claim to the employment tribunal because they were so worried about losing not just their role but also their tier 2 visa status at the time. Is it practical to get to a system where they do have the ability to transfer where exploitation, or a claim that could have merit, is demonstrable?

There are different ways of designing it. If they have to prove exploitation, that is harder than options that allow people to move more easily regardless of whether there is exploitation. I suspect you might see similar problems if people had to prove exploitation before they were able to switch employer. They would have similar levels of concern, because it is going to be difficult to do that totally without the employer’s knowledge. There are other models. For example, Sweden has a model where people are initially sponsored, but after a certain amount of time they can then work in the same occupation. That model relinquishes a little bit of control of the worker. There are pros and cons if the Home Office does that, because there is a risk that workers would not be overseen and would move into occupations that were not eligible. There are options that would allow people to secure a new job first and then sort out the paperwork afterwards, as long as the new employer is willing to sponsor them. It gets down to the technical details of precisely how you want to do it and how you deal with the trade-off between giving employees more freedom to move around, and the desire to be totally sure they are working in an eligible job that meets all the salary criteria and all the rest. Ultimately, there will be a trade-off between those two things.

Did you say it was the Swedish model that allowed flexibility within the same sector?

Yes.

Very briefly, you referred to the establishment of the Fair Work Agency. Do you have a view on how effective that is going to be in tackling exploitation—on migrants specifically, not in the wider labour market? Are you making recommendations to Government about how it should be set up or structured?

Not thus far. Fundamentally, it will all depend on the funding arrangements, because the real key problem is that the constituent parts going into that organisation—the one from the Home Office was the Gangmasters and Labour Abuse Authority—simply did not have the number of people needed to raid offices, factories and the like. Unless you do that, it is a fairly pointless exercise; if it is all just trying to get employers to do the right thing. Most employers do the right thing, but for the ones that do not, you need the enforcement powers. One area on the migration side where the Government could think a bit more is addressing the funding issue by putting more of the onus on the employers. To an extent, we do that with the seasonal worker scheme. As I understand it, some of the fees that are used in the seasonal workers scheme are used to fund people in the Home Office to do the enforcement. That seems like quite a good model for making employers as a whole pay to make sure the sector is well enforced.

We will now move on to family migration.

Dr Sumption, you mentioned before that, following your family visa financial requirements review earlier in the year, it was for the Government now to make a set of fundamentally political choices about balancing different considerations. Could you outline in brief what key considerations you think the Government need to balance when making their decision, hopefully later this year?

One of the challenges that the Government face on the minimum-income requirement in particular is that, if they care about the economic impact of family migrants with spouses coming to the UK, then what really matters is what the applicant migrant does after they arrive—do they work, and how much do they earn? That is really the key thing, especially if you are looking at fiscal impact, but also if you are looking at other economic metrics. The challenge when scrutinising people who are not yet in the country is that we do not know very much about them; we know about their UK citizen sponsor, which is not a brilliant predicter of what the applicant is actually going to do. You could, for example, have someone who is earning £60,000 and they marry someone from overseas who comes over on a spousal visa. If that person does not work, they will have a negative fiscal impact even though their partner is earning a lot. The challenge is that the minimum income requirement in itself is not very well targeted. Because the overall fiscal impact of the route is expected to be negative, you can have a higher minimum income requirement and expect to reduce the fiscal cost, but you are not necessarily reducing migration of the people who would have been most fiscally costly. It will have some impact, but it will also affect people who would have come into the UK and done perfectly well. There is a question of whether all the weight of Government policy should be on this one metric about the income of the sponsor, or whether other things more important. I know the Government are looking at language requirements currently. We were just asked to look at the minimum impact requirement, but it may be, for example, that language requirements are more important than the income of the sponsor at the time of the application. That was not something that we assessed, but there will be other metrics as well that will have an impact.

Building on that, I have been thinking about the potential for the work plans. We have got a situation at the moment where people might have an active job offer in the UK for an amount of money that would not meet the skilled visa entry route requirement in itself, but is clearly in excess of what their partner is earning in the UK. They are not eligible to come to the UK through that route, so have you any thoughts about how job offers for partners potentially coming into the UK could be considered as part of that?

We did look specifically at the family route in which British citizens sponsor their overseas partners. One of the challenges is that, at the point they are applying, most people do not have a job offer. If you enabled job offers to be considered, I am sure that more people would go out and get job offers, but the Home Office is anxious about enforcement around the job offers. Currently, we have a discrepancy where British citizens who are moving back from overseas, such as a British person living in Australia moving back with their spouse, can have a job offer in the UK counted, but the applicant cannot. The Home Office position is that it is harder to tell whether it is a legitimate job offer if it is the foreign national who has the job offer than if it is the British citizen, or that the scale would just be so much greater that it would be harder to work out which job offers were real and which were not. There is a related challenge: in theory, you could say, “Okay, this person has a job offer. They are going to come to the UK”, and then we could check in a couple of years’ time whether they are doing the job and earning whatever level. The challenge is that, in practice, once people have arrived in the UK, they tend not to leave. If they do not meet the requirements, they will just shift on to the 10-year route, rather than being removed. In practice, we end up making the decision about whether someone will be able to come to the UK at a point when we know almost nothing about them.

On the evidence base that underpins the policy setting that the Government will have to do, can you say a bit about the evidence base that the Government have at the moment, and how you think it could be improved? I am particularly thinking about the welfare of British citizens and their right to family life, and how far you took evidence from British citizens and heard the voice of constituents such as mine—and I am sure we all have such constituents—who have a partner overseas that they are currently not able to live with, who is sometimes the parent of their child.

We took a lot of this on this, and it was very clear that the impacts on the families who are separated by the minimum income requirement are extremely negative, or at least can be extremely negative. There were a number of people who were separated for short periods, maybe a couple of months, who said, “Well, that was negative, but broadly speaking, it was fine.” With longer separations, particularly in cases where people did not know how long it was going to be and were not sure that they would ever qualify, that is where we saw the most negative impacts. The subset of cases where there was very significant evidence of harm was where families had children who were then separated from one parent. You would have a parent in the UK, usually with the children, who was then effectively solo-parenting their child while also trying to work enough hours to meet the income requirement. There was some quite convincing evidence on the negative impacts on the mental health of the children, some of which were sustained for long periods even when the spouse did get to come to the UK.

Given what we know about differences in average wages across the UK—I represent Sunderland, which has fairly low average wages—does it follow that the current policy is essentially creating further harms in places such as Sunderland, where the average wages would be lower, because the minimum income requirement is a nationally set level?

It is clearly harder to meet the income requirement in some parts of the country than others, but it is not easy to vary policy depending on where people live. First, you have some couples who are coming home from abroad, and they do not live anywhere in the UK—they are overseas, so it would be difficult to attach them to a particular place. There is also some concern about enforceability and that people would move temporarily to one area, to benefit from a lower threshold, and then move back again. You would create a certain amount of complexity, so we did not recommend regional variation in the salary threshold, although that obviously is a concern.

I have a couple of points to add. One thing we suggested that the Government could at least think about—a broader question that we are thinking about across various routes—is that one could do a calculation in which you exclude just London from any calculation that you make. You would produce a national figure of something, but just take London out to make it more representative of the rest of the country. That is an option that the Government would have. Obviously, you cannot do that for the national living wage, for example, which does not vary, but some measures do. The question about evidence base relates back to data, and it is a test for the Government. We gave them the report in June, and one of things we said that they should do as a matter of urgency was to include in the application that the applicant, the person coming from abroad to join the British citizen, makes a requirement that they put down the national insurance number of the sponsor—or the British person—so that we could link that to HMRC records and see how much the family earns. Let’s see whether the Home Office has done anything about that.

I have one small follow-up on that. What is the evidence on the impact of the changes to family migration on different ethnic groups in the UK? Obviously, a lot of family reunification in some communities can involve arranged marriages and things like that. What is the evidence of the impact in different communities?

I do not think we found big differences. We looked at how the composition of people using the family migration route had changed, and we also did not see huge differences based on protected characteristics, as far as they were available. Obviously, we do not have ethnicity in the Home Office data, but we have nationality.

If I might add a quick advertisement, in our annual report this year we will produce a chapter on English language. One of the things that is really noticeable is the gender differences in the ability to speak English for south-east Asian countries in particular. For many countries you see absolutely no difference between men and women in their ability to speak at least a decent level of English. What you see from the census is enormous gaps for Bangladeshi, Pakistani and Sri Lankan women, who have much lower levels of English than male applicants. That is a big group on the family route, so there must be concern about whether language may be an alternative way of trying to improve outcomes for people on the family route. That could be quite powerful given where people come from.

We will now move on to some questions about indefinite leave to remain.

First, the Government have already made it clear in the White Paper the groups that will not be affected. Importantly, given the discussion we have just had, partners coming to join British citizens will not be covered by any potential change. The Government guaranteed that they will maintain their current settlement rights after five years. We estimate that about 80% of people who come on the partner route come to join British citizens, so they will be exempt. On the family route, only the smaller group of people—those who come to join people who have indefinite leave to remain—may be affected by the policy change. The real groups we are going to talk about are those on the work routes, where the Government could change the rules, and on asylum—I guess they can change the rights associated with asylum. The work route is the one that we are most focused on usually, and the thing that the Government need to think about is the trade-off here. It is a difficult trade-off for the Government to work out. If we make it harder to get settlement, there is a benefit to that, the main one being fiscal, in that we do two things at the same time. As a simple example, instead of five years, say it takes 10 years before people are allowed to apply for ILR on the skilled worker route. There are therefore paying the fees associated with that for another five years. They are also paying the immigration health surcharge. Firms might be required to pay the immigration skills charge, although I am not sure why that would make any sense, but they could be asked to pay it for another five years. The NRPF rule is also being continued for another five years, so people will not be able to claim benefits. All that is a real fiscal gain, so it is attractive from that perspective. On the downside, there are two things to worry about. One is whether you discourage some people from coming in the first place, either because they worry that they will find it harder to get settlement, or because they think the country is less attractive because it has harder settlement rules? Also, if that affected the people at the top of the earnings distribution, it would be quite easy to turn the overall effect of the policy negative very quickly, because it is the really high earners in the skilled worker group who are contributing fiscally. If you discourage a certain share of those, the gains you think you are making on the benefits and the fees side would quickly be lost. There is then the broader issue: if you make it harder to get indefinite leave to remain and to get benefits, you have to take seriously the question not that these people will leave, but that they are just going to stay here longer without permanent residency rights. Is that good from a poverty perspective, because if they are not able to claim benefits, they may just live in relatively more poverty than they would otherwise, and are they more likely to be exploited, for exactly the reasons that we have talked about? If someone is tied to a visa, they have less power in the labour market and in rent negotiations with their landlord, because they are always worried about their immigration status. One of the advantages of settlement is that it gives people more power in such relationships, because now they are able to freely move employers—if they have an exploitative employer, they just go to the one next door. They cannot do that easily if they are on a visa. Those are the challenges that the Government have to balance.

It is interesting, because the Government have not defined what they mean by “earned settlement” yet. If I were a betting man, I would say that the likely outcome will be something along the lines of this: if you are on the skilled worker route and paying taxes for five years, you will get earned settlement after five years. In other words, it would not surprise me if the net result were not much change for skilled workers. I think that that would be broadly the right answer; skilled workers are, on average, fiscally positive for the UK. If we discourage them from being here, we will have to pay higher taxes or have lower public services. That is just a fact, so we have to live with it. Frankly, too, I just cannot imagine the Government turning around to say, “We do not think a nurse has contributed to the UK after five years.” Perhaps they will say that, but if they are not willing to say that, it is very hard to say, “Well, what about an engineer?” What about an investment banker, even? On skilled workers, therefore, I think we will end up there. Fundamentally, the only groups who will really be affected are family members who are not coming for a citizen, but for someone who has settled status, and those seeking asylum, for whom the Government could change the rules. The problem is that those groups are, on average, poorer households. The risk would be that you do not change the number of people who stay in the long run, but they have less secure time while they are here, and they are poorer overall.

The trade-off that Brian mentioned is an integration trade-off. In the short run, if you make it harder for people to get settlement and you restrict their access to benefits, you get that short-run financial benefit. In the longer term, there is a surprising amount of evidence from other countries that having access to a permanent status and then going on to citizenship is beneficial for both economic and social integration. People earn more when they move on to citizenship. That is not necessarily what you would expect, but there are quite a lot of studies that show it. There is a risk of undermining integration by making people wait longer. I add that this is a policy where it is really important to look at it route by route. We have talked about the potential fiscal benefits of people having no recourse to public funds for longer, or paying immigration fees. That does not apply to refugees, because they are not paying the fees and they already have access to benefits. It is less obvious what you are doing with refugees by making people wait longer. It might be that the main impact on refugees is to undermine their integration by giving them a less secure status. The situation could be very different for work migrants, where there is a realistic chance that people will leave the country; if the settlement criteria become more restrictive, you may well see people emigrate. That is not something that we expect to see for refugees.

That is correct.

Obviously, it is for the Government to decide whether to commission us to do any work. The key thing that the MAC would be able to contribute would be primarily on the fiscal-economic side, focusing on the trade-offs: if you change IR rules, and therefore change fees and access to benefits, what might be the fiscal implications of that, under various assumptions about how many people might choose not to come in the first place, and what might the trade-off for that be? We could do some work on that, but I would caution that that would be pretty speculative work by anyone, because—to come back to data—we literally have no idea which visa categories claim which benefits. We simply cannot answer the question of whether people on the family route, once they have indefinite leave to remain, are more likely to claim benefits than British people of the same age and gender. We have no idea at all.

We do not know. There is just no data that allows you to answer the question whether people on the family route or people on the work route, once they have ILR, are more or less likely to claim benefits than British people? The only evidence that I have ever seen for the UK is quite old. It looked at whether migrants overall were more or less likely to claim benefits relative to British people of the same age and gender. The evidence there was that Europeans were significantly less likely to claim than the equivalent Brit. Non-EU citizens were slightly less likely to claim, but it was a marginal difference. The problem is that that does not account for all the other things that you would want to know. If more British people were disabled than people who had come as migrants, that would explain the difference, so you need to delve into the detail of the data, and we just do not have good data.

We do have data on the salaries that people earn, and some of the fiscal modelling that the MAC has done looks at that. We can assume, basically, “If a household is on this income, they would be eligible for benefits”. We do not know whether, if they have the kind of income that makes them eligible, they are actually claiming those benefits. We do not always see people claiming benefits that they are eligible for, especially if the amount of benefit is not massive.

It all exists in Government; the question is whether the different data sources are linked up. I do not want to say that it is super easy; there is obviously a fair amount of work to be done, but it is eminently feasible.

The key thing is that now that we have linked visa records to HMRC records, it requires a national insurance number, which is all you need to get into the DWP dataset as well. It is completely feasible to do.

We were responding to a commission—this is the temporary shortages review that we are currently doing. We were commissioned by Yvette Cooper, the then Home Secretary. She was clear in her commissioning letter that we could not consider occupations that were not critical to the industrial strategy or the building of critical infrastructure. We took that to mean construction. We took it to not mean the broad critical infrastructure. The difficulty for the Government is that if you type critical infrastructure into the web, you get 13 or 14 sectors, including transport. It includes Government, so that presumably is the entire public sector. It includes food and health—it basically includes the economy. So if they decided to ask us to review that, there would have been no change in net migration. Part of the White Paper’s approach was to stop mid-skill level jobs from being able to have general access to the immigration system, and restrict it to only some sections thought to be crucial—that is the political decision that the Government made. We said to them, “If you really meant critical infrastructure to mean the very broad definition of transport and health, just write to us and tell us, and we will put them back in.” The Home Office has not written back to us to ask us to do that, so that is a question for the Home Secretary.

It is important to remember that anyone who was employed before April 2024 is essentially grandfathered into the system. So if you were in a job on the skilled worker route prior to that, for example, in customer service operations for TfL, because you were on a visa that was legitimate at that time, you will still be able to continue that visa, renew it and apply for settlement. It is only new people that are being stopped from using that route.

The answer is no. We cannot deviate from what the Home Secretary commissions us to do. The Home Secretary told us which occupations we were allowed to consider. It is literally impossible for us to say, “We are going to consider other sectors because we think you are wrong.” That is not our job; our job is to respond to a commission from the Home Secretary. It is for the Home Secretary to say, “Actually, I have reconsidered, and I think these other sectors are important and I want to include them.” If they do that, we will happily go back and do it again. Inevitably, that would affect the reduction in net migration that the Government planned in the White Paper.

We did, yes. We had a roundtable with all the trade unions.[1]

Sticking with the proposal on indefinite leave to remain, am I correct in thinking that there is not much distinction at all between indefinite leave to remain and citizenship? The only one would be voting, and if you are a Commonwealth migrant, even that does not make a difference. So the system does not draw any distinction between those two things.

Voting is probably the most important one. Citizenship gives you a passport, and a lot of people find that very attractive. If they come from a country where you cannot travel very easily around the world without a visa, having a British passport and being able to travel on it is quite desirable. Even though indefinite leave to remain is, in theory, permanent, it can be lost if you leave for a long period of time. If you accepted a job somewhere else for three years, you may lose your indefinite leave to remain. Citizenship is more of a cast-iron guarantee that it is permanent.

That is a totally fair point. The point I am driving at is that the immigration system in the UK is pretty agnostic about whether someone who comes here ever becomes a citizen. If you are in the UK, you can leave and things like that with citizenship, but you do not actually get any benefit from citizenship—there is not much distinction between citizenship and ILR. What would be the MAC’s view? Have you ever looked into drawing a bigger distinction between ILR and citizenship? Do you think that would make a difference in the integration indicators that we are looking to see?

The MAC has not looked at citizenship, but there are several key considerations. Some people have ILR and they do not want to become citizens, one of the major reasons for which is they come from a country that does not recognise dual citizenship. Obviously, we do, but if their country of origin would make them renounce their citizenship, a lot of people are reluctant to do that. They may feel that, if they are going to go home and look after elderly parents at some point in the future, they want to be able to have those rights there. That would be one of the main impacts. If you had a policy that tried to discourage people from being on ILR without becoming a UK citizen, that would be the biggest challenge. If you look at the rights, it does not look on paper like there is much difference between ILR and citizenship. There are some studies from other countries that show there is some benefit for people in both economic and social integration outcomes when they become a citizen, which is actually quite surprising. There is then a whole scholarly debate about why that would be, but we do seem to see an impact. Some people then look at that and say, “We should make it easy to become a citizen because, for whatever mysterious reason, citizens seem to do better.” I suppose the other argument is that, if you want to try to encourage people to adopt certain behaviours—do the “Life in the UK” test and learn a higher standard of English—do you use the ability to obtain citizenship as an incentive to do those things? There is a debate about how you should balance those two objectives.

That is really interesting. I suppose what I am trying to drive at here is: what would be the benefits of saying to migrants, “Okay, if you get indefinite leave to remain, you can stay here indefinitely, and you can move around different jobs, but you cannot claim benefits or some of the things that you get with ILR. To move from ILR to citizenship, we are going to put in some more tests and different kinds of indicators, including a proper “Life in the UK” test and properly speaking English, as well as the Home Secretary’s point about contributing to your community”? We are trying to tackle the point that you were driving at, Brian, about exploitation of some people, and we are also trying to make the move to permanent status meaningful, with access to a lot of the “literal” benefits and more generic benefits. If what I said makes sense, what do you think about it?

I suspect it would create an incentive for people to move to citizenship. In the short run, there will be some people who cannot do that, and you would have lower welfare spending on that group, potentially at the cost of higher risks of poverty for those people, if they cannot move to citizenship. The main question would be what happens to this group of people who are unwilling to become citizens because they do not want to renounce their original citizenship. They would obviously be the main losers from that policy.

While the Government are thinking about indefinite leave to remain and the Committee are doing a consultation on it, it is helpful to distinguish the immigration control perspective of indefinite leave to remain from all the other things that ILR gives you, and there is no reason why they have to be at all related. Our view would probably be that it is really good for workers and people to be freed from immigration control, because that gives you power in lots of dynamics within society, because you are no longer worried about being kicked out. That does not mean in any sense that that should be when you get benefits. It so happens that that is what we currently do, but there is no principle there. There would be nothing wrong with the Government saying, “We are going to have a different rule for when you can get ILR, in terms of no longer having to have a visa to stay here, and when we let you have other benefits and access to other services.”

That is really interesting, and it takes me to my follow-up question. Is my interpretation right here? Net migration peaked a couple of years ago at just over 900,000. As that cohort of people inexorably move through the time periods they have been here, they will start hitting the point at which they become eligible for indefinite leave to remain, and therefore will become eligible for benefits as well. Is that right? What would be the implication of detaching ILR from benefit access for that cohort of people?

That is absolutely right. I am sure that part of the reason for the momentum of the White Paper is a recognition that there will be a bulge. You move five years along, and you see where you will suddenly get a large number potentially eligible for ILR. Importantly, however, the majority of people who claim ILR do not claim at year 5; it takes a number of years. It is more likely to be between five and 10 years that you eventually get almost everyone. You would think you would, but lots of people do not because it is a sudden £3,000 bill just to get ILR. If you are constrained, you may have to keep on paying the visa fees until you have saved up enough to get the ILR fee. It is inevitable, therefore, that the Government should think about this, because there will be that bulge in numbers. My view would be that one should think very carefully about stopping ILR itself and settlement, where all that means is not being subject to Home Office control. The only benefit I can see of staying within the immigration system is if the focus is really just on immigration fees, if the Government think that that is worth it. I would be surprised if it was worth it. To me, that is a completely conceptually different question than when you allow people to get benefits. There, I think you could legitimately say that there will be some categories of migrants who we decide will not be able to get benefits for x number of years further. The difficulty with doing that—it is a bit like the immigration health surcharge—is that we are asking all these migrants to pay all the taxes that British people have to pay. There must surely come a point at which you say, “It is a bit harsh that you’re paying for all these benefits and all the health service, and we are charging you additionally for the health service—we are double-taxing you—and making you pay for these things that you are not eligible for.” There must come a point at which that feels an uncomfortable exercise.

I just want to check I have understood you. Are you saying that, if you kept charging people visa fees for another five years of ILR, you do not think that actually adds up to very much?

It depends which visa fees you make them pay. There is one that I would strongly urge the Government under any circumstances not to do. I think it would be extraordinarily unreasonable to make firms continue to pay the immigration skills charge after year 5. This is not an additional migrant being brought in; it would just be a tax on business, and that does not seem very sensible.

This is my final question. One of the analyses coming out of the Home Office is that, essentially, in the British immigration system, we do not make very much distinction between types of migrants. As long as you are here, you do not commit a serious crime and you do not leave the country, you are inexorably eligible for ILR after five years. Some migrants come here and make enormous contributions and do loads of voluntary work, while some who do not—just like any British citizen, frankly. Does the system reward or incentivise any of those behaviours? Does that sound right to you? Do you think that the system could be adapted to take greater recognition of the contribution some people make and help them speed up the process to ILR as a result?

It absolutely could. There is a bit of it already. The good example is the global talent visa, which allows you to get settlement after three years rather than five. Hopefully, we are selecting the right people, and that is a faster route. On the other side, if you are on the family route and, at any point when you try to extend your visa once you have arrived in the first place, you are no longer meeting the minimum income requirement as a family—although that is less likely once you have arrived—you get moved on to the 10-year route to settlement, so there are some distinctions. I do not think there is anything obviously wrong with thinking about an earned settlement model. I hate to say you could have points and trade them off, because everyone believes in a points-based system, but it does not always work. There is something attractive about that. You could easily have a model that says that if you are a higher rate taxpayer, you get faster access than if you are just a basic rate taxpayer. You could do that, and I think there is an attraction to doing so.

You have to think carefully about what you are trying to achieve with this thing. First, we do not really have much evidence. There is a gap in the research on how much incentivising happens as a result of that. We would like to have a model where you say, “Okay, this is the criterion, and then everyone’s going to meet it,” because people will work harder to meet a higher level of English or whatever. But you have also got to think about the impact of what happens to the people who do not meet it. There is obviously the principle of wanting to reward people who meet whatever criteria you are interested in, but the other side of that sometimes gets forgotten. You must make sure you are happy with the impacts on the people who do not meet the criteria—especially where there are going to be negative impacts on them or their integration.

I have a quick follow-up question, not so much on the incentives, which Chris covered, but a potential disincentive. You mentioned earlier that refugees would probably be more likely to stay in a state of poverty if the ILR qualifying period was extended or they were excluded from it. Is there evidence of previous visa or migration status changes that have acted as a deterrent on refugees? Although you have said that it would not affect them, could those also potentially act as a deterrent to skilled workers even if they are not brought into it but if it were part of a general unwelcoming environment?

Are you talking specifically about whether the ILR changes affect people’s decisions?

To come here, yes.

We do not have a lot of evidence for that. This is again a bit of a gap. There is a relatively recent paper that particularly looked at a group of people in the refugee or unauthorised migration cohorts. It surveyed people overseas and asked what kind of policies they would find more attractive. It found basically no difference between a five and a 10-year path to settlement.[2] What people actually cared about was the ability to get a legal status rather than how long it was going to take them. People will often have quite a long time horizon for that. In theory at least, even though we do not have great evidence, we would expect that people who have more options—for example, skilled workers or those with very high salaries—might be more likely to pick and choose and decide to go somewhere else with more generous criteria. We should not overstate that. People do not line up five countries and decide whether they should go to Canada, the UK or wherever—although international students do, to some extent. Instead, a lot of people have a particular job offer in a particular place. They have a particular reason to come to the UK and they are not necessarily just going for the country with the best policies.

Chris has mainly covered my point, but how can community responsibility be measured? I can imagine myself or other MPs suddenly being approached to write references for people who have been doing good things in their community. What standards could be expected of people to achieve that?

This is an area that is going to be very difficult. Some of the other bits of the earned settlement model are quite straightforward—your tax contributions and tick-box things about criminal records are fairly easy. This is the one area where an awful lot more thought will have to be given as to what we mean by that. The FT had a rather obvious point, which was that what you will create is a cottage industry of organisations that set themselves up to certify that a migrant has done community work, and it will be meaningless. This is perhaps a broader question for British people as well, but we also need to think about what we mean by contribution to society and community. Are we really saying that a migrant worker who is merrily working at Oxford University developing a vaccine should also be singing in the local choir or picking up litter at the weekend? If they are not, should we refuse their application? That seems an odd measure. If it is going to be a tick-box exercise in which you just must tick everything, you will get a lot more people joining local choirs and that kind of thing. However, if we are going to have those criteria—which will be very difficult to measure—it would be more valuable for those who cannot do the other things so that they can show a significant contribution in that way. That would be beneficial. Otherwise, if we apply it to everyone, it will just be a fairly pointless tick-box exercise.  

We are moving into the final two areas of our questioning, starting with student migration.

Professor Bell, I think earlier you expressed some queries about the current design and use of the student and graduate visa system. Can you say a bit about your views on that?

In 2018, the MAC wrote a report for the then Government about the student route, and part of that was about whether we thought there should be a graduate route. Our view was there should not be at the time. Our recommendation was to make sure that foreign students who come to the UK and graduate have a reasonable length of time to get a graduate job once they have finished their degree. We have simply suggested that there should be more flexibility at the end of the student visa to allow a bit more leeway for students to stay to find a job, but that that job should have to be a graduate job, and therefore eligible for the normal skilled worker route. The Government under Boris Johnson decided instead to create a graduate visa. Looking at the intention of the policy when it was launched, anything to do with immigration is right at the bottom of the list. The objectives of the policy were explicitly to have more international students in the UK and make UK universities more attractive to international students. In our review in 2024, we said that it completely succeeded in that objective; it was one of the greatest successes that the Government of the day achieved. The international education strategy said we should have 600,000 students by 2030; that was achieved in 2023. It succeeded, but it succeeded because it essentially created a model in which, if you were a student coming particularly from poorer countries around the world, you were essentially paying a fee of one year of a master’s programme at a university, and that fee bought you and your partner three years in the UK. Your partner could work three full years without any restriction in any job, and you could work two full years on the graduate visa, and 70% when you were a student, because you can work full time during holidays and 20 hours in the rest of the time. We created a model in which we said, “There’s a visa here that is about being a student, but fundamentally it is about getting the right to work in the UK labour market.” Our view at the time, and my view still, is that that is not a very good model, and what we should be doing is making it easier for students who graduate to go on to graduate jobs.

The Government’s immigration White Paper states that it is “essential that any individual that our country welcomes here to study is…not using their application for a student visa simply as a vehicle to move themselves and their dependants to a new life in Britain.” How far do you think the current student and graduate visa system meets that policy intent set out in the White Paper?

To a large extent, the policy now succeeds on the family thing, because only master’s by research and PhD students are allowed to bring dependants. If you cannot bring a dependant on the student route, you are then also not allowed to have them on the graduate route, so it has essentially closed that down. My only concern would be that one hears whispers that some universities are redefining their master’s degree to call it a master’s by research simply to get around that rule. I think the Government need to monitor that quite closely and be willing to take action quite fast if there is an abuse of the system in that way. It is still failing, but it is a marginal failure in some sense, in that we are still allowing 18 months after graduation for a decent chunk of those graduates—about 40% in our report in 2024—to be working at close to minimum wage jobs. That surely was not the intention of the graduate route; they are graduates, but they are working in minimum wage jobs. In reality, although they are still getting those 18 months—it will be 18 months from next year—the one big thing that has changed is that we have now restricted the skilled worker route to not recruit care workers and only recruit at graduate-level jobs. All we are really saying now is that you can get an extra 18 months in the UK once you have graduated, but then you will have to leave, because the vast majority will not be able to get jobs on the skilled worker route at the salary required. It is a question of whether you think that 18 months is worth it from the perspective of the country as a whole for the benefit it undoubtedly gives to universities in recruiting international students.

What do you expect the impact of the planned changes to move the graduate visa length to 18 months from 24 months to be?

I think the Home Office in its White Paper assessment—its annexe to the White Paper—had a very small number: I think it was the order of 7,000 fewer students. Essentially the question is how many fewer students decide to come in the first place, because it is a less attractive option. I think they are basically right that it is a really small number. The reason it is a really small number is that 18 months versus two years is not that dramatic a difference. I think you will get a very small fall-off, but it will have very little effect. What should have a bigger effect is the fact that, because we have now made it much harder to switch out of the graduate route into other jobs, a lot of students will come on the graduate route thinking they will be able to stay and then find they cannot. We had a research company interview 30 students on the route for us, and one interesting thing was how many were quite optimistic that they would get a job that would meet all the salary requirements—even though, when you looked at what they were doing, with the best will in the world, that was hard to believe. That might eventually percolate into more significant effects, because people might begin to realise that it is not worth doing the graduate route because they are not going to be able to stay long term.

What is your assessment of how widespread the active misuse and exploitation of student visas is? The Government have also suggested some changes to the compliance assessments they carry out with universities.

There are two aspects to that. One is the compliance requirements themselves. I think the Government want to increase them by essentially five percentage points for each of the three criteria, which are about visa rejection rates, completion rates and course enrolment rates. Those are all very sensible. It is reasonable for the British public to expect that, if universities are recruiting people who will come on visas, those people will not have their visas rejected in any significant numbers, they will be enrolled on their course and they will actually complete the course. Frankly, if they are failing that for British students, action should be taken. They are all very sensible criteria. That data should be published by universities. Universities should not be ashamed to say, “These are the numbers that we achieve.” The bit that the White Paper highlights that is not in the assessment criteria at the moment, and that is a real concern, is the number of people who are coming on the student route and then applying for asylum. In particular, there is a very real concern about the number of people who do their course and then discover that they are fleeing persecution. The Government need to think carefully about the broad policy question as to whether it is okay to arrive in the country, not claim asylum and wait until you have finished the course; if nothing has changed in your home country, is that an acceptable behaviour? Again, my preference would be to publish the numbers, for every university, of how many people are claiming asylum after studying, because I think there is a real problem in some universities.

We will finish off with the seasonal worker scheme.

What progress have the Government made in addressing the recommendations in the MAC’s 2024 review of the seasonal worker scheme?

The honest answer is that we do not know because the Government never replied to the review. I should say that, if I recall rightly, we published the review about a day or two after the general election, so they were probably busy with other things. In one sense, it is a slightly odd review: it was the first review we had ever done that we commissioned ourselves to do. We are allowed to do that under the agreement we have with the Home Office: essentially, if we do not have enough work from the Home Secretary, we can decide to do our own review. Therefore, it is not entirely clear whether the Government feel that they should respond to us, because they did not ask for the report in the first place. As I understand it, they are progressing this internally. Of course, it needs to be with DEFRA, and therefore there are the usual cross-Government difficulties, but I think they are progressing it. But they have not written to us, so I cannot honestly tell you whether any of the recommendations we made have been accepted or whether they are acting on them.

Do you think it is right to reduce reliance on, say, migrant workers coming into food production? Do you think there will be problems up the road?

There is an ideal world we get to where there are robots on the field collecting all the fruit and veg. That is the right model if you could get there, because there is a good reason why we need seasonal workers: it is an extremely unattractive job to do. It is usually in remote locations, so there is not an available pool of labour. I remember Boris Johnson, I think, saying, “Oh, I remember when I was a youth, and people would go to the farms and pick on their summer holidays.” I am not sure that he ever did, but that was the idea. That is just not the world we live in any more, because why would you do that when, if you want a job over the summer, you could work at Starbucks? It is a job that can be decently paid, funnily enough. If you are productive enough in this job, you can earn quite a lot more than the minimum wage, but there is a lot of minimum wage work there, in unpleasant conditions. It is hard work. We saw that in the pandemic when we had Pick for Britain, and two people applied or something. We should live with that—that is the reality. On the other side, it is an extraordinarily attractive job for the people who come from abroad. It is now mainly the ’Stans that we recruit from. For them, coming for six months over a two-to-three-year period is enough to buy a house or to start a business back home; it is really transformative. Interestingly, the Governments of those countries are also very keen on the scheme. Broadly, it is a well-structured scheme. It is interesting that we were talking about exploitation earlier. It is a sector where you would expect mass exploitation—all the conditions are there for really bad work practices. I am not saying that there is not exploitation on the seasonal worker scheme, but there is probably much less than you would think for that type of scheme. Part of the reason for that is that we have separated the recruitment of workers from the employment of workers. Breaking up that link between the visa and the work is really important. It is a scheme that we will probably have to have until the magic robots come along. The only risk—and it is hard to think how you would design policy well here—is whether, if you make the scheme easy, you lessen the incentives to invest in that technology. There is a bit of a worry that if we continue to do that, firms will say, “Let’s not spend the quarter of a million pounds it will cost to do the technology because we can still get access to the workers.” The previous Government had a plan of essentially reducing the cap year after year to try to encourage that. Whether that is politically sustainable—is it realistic that if the cap were hit, the Government of the day would say, “We’re not going to bring in workers and we’re going to let the fruit rot?” Probably not.

Going back to your earlier point, do you think it is about the employers? If you read the studies of what actually goes on, you can see that some really value their workers and make sure they are paid well and in good accommodation, but there is a group, going back to your reinforcement point, that does not. The properties are not licensed and the enforcement base is not good. Do you think that, until the AI arrives, that will not be dealt with and will still be out there?

It is definitely a challenge. There are two elements to it: the agencies that are recruiting from abroad and the farms that are licensed with the agencies to employ. On the agency side, I would say that the Home Office has revoked and suspended licenses in the past, so it is taking a fairly active approach. The good thing is that there are only five or six agencies, so it is quite easy to monitor them. Interestingly, a few years ago, the agencies themselves were recruiting from Nepal, for example, and saw that there were lots of problems with this bonded labour where they were paying fees that they should not have been paying to local middlemen, so they pulled away from Nepal and move to the ’Stans where there seemed to be better control. The Gangmasters and Labour Abuse Authority has been quite active in managing that relationship, so that part is quite well managed. The link between the farm and the agency is fundamentally for the agency to get right. That is part of what we regulate. It does not work all the time. The good thing is that workers—again there is a question about being linked to your employer—are allowed to freely ask to move employers. They cannot be refused unless there are really good reasons, so you get quite a lot of movement across farms. It is not perfect, but my view is that it is much better than you would have expected if you said, “I am going to let migrant workers work in agriculture.”

We have recently heard from the Government about an allocation of 41,000 visas for the horticultural sector and 1,900 for poultry, which deals with many of the challenges that have been raised in terms of wanting to get that additional workforce for picking fruit and veg and for the poultry sector. The cap has been set for 2026. Is the cap right, based on research that you have been doing? Secondly, the flexibility for that visa is for working for six months within a 10-month period. Is that right, and is there enough flexibility for employers to be able to meet demand?

The cap, I think, is broadly right. In our report, we showed that by and large if the cap ever bound, it only just bound, so it is broadly at the right level. I have more concern about whether there is really a plan to get that cap down. If there is, it relies on unproven investment in technology, which is more problematic, but the cap is broadly at the right level. In terms of the flexibility of the visa, there is a trade-off. One group we hear from a lot is the Northern Ireland Mushroom Growers Association, and its point is well made: mushroom farming is not seasonal; it is done inside, and it goes on year round. Our view is that that is therefore not a seasonal job, and if it is going to survive, it has to survive against normal jobs in the local labour market. My personal view is that we should not be giving protection to a particular industry to bring in migrants at low wages when we are not letting the coffee shop down the road from a mushroom factory do it. What is the difference? You could make a difference. You could say, for food security reasons, we value mushroom growers much more and we are going to make an exemption, but we should be clear if that is what we are doing. There is the challenge of whether work is really seasonal. Six months for most firms was perfectly fine, and actually most firms use more like five months on average rather than six. There is a wrinkle in the immigration system that makes anything more than six months difficult. You could change that, but one of the problems is that, as soon as you pass the six-month mark, the immigration health surcharge kicks in, and firms really do not want that to be paid. You could vary it and say, “We are going to have a different scheme for seasonal workers,” but there is a fundamental question there. Our view in the end was that the balance is broadly right. The only thing we suggested was a bit more flexibility in that, once you have done six months, you have to be away for a full six months. We were saying that it could instead be six months within any one-year period, because it might be that the season is slightly different from one year to another and you might therefore want to bring the workers back a bit faster.

You touched on that point about seasonal workers versus investment in technology, machinery, fields and stuff. More broadly, what is your view on how migration—seasonal migration, but also general migration—has affected the UK’s productivity challenge? I know that is a huge question.

In the five minutes we have left.

No, I have a final question, so let us save that time for the last one. What should we as a Committee be thinking about as we scrutinise Ministers on that?

I am aware that productivity numbers are very important at the moment. The evidence we have is perhaps unsurprising. In general, the evidence not only from the UK but from around the world is that things like the skilled worker route tend to be positive for productivity. The best way to think about it is this: the best people you can bring in for productivity purposes are people who are not like us—those who have skills that are complementary to ours, but that we do not already have. The archetypal example is a banker from New York who has a different product skill than you have in your London office that enhances the productivity of the London office. He or she might be able to train the British workers to be better in their jobs. That is the archetypal good outcome. It is not very useful if you bring in people who are just like us, because if they are just like us, they are as productive as us, and that does not change productivity. That tends to favour high-skill workers. There is an interesting trade-off with low-skill workers. They do not directly increase productivity, but as in the example I gave earlier, they can reduce the cost of childcare and let high-skilled workers—often women—go out to work more. That can increase productivity. To be honest, the overall evidence—this is the OBR’s view broadly—is that when you sum up immigration from all routes together, it is all a bit of a nothing. Basically, migrants are just like us. They are about as productive as us. At the margin, they might be a bit more productive or dynamic, because they have travelled the world, rather than stayed in their own country, but it is really small fry. You can tilt your immigration system and make it slightly better, but it is not as if there are millions of Nobel prize winners who are desperate to come to the UK to work who will suddenly transform our growth rate. It is best to think of it as a fairly neutral effect.

That brings me to my next question. Do you agree with the OBR’s methodology of just treating migrants all the same?

I think that the overall aggregate effect is probably correct. To be fair, I do not know whether the OBR goes through it route by route and averages it out to a number that gets to about zero, or whether it just says, “We cannot do that, so we are just going to assume it is about zero.” It turns out the answer will be about zero, but the reason it would be useful to do it route by route and to try to think about it a bit more is that as we change our immigration system and favour certain routes over others, which could change things. It is marginal, though, so let us not get carried away. If we switch to the skilled worker route and away from the student route, it will change productivity a little bit, but it is not going to get you £30 billion.

The OBR’s approach is interesting, because it is both reasonable and outrageous. It is outrageous in the sense that we know that migrants are not all the same, right? There are massive differences, especially if the OBR is looking at public finances, between someone coming in as a refugee and someone coming in on the skilled worker route. There is an absurdity to a model that just says, “There is just another person in the economy.” The reason that the approach ends up being reasonable is that for complicated reasons, regardless of what we do in the immigration system, we do not tend to see the composition of migration change as much as you would expect. With Brexit, for example, we have data on the average earnings of non-citizens before and after Brexit. What you would expect to see is a big increase in the average earnings of EU citizens coming in, because they faced a much more selective system. Actually, the impacts are not that big. There is this puzzling thing. I think it is because a lot of the policy changes end up cancelling each other out, or there are other things that happen that are nothing to do with policy that cancel each other out. For example, we have more skilled workers, but we also then receive more asylum seekers, or perhaps the Government ends the care route, but then it also has some restrictions on higher-earning skilled workers that mean there is lower migration there. At the end of the day, the OBR ends up in a place that is sort of okay, despite the fact that the model does not really make any sense.

Thank you very much for your time. We have come in just on time, so thank you again. I am sure we will see you again, but this has been incredibly helpful for the work of the Committee.     [1] This statement was made in error, for which I apologise unreservedly. The MAC did not run roundtable events with stakeholders, such as Trade Unions and Employers, for Stage 1 of the TSL. We did run such roundtables, including a session with Trade Unions, for the concurrent review of Salary Thresholds. Stakeholder engagement in TSL Stage 1 was limited to Government Departments that oversee Industrial Strategy sectors and the building of critical infrastructure. Engagement was limited because Government Policy was clear on the scope of occupations for possible inclusion on the TSL and the MAC are bound to comply with guidance on policy from the Home Secretary. [2] Dr Sumption shared the following correction after the session: the study referenced here found that lengthening the path to permanent status from zero to five years had no impact on a country’s attractiveness.