In a judgment handed down last Friday, the High Court has cast doubt on the British citizenship status of children born in the United Kingdom before 2 October 2000 to EU citizens who did not at that time possess indefinite leave to remain. The case is R (on the application of Roehrig) v Secretary of State for the Home Department [2023] EWHC 31 (Admin).

The claimant was born in the UK on 20 October 2000. His mother was a French national who had been living and working in the UK since June 1995. Between 2006 and 2020, while the Citizens Directive was in force, an EU citizen living and working in an EU country for five years automatically acquired permanent settlement. Back then, there was no directly equivalent provision of EU law, nor of UK law. The claimant’s mother was eligible for and could have applied for the only explicitly permanent status then available, indefinite leave to remain (ILR). But she had not, at least at the time of the claimant’s birth.

Historically, every child born in the United Kingdom was automatically born a British subject. That rule was scrapped by the British Nationality Act 1981, which came into force on 1 January 1983. Instead, a child born after that date would only be born British if at least one of their parents was themselves British or was settled at the time of the child’s birth. At this point, you can probably see where this is going. Was Mr Roehrig a British citizen?

What does British nationality law say?

The precise wording is of the British Nationality Act 1981 is as follows, at section 1(1):

A person born in the United Kingdom after commencement … shall be a British citizen if at the time of the birth his father or mother is—
(a) a British citizen; or
(b) settled in the United Kingdom…

Section 50(2) of the Act then goes on to define “settled” as meaning:

ordinarily resident in the United Kingdom … without being subject under the immigration laws to any restriction on the period for which he may remain.

There is a bit more to it as the definition goes on explicitly to exclude certain people from being settled, including diplomats. It does not exclude EU citizens resident under EU law.

Section 50(1) defines “immigration laws” as:

the Immigration Act 1971 and any law for purposes similar to that Act which is for the time being or has at any time been in force in any part of the United Kingdom.

The question for the court was therefore whether Mr Roehrig’s mother was “settled” at the time of his birth. The key issue was really whether any immigration laws imposed a restriction on the period of time she could remain in the United Kingdom. If she was “settled” in this sense, he was British. If she wasn’t, he was not British. The court concluded that she was not settled and therefore Mr Roehrig was not a British citizen. There were two main strands to the court’s reasoning.

Were domestic regulations transposing EU law “immigration laws”?

Until 2 October 2000, the relevant domestic laws had been the Immigration (European Economic Area) Order 1994, which very belatedly had attempted to implement the European Court of Justice case of R v Pieck [1980] ECR 2171 and give effect to section 7 of the Immigration Act 1988. After 2 October 2000, the 1994 Order was succeeded by the Immigration (European Economic Area) Regulations 2000 then, eventually, successor regulations in 2006 and 2016.

The claimants argued that the domestic secondary legislation purporting to implement EU law was not “immigration law” as such because it had a different purpose. They did not quite put it like this, but the Immigration Act 1971 is basically about keeping people out by imposing restrictions and EU law is basically about letting people in by conferring rights. EU is permissive not restrictive, in that it is all about enabling people to do things. It can certainly be argued the two bodies of law are very different.

But the judge disagreed. Ignoring the nature and direct effect of EU law, he focussed on the domestic secondary legislation, which he held to exist for the purpose of regulating immigration, just like the Immigration Act 1971. All the domestic legislation literally says “immigration” on the side of the tin, so to speak.

In this conclusion, the judge was fortified, apparently, by an arguable implicit concession to this effect by then-counsel Nick Blake in the unreported tribunal case of Gal (10620) from the 1990s. The judge preferred this authority (if that’s the right word) to the position adopted by Mr Justice McCloskey in the reported decision of Capparelli (EEA Nationals – British Nationality)[2017] UKUT 162 (IAC). To be fair, no-one I know thinks Capparelli was correctly reasoned and decided, and I was pretty sceptical when it was promulgated, but it certainly highlighted the issue.

Was there a time restriction on residence?

In the key legal textbook, Fransman’s British Nationality Law, the authors argue that even before the creation of the right of permanent residence, an EU citizen living and working in another EU country enjoyed an unlimited right of residence, in the sense that there was no time limit to their period of residence. A similar argument was put to the court in Roehrig: that EU residence is “open-ended and not limited either to a single employment … or to a single category of qualification” (para 103).

Following Gal and some cases of dubious relevance, given they did not concern EU law status, the judge concludes that residence as an EU citizen was conditional on remaining a qualified person by exercising free movement rights. He went on to find that conditional residence is residence subject to a restriction on the period for which the person may remain. The judge adopted the defendant’s submission that

a person will be restricted as to the period for which he may remain whenever his continuing right to remain is contingent upon his continuing to meet the qualifying criteria.

Repeating the same point as I made earlier, this is not really how EU law works. In a very real sense, EU law — which is the legal basis on which an EU citizen resided in the UK, not domestic law — imposed no limit to the period for which an EU citizen might remain in an EU country other than their own. At the time of Mr Roehrig’s birth, there was no time limit on his mother’s residence.

But the judge was not persuaded. The claimant’s mother was therefore not settled, Mr Roehrig is not British and nor, by implication, are any of the children who derived their British citizenship from the incorrect Home Office understanding of how British nationality law worked prior to 2 October 2000.

Home Office change of position

The argument put by the claimants in this case and rejected by the judge was also, for many years, the position of the Home Office – seemingly from 1 January 1983 through to 2 October 2000, a period of 17 years. A lot of children were born in the UK to EU citizen parents during that time. Those children have, until now, been considered by the Home Office to be British citizens.

The Home Office position on what was required for a parent to be considered “settled” changed on 2 October 2000. From that date onwards, the Home Office took the view that an EU citizen had to have applied for and been granted indefinite leave to remain to be considered settled. From 2006, the Home Office maintained that the parent had to have acquired permanent residence. This is why Mr Roehrig, born on 20 October 2000, was refused British citizenship by the Home Office.

But did the law really change? The only law that had changed was the transition from the 1994 order to the 2000 regulations. But there was nothing in the new secondary legislation to justify the change in Home Office approach. Essentially, the Home Office changed its mind about how EU law worked and how it interacted with British nationality law. The Home Office as a government department unilaterally decided as a matter of executive discretion that relevant children born before 2 October 2000 were British and relevant children born after 2 October 2000 were not.

That’s not how nationality law works, and the Home Office well knows it. British citizenship is conferred by law, not executive fiat. So, which position was right? Was it that children born to EU citizens exercising EU rights of residence were British, or weren’t they? The conclusion would apply to all children born between 1 January 1983 and 31 December 2020 (the end of the transition period) who claimed British citizenship via a parent who was an EU citizen and lived in the UK without indefinite leave to remain. If the Home Office was right prior to 2 October 2000, the creation of the right of permanent residence in 2006 did not really matter other than providing assurance that an EU citizen was settled; any children born after 30 April 2006 to any EU citizens exercising free movement rights would be British irrespective of whether the parents had acquired permanent residence or not. I have no way of estimating how many children that might be, but it must be hundreds of thousands.

The court decided in Roehrig that the Home Office was right after 2 October 2000 and therefore, by inference, was wrong before 2 October 2000.

I should make clear that the outcome of this case is only relevant where a child’s only claim to British citizenship is through a EU citizen parent. If the child’s other parent held ILR or British citizenship then the child would be British through that other parent anyway.

What now?

We have known about this issue in British nationality law for quite some time. It is flagged up and questioned in Fransman’s British Nationality Law (section 13.2.5) and it was raised by the former President of the Upper Tribunal Immigration and Asylum Chamber, Mr Justice McCloskey, in the case of Capparelli (EEA Nationals – British Nationality)[2017] UKUT 162 (IAC).

The idea that the Home Office could from one day to the next, as a matter of simple policy, change the basis on which British citizenship was acquired as a matter of law, was always highly suspect.

It might be said it was unwise to deliberately seek to litigate the issue. That’s certainly what I was told by someone senior several years before Capparelli. On the other hand, lawyers are bound to act on their client’s instructions and this was all going to need to be sorted out at some point.

The next step is for this to go to the Court of Appeal. I assume the claimant will be seeking permission to appeal. These are meaty issues and one would expect permission to be granted. The judge, for example, assumes and indeed states that the claimant’s mother’s status in the United Kingdom derived from United Kingdom law, namely the Immigration Act 1988 and the 1994 Order (e.g. paras 42, 49, 89). That’s arguably wrong at a fundamental level; her status actually derived from EU law, which has direct effect and is of a different nature to United Kingdom law. The problem with Gal is arguably not just that the tribunal misunderstood British nationality law and conflated conditional residence with limited residence, but that it also misunderstood EU law.

Appeals are very expensive, especially once the potential legal costs for the other side are taken into account. The cost of taking this to the Court of Appeal could run to tens of thousands of pounds. Some lawyers may be willing to act for free and the Home Office might conceivably be persuaded to agree not to seek their own costs or a protective costs order might be sought. But it seems perfectly possible that the litigation may fall away for costs or other reasons.

If the outcome remains the same, appeal or no appeal, the Home Office is surely going to have do something this time. I’ve acted for people who thought they were British citizens, were recognised by the Home Office as such for years and were issued passports on that basis, only for the Home Office to turn around years later, say a mistake was made and nullify the citizenship on the basis it was never in truth held. The Home Office rationale for doing so is that citizenship can only be conferred by law, not by executive discretion or by mistake.

So, it was very surprising to read in Roehrig that the entire basis for recognising children of EU citizens born before 2 October 2000 as British was “as a matter of policy and fairness” (paragraph 41).

What should an EU citizen do if they were born before 2 October 2000 and derive their claim to British citizenship from one parent only, who was an EU citizen at the time of their birth living in the UK but who had not applied for and been granted ILR? Will they be able to re-enter the UK if they travel abroad? Will their citizenship be nullified and their passports withdrawn? Should anyone in this position be spending money to apply for a passport at this stage? We don’t really know.

The reality is that the Home Office has shown a reckless disregard for the interaction of EU law and British nationality law. The solution may have to be a retroactive statutory recognition of the British citizenship of those affected, who probably can be defined as a class for this purpose.

I’ll end by pointing to another different but related issue. If the outcome of this case stands, a significant problem awaits the children of EU citizens born between 2006 and 2020, who will often have no proof of the status of their parents at the time of their birth. They are automatically British according to section 1(1) of the British Nationality Act 1981 because their parents will have automatically have acquired permanent residence under EU and transposing UK law. But finding and sending in five years’ worth of pay slips to prove one’s parent’s status many years ago will be impossible. It’s another major problem affecting some people already and increasing numbers over time, as they try to apply for passports, comply with hostile environment checks or resist deportation on the basis they are actually citizens.