Neglected last %

When the British government starts opening up its digital gates for applicants soon, it hopes that everyone of millions of eligible EU nationals turns up. Mere esperance can be dangerous. Foremost, it requires exact numbers to go about it. At present, there is a rough estimate of a base population.

But, data is partly resting on old census results or on-going surveys that lack precision. Migration Observatory states that the majority of 3,568,000 could be eligible but no one dares wagering a definite result. It is simply impossible to know without adequate data, experts bemoan. However, two things appear bright as day. Firstly, there is a considerable margin of error. Second, whoever does fail to turn up for the test, will pay dearly - e.g. in the form of giving up their rights as citizens.

Despite those potentially missed might appear little and insignificant - next to an enormous 3.5 million figure -, considered in isolation the single percent number is vast and unprecedented.

After those eligible managed to apply, the risk shrinks. In a previous trial, no one was outright refused and everyone was awarded a settled or pre-settled status. The Home Office pledged to be lenient. But even those results come with a caveat, as explained later.

One truth is likly to persist: any given immigration scheme will turne down some - either due to their criminal records, other reasons or due to plain error. the3million campaign asserts that mistakes could be at play for around 10% among those being denied - on the grounds of government failure such as inaccurate records held. It is based on results from a previous study on blacklisted UK banking customers.

Overly hostile in a no-deal future

The consequences for those remaining unaccounted for after a hard decline - 31st of June, 2021 - could be detrimental. The present consensus is that a hostile environment will fine employers, banks, and landlords if they decide to do business with EU national without an immigration status.

With the cost of an estimate of £500 to £600 million, the upcoming EU Settlement Scheme, it will equal to one of the largest endeavours in the history of the UK Home Office. Even with a Brexit deal in the pocket, most EU nationals - like your author - feel wary. With no deal, it could get far worse. MPs' non-binding vote against a no-deal Brexit recently brings little solace. The PM clarified that the prospect of crashing out of the EU is not off the table'.

No-deal could be a nightmare for how the settlement scheme is executed. Not without reason, are there fears that no-deal could stifle capital investments by the government that finances campaigns to get everyone to apply, especially those groups at risk. In addition, in such a world, entitlement to social security aggregation from previous periods in the EU and pension aggregation would cease to exist - a considerable issue for EU nationals with previous work history in EU countries (an unappreciated burden that future governments may have to pay dearly for). Also, apart from a more sudden end of free movement, only EEA citizens and their family members (including the Swiss) are allowed to come to the UK.

Concession on scraping the fee which took a great load off many minds, do signal: The government may comprehends the perils of missing vulnerable groups and it may reconsidered high application barriers after all. Axel Antoni from the3million campaign explains that it will sensibly remove one major hurdle for vulnerable groups. 'It will take thousands of EU citizens out of the risk of ending up with no status and [removes] the need to have funds to apply. Groups that earn little or nothing, due to disabilities or are economically squeezed, now having an easier route', he says.

Critics doubt that the true reason - showing in her long deferment to finally abolish the fee - is really to be found in her empathy for those groups at risk. She would have waited to play her cards right at the climax of negotiating a deal with the EU. After all, it would only exhibit a master’s diplomatic gambit to leave certain options open - to be deployed for last-minute concessions.

First of April - Now what?

From the 1st of April - independently whether you refer to EU nationals as expats or immigrants - both will need to apply for settled status, via an Android app. An iPhone app is not in the planning. Apple was reported to have decided against enabling its technology to read the chip on modern passports. With nearly half users in the UK estimated to use an iPhone, the extra work will drop on the heads of Home Office caseworkers.

EU nationals with sufficient evidence for UK residence - a bank account, an address, a national insurance number, a passport, a work history, an android phone, no confusion with previous permits and no fear to deal with the authorities - are are anticipated to sail through the process without trouble.

On the contrary, EU migrants experiencing language problems – estimated at a quarter of a million non-Irish EU adults who struggle with English – might hit a wall. The government pledged to translate the EU Settlement Scheme overview document into 23 EU languages. Yet, despite the goodwill it is unclear if the application itself will offer other languages. Mr. Antoni says there is no sign of it.

Earlier studies have shown that around 19% (first trial) experienced difficulties with some of the terminology in the application. 9% agonized over inputting their names, sometimes due to different naming conventions. Language proficiency, better mental health, good education and deft with technology is unlikely to fail young Londoners.

Young Londoner especially, might have also the advantage of a larger social network. It could be easier to share Android phones if there are many to call on if one has only iPhone to offer. To borrow their friend's device for applying was mentioned as a suggestion by the administration. How feasible this advice really is, remains uncertain. Meanwhile, media coverage gravitates towards younger, educated and outspoken EU nationals in London. Few appearances are made by individuals from vulnerable groups such as the elderly, children, or EU citizens living in isolation.

Evening Standard story: 'The EU 27 speak: adopted Londoners tell how Brexit has already changed their lives' – Millennials in the focus. So far, vulnerable EU groups in rural or poorer areas received little attention by the media how Brexit affect their lives.

Being grown-up but young, agile, highly-educated with residence in a major UK city, all diminish chances to run aground in the application process. Recent trial results suggest that applying might be as simple as ordering a burger at McDonald's – four out of five completed the process within 10 minutes.

However, there are signs that it might not be as smooth. Figures by non-partisan think tank British Future found that if only 5% fail to apply or are refused, 175,000 could end up with an insecure or no immigration status - roughly the size of Rome in 1850.

An analysis by your author shows that if only one in ten EU nationals categorized to be at peril, misses applying, it would raise the bar to 231,200. This volume merely constitutes 6.6 % of the total but compares substantial next to potential Windrush cases: 57,000 Commonwealth migrants according to the estimates by the Migration Observatory at the University of Oxford.

Fear of applying

Studies show how women suffering violence could be affected. A report by think tank Freemovement.org comments on the legislation of the EU settlement scheme: "The suitability grounds, especially at EU16(a), are likely to adversely affect vulnerable individuals such as victims of trafficking or domestic violence. This would have an indirectly discriminatory effect on women".

Conceivably, some overlaps in groups may curtail the base. But one in ten may even be on the conservative side, says Make Bohn from the3million: 'A recent trial with the Roma resulted in only 3% of people who didn't need assistance but of course that doesn't mean that these people won't get the status in the end, if assisted properly'.

Fear-mongering might be the Home Office's first and obvious choice to nudge and motivate the broader base of EU nationals into appling. It might even work for the large part of it. For those without a status after June 2021, Mr. Antoni expects contingency plans to appear. It would be likely that ‘even in a hostile environment, in reality, whoever didn't apply after June 31st will still be able to do so', he says. However, employers, banks and landlords will simply refuse those without a status' – - arguably a debatable incentive and a risky bet to catch the remaining few. The amount of discrimination against EU nationals, who fail to apply, would also increase, but would not kick in all at once, rather gradually, after June 2021, he says.

Background: The 2016's Brexit vote

To support her Brexit deal, the PM announced plans to woo MPs with cash injections to deprived areas that supported Leave. The debate on how much vulnerable local EU nationals can expect, leads this investigation back to the Brexit vote, nearly three years ago. Leave share at the Brexit referendum per local area is one piece of the puzzle explaining how an influx of EU nationals into areas – already feeling squeezed and suffering lower living standards – deepened the level of fear and concern among UK citizens.

EU Immigration concern played a fundamental role in the Brexit vote. Anti-immigration sentiments went hand in hand with rejoice for leaving the EU. It came down to what UK nationals, mostly poorer ones, felt. Earlier work shows that immigration, per se, was in fact negatively correlated with share in Leave-vote.

For the discussion to prevent EU citizens to remain unaccounted for after the hard deadline, it could depend on how much responsibility and control the Home Office passes on to local areas to 'help' the between 10% and 30% who are predicted to need some sort of assistance - e.g. via a human caseworker. Also, how much funding there is. To find sufficient financing for local areas will be difficult and is complicated by absent capital that could have cashed in from application fees. In this light, bespoke additional local support - beyond London - remains uncertain.

How support for helping EU nationals apply could falter in poorer, Leave-vote dominated local areas

EU-15 countries, which were part of the EU before 2004 (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden, United Kingdom).
EU-8 countries, which joined the EU in 2004 (Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, Slovenia)

NRS social grades by the UK demographic classification system ranging from A to E. The graphic considers the share of those made up by D and E – consisting of semi-skilled/ unskilled manual workers and state pensioners, casual and lowest grade workers, unemployed with state benefits only.

Where support for locals assistance could falter: Poorer, Leave-vote dominated local areas

A British Future analysis on immigration found that people might comprehend immigration as something positive on a national scale, yet inclined to feel more concerned about it on a local level. Where in the UK vulnerable EU individuals reside - either by choice, faith or force - might matter in how much support in the application process they can hope for from the hands of local constituencies.

Personal assistance for struggling applicants in the form of local help-desks could resolve errors and misunderstanding more swiftly. The fear is that locals’ opinions towards EU immigration could cloud judgment – and in its worst, diminish the level of assistance.

On the contrary, to rely on the 2016's vote entirely in judging immigration sentiment may not be entirely fair. This has to do with a witnessed change in how UK citizens perceive immigration, today. With a heavy drop from a little more than the majority to now 16 per cent - far fewer Britons than before the referendum say they are concerned about immigration. This could produce some relief.

On the other side of the spectrum, there is London. Much support is expected to be available here. There will be few EU national Londoner requiring help. Its mayor is lobbying for a second referendum and large companies employ thousands of highly-educated EU nationals - with employers pledging to be of assistance in any possible way. Here, problems for some to fall astray prior or during the process are expected to be minimal.

More complicated cases – for the lack of work documentation or other reasons –could originate elsewhere. If the Home Office can read the signs, it would be prudent in exercising greater caution and support in locations where average earning levels are much lower, the share of EU8 country immigrant higher and the proportion of lower social grade population greater.

Data quality:

The first step to motivate millions to apply by a deadline is to have exact numbers. A whopping 91,000 in margin of error for the estimate number issued by national statistics for the whole of the UK - and hypothetically could fill Wembley Stadium to the upper ranks. But, confidence intervals are not fixated for areas the same way and can substantially vary across locations.

Errors

A recently held second dry-run by the government to test the EU settlement scheme included 30,000 civil service workers. PB2 – an acronym for 'Private Beta Phase 2', covered 0.86% of the population in question. Some criticized that the sample failed to feature any vulnerable groups outlined in the first section.
While the government’s press release used the word 'success' six times to express its triumph (no-one was denied), criticism goes beyond denial-rate and its upbeat rhetoric.

One of critics' major concern has to do with testing on civil service workers. The majority is expected to have a cleaner slate than the average EU national. Records are likely to be more in order. They might have passed a security check. Many will have a continuous work history and a bank account. None of the workers at the trial were allowed to apply for any family members – which are expected to complicate cases.

Issues with names and hacks

While it is true that ‘technology may only be as good as its user', there are concerns that infrastructure could take a hit and fail due to peaks in demand. Neither are hacking attacks out of question. With over 500 different types of android device tested in the trial from 52 different device manufacturers for verifying applicants via the app, it may leave the wrong people a broad field of attack.

Failure as a result of the collected data includes name convention problems – major concern such as for German umlaute (ä, ü), and similar, resulting in incongruities in matching names to official records. A first pilot threw errors when applicants' names could not be properly processed - names on passports didn't match HMRC records. 9% had trouble inserting their names, according to the British Future report.

Technology errors: When the computer says No

For 10% of cases that intended to use the mobile app for the ID document check, the system failed. In those cases, human caseworkers needed to intervene. 16% needed to provide additional documents – translating into over half a million EU nationals.

So far it remains unclear if the Home Office could lift the workload. A point in its favor: It did pledge to hire more people - an additional 1,000 of caseworkers. To have more personnel is good news. On the contrary, they also need to work at the right moment in time. It ties into a major issue. How to prepare for inconsistent demand?

Congestion: Sudden onrush expected

Sudden onrush of demand could quickly flood the service. Will the app’s infrastructure withstand a cascade of hundreds of thousands of applications, coming all at once? Recent trials, due to their limited size, give no clue on the matter. Demand may vary and diverse scenarios are conceivable.

The unlikeliest of all is perhaps one with equal volume over time - this alone could be nearly 130,000 applications per month. A simple ‘exponentially decreasing’ model believes in encountering the highest load in the first month, around half a million applications. If this were true, demand would gradually abate, reaching a mere trickle of around 20,000 per month by June 2021.

Migration Observatory argues in favor of fluctuating volume. In addition to the initial onrush at the start, they also expect a major peak shortly before the deadline. At play here: salience and deadline thinking, - behavioral economic theories, determining demand.

In this model, the last minute rush would translate into roughly a quarter of a million applications in the final month. The truth is likely to lay somewhere in between. Considering different scenarios could aid hiring caseworkers and prepare to avoid being smitten by peak periods. A possible result of negligence in this regard? Next to the Home Office reputation, the service could suffer irreparably and leave many applicants beyond frustrated. It could also bear a heavy weight on Home Office staffs' morale.

In the following, a graphical representation of three scenarios until the dealine - number of applications ('000):

Unpublicised information emerged suggesting that a share of applicants may have received a status they did not anticipate. Sajid Javid, the UK Home Secretary mentioned at the EU Justice Sub-Committee on 22nd of January: that 'Over 90 % of decisions have been made as expected'.

Inversely, this means roughly 10% (or less), extrapolatable to around 350,000, that could receive unexpected feedback – many of those to be granted pre-settled status instead of settled status (also possible the other way around). An interactive game puts you in the shoes of the Home Office. Can you dodge the 10% of cases that grant applicants an unexpected status?

Impossible to win? Play the UK Home Office (Please click!)

Can you mind EU immigration cases where applicants would receive unexpected results?

Can you locate the 90% of applicants who received the correct immigration status they anticipated, and avid those who did not (in the form of mines)?
Tip: like in real life, not all cases are straightforward. For some, you might need to work harder (e.g. only receiving the right results after several repeated applications). The Home Office stresses, that their findings at this phase cannot be used for extrapolation. This is worrisome, as only two months remain before the scheme opens.

Conflicting data

Many of the 3.5 million EU nationals are children, born in the UK or immigrants to previous periods more than 40 years ago. According to the numbers, around 2.3 million, may have not immigrated recently. A calculation mixing various open datasets throws up important questions on how reliable official estimates really are. 2.3m is the result of subtracting the cumulative count of net migration since 1975 from the total number of EU nationals.

Many of those will be children

A considerable spike in birth rate by EU national parents in 2008 was noticeable, says Mr. Antoni. 25% of children born in 2008 to 2010 were of EU national parents, despite the fact that they only make up of 5% of the UK population. ‘Young central Europeans arrived as couples to the UK, settling down after 2004, feeling so comfortable that they decided to have children. Back then government – notably Labour - seemed enthusiastic about it. It would have propped up a poor birth rate. But the feeling lost its force by the time of the financial crisis, when birth-rates fell back to normal’.

Timely and informative data sources are in short supply, argues the Migration Observatory. Urgently required but absent: data that exposes EU national's residence status. Survey questions inquiring about individuals' plans to stay in the country should resolve discrepancies, too. It would already be a common practice in countries like Canada and in English Commonwealth Australia.

For those who are hoping that the upcoming census in March 2021 will bring clarity, crowed too soon. An updated count on how many EU nationals lead their lives in the country at present will not become available before the end of the grace period in 2021. By then, it could be too late. Another general issue is that people are unlikely to be asked by the census about their nationality (it did not inquire about it in the last census). In addition, ONS prediction might even grow less reliable. Documentation quotes a decline in the response rate for the Labour Force Survey – an essential device to count EU nationals within the labor market and elsewhere.

Lastly, ongoing surveys may also become less reliable, as OM reports. The number of respondents answering all the questions in the Annual Population Survey questionnaire (APS) fell in recent years.

Keeping records clean by offering transparency

Accuracy in individual records is vital. Barring EU nationals from accessing their own data which the government keeps of them, complicates matters and is widely seen as bad practice.

When data with the Home Office is, in fact, incorrect - whether mucked up by the government or not - cleaning such errors would be much harder and take much longer, if it even comes to that.

Some even fear that if the Home Office - left to its own devices and unable to obtain accurate records - could feel motivated to explore more drastic measures. This could include accessing citicens' alternative data - such as checks on individuals' smartphone device data, geo-locations, membership data or traces, citicens leave on the web.

The old and the young

Children and older EU nationals, as quoted by multiple sources, are at greater risk to reliably apply. Older generations tend to have poorer digital literacy skills, which can affect the way they use online government services. For the young, the limiting factor appears to be out of employment, education or training. As a result, they may lack the required paperwork. One source of remedy are school records, vouching for a child’s UK residency.

To aid the old, a British Future report says that the Home Office is considering giving extra face-to-face or other assistance to vulnerable groups such as elderly care home residents.

EU nationals in your local area

Find means that help EU nationals who need assistance to apply, could rise as an agenda point on government's overstuffed To-Do list, as early signs suggest. Alongside the PB2 trial, the Settlement Resolution Centre was opened to accept email queries since last October as well as phone calls.

Applicants without access to an Android device were offered - by thirteen locations, mostly, Local Authorities - an identity document scanning service to complete the identity verification process via the EU Exit: ID Document Check app”. Sadly, EU applicants had to pay a charge for the service - another limiting factor to those who can not afford it.

Ben Heubl is a data journalist, previously with the Financial Times and The Economist. This piece is independently produced. Great thanks for answering neverending questions go to Axel Antoni, Maike Bohn, Madeleine Sumption and Robert McNeil and to bali_balo for an idea to present Minesweeper in CSS.

Datasources: ONS, Migration Observatory, the3million campaign

Looking for a skilled data journalist? Reach out via email.