Immigration and border control signs at Edinburgh Airport. Photograph: Getty Images.
Show Hide image

Is illegal immigration really "getting worse"?

Contrary to Yvette Cooper, the evidence suggests that the problem is less severe than commonly thought.

The shadow home secretary, Yvette Cooper, recently made a major speech on "Labour’s approach to immigration". This blog isn’t about the speech as a whole (much was fairly sensible on the benefits of migration and the damage being done by government policy in some areas, but it lacked substantive policy proposals to improve matters). Instead I want to focus on one specific paragraph, because it exemplifies the sort of alarmist, yet evidence-free, rhetoric that is increasingly common from politicians of all parties on this topic:

"At the same time illegal immigration – which isn’t included in the target – is getting worse. More people are absconding at the border, fewer are being caught and sent home, and the number of people here illegally is growing."

What does she mean by this? It appears to be true that more people are "absconding at the border" and removals are down. But so what? Only a very small proportion of those not legally present in the UK enter illegally - the vast majority are asylum seekers whose claims (and appeals) have been denied, and those overstaying their visas. Are there more of those? The fact that (slightly) fewer such people are being deported could indeed mean that there are more who aren't. But it could equally, and just as plausibly, mean exactly the opposite, if the reason is actually simply that there are fewer out there.  

So what does the evidence say? Is it correct that "illegal immigration..is getting worse..and the number of people here illegally is growing"? The last serious estimate of the number of those illegally resident in the UK was made in 2009, with data to the end of 2007; the estimate then was for a range of 420,000 to 860,000, with a central estimate of 620,000; this is the number most commonly quoted, although Migration Watch, assert the number is more likely to be about a million, by adding an arbitrary half a million overstayers and illegal entrants.

But what has happened since 2007? The 2009 report, itself based on an earlier Home Office study, found that the most important group of illegal residents was failed asylum-seekers, resulting from the period of very high flows in the early 2000s, and the resulting breakdown in the administration of the asylum system. Since then, however, while the flow of asylum seekers has remained fairly stable at much lower levels, a major regularisation programme, beginning in 2007 – amnesty in all but name for well over 100,000 applicants, especially those with children – has dealt with much of this issue. Indeed, as a result of this programme (begun by the previous government but mostly implemented by this one), in 2010 more immigrants were granted permanent residence in the UK than ever before.  

Although it is almost impossible to make sense of Home Office data on this topic, this Home Affairs Committee report suggests that of about half a million unresolved cases (in some cases, people whose claims had been denied; in many other cases, people who'd been lost in the system) more than 70 percent have been resolved (either by giving the applicants leave to remain; or because they had already left the country; or because the files were errors or duplicates). In any case, these cases no longer represent people who are here without legal residency. It is impossible to say how that affects the numbers  - but it does seem probable, therefore, that the number of failed asylum seekers is now considerably lower than in 2007.

This effect also shows up in aggregate immigration data. In the period 2001-07 inclusive, net non-EU migration to the UK was more than 1.5 million; during the same period, about 1 million people were granted settlement (indefinite leave to remain or similar). While the discrepancy is not in any sense a measure of irregular migration (there are lots of reasons why someone might immigrate here and not show up in contemporaneous settlement data) it is indicative of the overall balance between the number of people moving here and those given legal residency. By contrast, in the period 2008-12 inclusive, the two figures were not far apart, at somewhat over and just under 900,000 respectively. This does not look obviously consistent with a growing irregular population in the latter period.

Similar confirmation that there is not a large illegally resident population that has somehow been missed is provided by a comparison between the immigration data from the International Passenger Survey (IPS) and the 2011 Census. The fact that the immigration data "missed" a lot of EU migrants has attracted a lot of attention; what has been less well publicised is that the IPS-based estimates for non-EU migration turned out to be surprisingly accurate, with hardly any discrepancy with the Census. If Migration Watch was right about the extra half million  (who would not have been counted by the IPS) then where exactly are these people now?

Finally, although it is not probative, another piece of evidence suggests that the number of illegal residents may actually be rather lower than we thought. Data from the Metropolitan Police continue to show, as I first noted here, that even when immigration officers are directly involved in the processing of all those arrested in London, immigration action is taken against a remarkably small proportion: slightly under 1 per cent. More detail in my article, but unless we believe that irregular migrants are remarkably law-abiding compared both to natives and legal immigrants (who offend at approximately the same rate as natives) then it may be that levels of irregular migration are considerably lower than previously thought. Again, Migration Watch's "estimate" that the illegally resident population of London is about half a million (more than 6 percent) looks implausible in this light.

To conclude: for obvious reasons, it is very difficult to be definitive about the number of people illegally resident in the UK, let alone whether the problem is getting "worse" or "better". Given that illegal and irregular immigration is, quite rightly, a sensitive issue among both native Britons and legally resident immigrants, this alone should have given Yvette pause before making a definitive statement on the topic.

Moreover, the balance of the evidence and data that we do have suggests that, if anything, the problem is perhaps rather less severe than commonly thought. Politicians, Yvette no exception, often say that "politicians didn’t talk enough about immigration" – to the intense irritation of those of us who’ve been talking about little else for the last 15 years.  But feeling that you need to talk about something is not an excuse for getting it wrong. 

Jonathan Portes is director of the National Institute of Economic and Social Research and former chief economist at the Cabinet Office.

Cameron in Nuneaton. Photo: Getty
Show Hide image

Why fewer of us want a long-term relationship ... with a political party

In 2015, 38 per cent of voters backed a different party to the one they supported in 2010. So what does the rise of swing voters mean for British politics?

For decades political parties have competed furiously for one of the great prizes of British politics: the affections of the swing voter. It wasn’t that long ago that there were relatively few political swingers: until the 1990s, fewer than a quarter of voters would switch parties from one election to the next.

Yet that once relatively rare breed is becoming increasingly common, which means party campaigners are going to have to come up with new tactical thinking. The British Election Study survey panels, conducted episodically over the last fifty years, are unique in that they are able to track the same voters from one election to the next, unlike more conventional opinion polls that only look at a snapshot of voters at a given time. Using these studies, you can identify the percentage of voters who switch their vote from one party to another between each pair of elections since 1966 when such data was first collected.

In 1966 only around 13 per cent of voters had changed their minds since the previous election in 1964. Since then, the proportion of swingers has been steadily increasing, and by 2015, 38 per cent of voters backed a different party to the one they supported in 2010.

The increase in swing voters is pretty consistent. The only exceptions are between February and October 1974, when (understandably) fewer voters changed their minds in eight months than switched in the preceding four years, and between 1997 and 2001, when the electoral dominance of New Labour under Tony Blair held back the tide for a time. These two exceptions aside, the increase has been constant election-on-election.

A lot of vote shifting can go on even between elections where the overall result remains stable. In 2001, for example, more people switched votes than in any election before 1997, with a surprising level of turmoil beneath the surface stability. While these largely cancelled out on that occasion, it set the stage for more dramatic changes in the parties’ votes later on.

So British voters now seem more likely than ever to jump from party to party. But who exactly are these swingers? Are they disillusioned former party loyalists? Or have British voters simply stopped getting into a serious relationship with the parties in the first place? We can get some insight into this using data from the yearly British Social Attitudes Survey, looking at the number of respondents who say that they do not identify with any of the political parties (party identifiers tend to switch much less often) when they are asked ‘Generally speaking, do you think of yourself as a supporter of any one political party?’ and then ‘Do you think of yourself as a little closer to one political party than to the others?’ if they say no to the first question. The graph below combines data from 1984 to 2013. Each line represents people who were born in a different year. Higher lines mean that there are more people who do not identify with a political party. So, for instance, voters born in 1955 started with very low levels of non-identification (22 per cent), which have gradually risen to 44 per cent in the latest survey. Most of the lines on the graph go up over time, which shows that almost all generations are falling out of love with the parties.

However, an acquired taste in swinging among the older generations is dwarfed by the promiscuous younger generations – shown by the dashed lines – most of whom never form an attachment to a party at all. Each generation in the data has been less committed to the parties than the previous generation was at the same age, with around 60 per cent of the youngest generation – those born since 1985 – expressing no attachment to any political party.

Since most of this change has been a generational shift, it may be a long road back for the parties. Loyalty to parties is often handed down in families, with children inheriting their parents’ commitment to a party. Now that this process has broken down, and younger generations have lost their attachment to parties, they may in turn pass on this political detachment to their children.

The majority of younger voters have simply never grown up with the idea of getting into a long-term relationship with a political party, so they may never settle down. Many Labour MPs were outraged when it turned out that lots of the new members who joined up to vote for Jeremy Corbyn had voted for the Green Party just a few months before, but this may simply reflect the political approach of a generation who see parties as needing to earn their vote each time rather than commanding lasting, even unconditional loyalty.

If Britain’s newfound taste for swinging isn’t going to disappear any time soon, what does it mean for party competition? In the past most people had settled partisan views, which seldom changed. General elections could be won by attracting the relatively small group of voters who hadn’t made up their minds and could very easily vote for either of the two main parties, so political parties based their strategies around mobilising their core voters and targeting the few waverers. While they worried about traditional loyalists not turning up to the polls, the parties could be assured of their supporters’ votes as long as they got them to the voting booth.

Nowadays, swing voters are no longer a small section of the electorate who are being pulled back and forth by the parties, but a substantial chunk of all voters. This helps to explain why politicians have been so surprised by the sudden rise of new parties competing for groups previously thought to be reliable supporters. The new parties that have entered British politics have also allowed voters to express their views on issues that don’t fall neatly into traditional left– right politics such as immigration (UKIP) or Scottish independence (the SNP). This in turn has posed a dilemma for the traditional parties, who are pulled in multiple directions trying to stop their voters being tempted away.

This may just be the start. If the number of swing voters stays this high, the parties will have to get used to defending themselves on multiple fronts.

This is an extract from More Sex, Lies and the Ballot Box, edited by Philip Cowley and Robert Ford.