One of the biggest issues in the EU referendum debate has been the question of immigration. One argument made by the “remain” camp is that immigrants make a positive fiscal contribution to the economy – paying in more than they take out.
Regardless of whether a vote to “leave” would actually make a difference to immigration levels or not, is this conclusion justified?
Below I take a look at the assumptions behind this view, and look at why a combination of those assumptions, alongside the nature of our benefits system, might mean that it’s very wrong.
But first, we need to pull out to take a view of the bigger picture. Following the 2008 financial crisis, monetary and fiscal policy in Britain has clearly failed to deliver a sustainable, productive recovery.
The national debt has risen by around £1trn since 2008 (going from around £560bn to more than £1.5 trillion today), yet GDP is only £360bn higher than it was back then. Debt that rises faster than GDP is, by definition, both misallocated and unproductive.
Mass migration, deflation, the collapse of productivity, and the ballooning of the UK’s national debt are all symptoms of this apparent failure. But what is less clearly understood is that these symptoms are all interconnected. Before the system can reset along more productive lines, we need to understand exactly why and where it has gone so awry.
The assumptions behind the immigration studies
Much of the Brexit debate has centred on immigration, so let’s start there.
In the year to the end of December 2015, 630,000 new migrants (both EU and non-EU) came to the UK. That brings the total over the last 20 years to more than 8.7 million (this is a gross figure, so excludes emigration). The figure for the preceding 20 years was just 2.7 million. These are unprecedented levels.
Academic studies claim that immigrants make a positive fiscal contribution to the economy. Even the New Scientist reported in April that “all the evidence suggests that migrants boost economic growth”.
This is fortunate, because the natives clearly don’t – government debt has risen by a cumulative £1.3trn since 2003, or £1,800 per year for every person in the UK. It would be a relief to think that immigrants at least aren’t making it worse.
But is that conclusion justifiable? How have the academics figured out that immigration makes a net fiscally positive contribution?
Economics likes to pretend that it’s a science. But the truth is that its theories aren’t falsifiable, and its statistics are usually only as good as the assumptions behind the model that they are being pumped into. So let’s take a look at those assumptions.
The main proponents of the “migration makes a positive fiscal contribution” camp are Professors Ian Goldin at Oxford and UCL’s Christian Dustmann and Tommaso Frattini.
But Professor Robert Rowthorn, Emeritus Professor of Economics at Cambridge, points out that studies which find a positive fiscal contribution only do so because they use what’s called the “marginal cost scenario” as opposed to “average costs” (see the chart below showing the migrant fiscal balance from Rowthorn, 2014).
(Source: The Costs and Benefits of Large-scale Immigration, Civitas, Robert Rowthorn, December 2015)
Marginal costs exclude large parts of government spending – such as defence, interest on the national debt, and transport – on the basis that immigrants “should only be held accountable for the extra expenditure that is the result of immigration,” notes Rowthorn, rather than spending that would have occurred anyway.
Clearly, this flatters the net contribution made by immigrants. And this “marginal cost” assumption is really only valid for estimating the immediate fiscal impact of a new arrival (the longer a person has been in the country, the harder it is to argue that they shouldn’t be ascribed a share of “other” spending).
Yet arguably, even the “average cost” assumption measure is unrealistically positive. It assumes that the average immigrant shares the same characteristics as the average native: such as contacts, language, appropriate education, earnings and assets – an assumption which is unsupported by the anecdotal evidence. Despite supposedly higher educational attainment, for example, the East German population dragged overall German productivity down for a generation.
Without the same support structures, language proficiency and cultural familiarity as natives, fewer immigrants find work than natives. And of those that do, a third work in elementary occupations and another quarter in other low skill jobs. So it is unrealistic to assume that immigrants earn as much as natives, and therefore it’s also unrealistic to assume that they pay as much in tax.
That said, first-wave economic migrants probably do make a positive contribution initially, due to the fact that they are mostly young, of working age, and single. Most studies also suggest that they don’t particularly “steal our jobs’, and nor do they push down wages much – studies for the Bank of England by Professors Stephen Nickell and Jumana Salaheen in 2008 and 2016 found that “‘a ten percentage point rise in the proportion of immigrants working in semi/unskilled services led to a 5.2% (2008) to 1.9% (2016) reduction in pay,” (in other words, it takes a huge jump in the number of immigrants working in a specific sector to drive down wages even marginally), which pretty much puts paid to the “immigrants-force-down-wages” argument.
Why debt matters in the immigration debate
However, these may not be the right questions to ask about immigration. Politicians love immigration (as long as it doesn’t lose them votes) because it adds to GDP. But what the native population should be asking instead is: does immigration add to GDP per head? Because if it doesn’t, then it’s destroying productivity.
In the 16 years to the end of 2000, the UK’s average annual real (after inflation) GDP growth per head was 2.6%, and immigration was 3.3 million. Over the last 16 years, GDP growth per head has shrivelled to 1.2%, while cumulative immigration has ballooned to 7.5 million.
If immigration was genuinely good for growth, then higher numbers of immigrants should have led to higher, not lower growth. The problem is this: although it’s true that immigration adds to growth, if this growth is funded by debt, then it is a chimera.
Think about it this way: if immigrants added 1% to the population in one year, but every single one of them started claiming deficit-funded welfare rather than getting a job, GDP would still grow by roughly 1% as a result. This is why you not only have to look at the impact of migration on GDP per head, but you also have to compare that against any rise in government debt.
UK government debt stayed at around £400bn from the end of 1995 until February 2003. However, debt then surged by 70% to almost £680bn in 2008 – and this was before the crisis. Tax receipts from the financial sector then imploded during the crisis, and debt went vertical: it hit £1 trillion by end-2009, £1.5 trillion by end-2013 and £1.7 trillion today.
What caused this? It wasn’t actually the cost of bailing out the banks, as you might assume. The bank bailouts, apart from RBS, are now showing a profit. Instead, April 2003 was the date that then-chancellor Gordon Brown introduced Working Tax Credits, which, despite the name, are nothing more than benefits for part-time workers. Then, in 2004, migration from the expanded EU took off.
Since 2007, there have been four million new immigrants (the net figure is roughly 2.2million). During that same time period, only 800,000 new full-time UK jobs have been created (a 3.9% rise). Yet the unemployment rate is now lower than it was then.
So what has plugged the gap? The answer is the rise in part-time work. A million new part-time jobs (a gain of 13%) have been created since 2007. As a result, 27% of all jobs in the UK are now part-time.
These new part-time jobs haven’t necessarily all been taken by immigrants. But if they haven’t, then it does mean that immigrants have been displacing natives from full-time employment into part-time. And the problem is that part-time workers definitely do not pay their way in fiscal terms.
So if immigrants have tended to go into part-time rather than full-time jobs – which is what the job creation numbers strongly imply – then contrary to what we are told, immigrants present a very real fiscal cost.
The real issue is not immigration, but in-work benefits
The real story here is not about immigration per se, but about benefits more generally – especially in-work benefits, which are mostly paid to those with low-skilled, part-time jobs.
The cost of the benefits system has gone up by 50% in inflation-adjusted terms per head since 2000 (from £2,200 to more than £3,300 by 2014). This is patently unsustainable.
If it is the case that immigrants are no better at paying their way than natives (and as I’ve argued here, it is possible that they may be significantly worse), then it is our out-of-control benefits system (which has already taken government debt above 80% of GDP), which is the real problem. Immigration is merely making a bad situation worse.
I’ve written about this before here at MoneyWeek, but here’s a reminder of how it works. In-work benefits kick in after a minimum of 16 hours a week of work. Coincidentally enough, the median number of hours worked part-time in the UK is 16.2.
Average part-time wages are £8.20 an hour, far below the full-time rate of £13.80. But once you add in tax credits, housing benefit and other welfare, part-time workers can net five times their wages in income. This is why, for most part-time workers, earnings are of secondary importance compared to their function as a gateway to benefits.
It is also why welfare payments to the second income quintile (and even to many in the third quintile) exceed those paid to the poorest quintile. It is also why, despite unemployment being relatively low at 5.1%, more than half the population now makes a negative fiscal contribution (ie they get more from the state than they pay in).
As data from the ONS shows (see chart below), the top quintile earns 15 times more than the bottom fifth before redistribution effects. But after taxes and benefits, this falls to just 3.7 times more.
If you then take into account the fact that as many as 50% of those in the top income quintile may be working more than 48 hours a week, while those in the bottom two quintiles work a 16-hour week, then it means that once taxes, benefits and hours worked have been accounted for, someone at the top of the UK pay scale only earns about 60% more per hour than someone in the bottom quintile.
Put in purely economic terms, the system is taxing productive endeavour whilst incentivising the low skilled to game the welfare system rather than work more than two days a week.
Why this matters
With unemployment in the UK back to pre-crisis levels, the vast majority of the net new jobs must have gone to immigrants. And of those 1.8 million new jobs created since the crisis, the majority are part-time.
Part-time workers account for almost all of the spending on in-work benefits which – along with pensions – are crowding out other government spending such as infrastructure, the NHS, education, defence, etc.
This non-welfare spending has been pushed back from 57.5% of total spending to just 46.3% in the last six years alone. To close the deficit, it would have to fall all the way down to 37%. And this is at a time when employment has been rising and so unemployment benefits have been falling.
Clearly, the problem of the benefits black hole is only exacerbated by mass immigration if immigrants either fail to get jobs or get low paying, part-time jobs which are eligible for in-work benefits. The trouble is that – despite the academic studies that assume otherwise – this does seem to be pretty much what immigrants do.
Moreover, although correlation does not imply causality, the surge in immigration has coincided with a post-crisis recovery that has been dominated by part-time jobs, a 50% after-inflation leap in benefit payments per head, and a £1.3 trillion surge in government debt.
The last three are not independent of one other, and it’s quite likely they aren’t independent of the first – the surge in immigration – either.