Winner of the New Statesman SPERI Prize in Political Economy 2016

Wednesday 1 January 2014

Economics and the Immigration debate

As the storm force winds blew, I wondered to what extent the debates on immigration and austerity shared a common feature. In both cases economists might feel like someone trying to walk against high winds: it is hard, perhaps painful, and you seem to be getting nowhere fast. To be less metaphorical, in both cases the economic arguments seem to be irrelevant to the public debate, and the politicians want to go in the opposite direction to the one suggested by the economics.

I have talked a great deal about austerity before, but not about immigration. A typical example of the economic arguments is this NIESR study by Lisenkova, Mérette and Sanchez-Martinez (pdf, blog post), which models the impact of the current UK government’s attempts to reduce net migration. (As this Bruegel post shows, the UK debate is fairly typical.) Although the paper uses an OLG model, and allows for some quite elaborate differences between migrants and natives, the basic results are intuitive. As migrants tend to be younger, reducing migration reduces GDP per capita (by about 2.5% in 2060), because there are less workers for each pensioner. For this and other reasons, migrants make less demands on the state, so a reduction in migration raises government spending per person (e.g. the elderly use the NHS more) which requires higher tax rates.  One interesting result is that although restricting migration raises pre-tax wages (less labour supply), after a time post-tax wages are lower because of the higher tax rate.

In short, migration is beneficial for the economy as a whole, and for households as a whole. For a short summary of other empirical evidence, see this article by Jonathan Portes, or this from the OECD. Yet the political debate presumes the opposite. It is taken as read that migration causes all kinds of harmful effects, and the debate revolves around measures to prevent these. It is summed up by this quote from BBC political journalist Nick Robinson.

“What Jonathan Portes has helped us do is define the difference between an economist and a politician," said Robinson. "A fine economist he might be, but I suggest he would not have a chance of getting elected in a single constituency in the country. It is a widespread view that there is exploitation of the benefits system by migrants.”

So you see why I think there is a potential parallel with the austerity debate. The evidence suggests that migrants make a net fiscal contribution relative to natives, just as all the evidence suggests that austerity is harmful in a liquidity trap. However the ‘public’ believe otherwise, and (by implication) economists should get real and stop going on about evidence so much.

There is a difference, however. Government debt is not a ‘doorstep issue’, whereas immigration is. In 2010 the Eurozone crisis brought the issue of debt to the fore, but since then in most countries it has been the politicians and sections of the elite that have kept the debt problem alive (to justify austerity). The ‘need’ for austerity is accepted by the public because it is portrayed as governments doing what households do in bad times: tighten their belts. But I do not think excessive government debt is an issue that many politicians encounter when they go canvassing for votes, but migration certainly is. Equally I doubt that the current UK government would have made migration such a big issue if it wasn’t perceived as a vote winner, and if it was not for UKIP.

Does this make a difference to how economists react in each case? We plug away at the economics of course, but how do we explain why the economics seems to be ignored? With austerity most explanations involve thinking about how politicians and sections of the elite think: why they may be irrationally worried about market panics, or why debt may be a cover for other agendas. With migration the focus has to be about why large sections of the electorate believe that migration harms them.

I see three strands of thought here, although none are mutually exclusive. The first is to acknowledge that there is a natural tendency for communities to be concerned about outsiders, but blames the media and some politicians for playing on this concern. This can explain why popular concern about immigration can be so high in areas where there is very little. The second is to grant that migration may be beneficial for the economy as a whole, but acknowledge that for some the immediate (and therefore personally verifiable) impact is negative (e.g. less unskilled vacancies, lower wages, higher rents). This article by John Harris exemplifies this strand. A third takes the concern about outsiders more seriously, and talks about the benefits and costs of social diversity.

A second difference involves politics. To the extent that austerity is a cover for sections of the elite to push for a smaller state, then austerity morphs into a standard debate between right and left. This helps explain why what on the surface should be a technical macroeconomic discussion about multipliers and the effectiveness of unconventional monetary policy is in reality so politically polarised. This is not the case with migration. As this letter from members of various UK right wing think tanks indicates, restricting migration runs counter to the neoliberal agenda. Others argue that public concern about migration reflects the failure of the left to oppose (or worse, pursue) this agenda.

Perhaps I can sum things up this way. While I find the macroeconomics of austerity interesting (it’s my field), I believe the reasons why the economics is ignored are fairly straightforward and much less interesting. In the case of migration, I think understanding why the economics is ignored is much more of an intellectual challenge.

A New Year Aside

Many thanks to Alex Marsh for saying nice things about this blog, but Chris Dillow absolutely deserves the No.1 spot. To see why, read this. I would have loved to have thought of this counterfactual, but whereas I would have ended with something predictable on the economics or politics, Chris talks about the importance of luck. Brilliant!


  1. The answer as to why the macroeconomics are ignored is in the post

    "Reducing migration reduces GDP per capita (by about 2.5% in 2060"

    Does it really need explaining why such small long term and diffuse benefits are not really much of a political issue?

    Rises in housing costs and pressure on unskilled labour wages are current, and at least in part attributable to immigration.

    Further, people hate being taken for suckers. Support for the welfare state is highest when it is seen as a form of insurance or club. We all pay a small amount into the pot, and those of us who find themselves in need take out. 'Outsiders' coming along and taking out of the pot without having been members of the club and paying in undermine support for the system.

    Now we know, rationally, that the number of welfare migrants is tiny. But there will always be some, and and they will be reported in the media as they are a story. That undermines support for the welfare state.

    So if, like say Harris in the Guardian (representative of the soft left Compass), you want support for maintaining and strengthening the welfare state, you'll oppose free movement of persons (as he does).

    It has taken a while, but we can see free movement of persons aligning as an issue in the way that makes sense in left/right terms. The 'neoliberals' being in favour, the left opposed.

    [Oh, and I know some people think it is dated, but using 'less' to mean 'fewer' still grates with me.]

  2. What I find lacking in the political debate on immigration is any recognition of the possibility that we British might at some time in the future find it advantageous to be able to migrate to other EU countries in search of better paid work. Anyone remember "Auf Wiedersehen, Pet"?

    1. But which British?

    2. UKIP voters probably - I reckon Oz would have voted UKIP.

  3. Do you think the answer to why the politics and economics of immigration are out of kilter is in large part down to the unequal distribution of the costs and benefits? It seems to me that although the benefits of immigration are greater than the costs, they are also more widely distributed, both across the population and over time, than the costs. That makes the costs seem more tangible and immediate to many people - and hence they rate them as more important than the greater, but more diffuse benefits.

  4. Quantifying the financial benefits migrants bring to the economy today in terms of a saving relevant to families right now would be more effective I believe than anything that might happen in 2060. The effect on pension age for instance which would be higher if immigrants were sent home is just one easily digested example.

  5. The debates on immigration and austerity share a common feature, a focus on benefit/welfare cheats. In both cases the non-cheating general public might feel like someone trying to walk against high winds when trying to keep a focus on high quality jobs in the mind of economists and politicians: it is hard, perhaps painful, and you seem to be getting nowhere fast. To be less metaphorical, in both cases the high unemployment rate and the generations of non-working families reality seems to be irrelevant to the public debate. In other words the general public has learnt that they will be ignored if they suggest that immigration is only beneficial if the number of jobs, especially high quality jobs increases, and they know that the phrase benefit cheat will capture the attention of a politician like saying bone to a dog, so that is the argument they use to protect the inadequate number of jobs that they have from further competition.

  6. The fact that immigrants are young is not a brilliant argument for immigration: immigrants eventually grow old!! In fact there are a couple of studies (one done by the UN) which show that the aging population problem was dealt with purely by immigration, the resulting population increase would be absurd: the population would double every 50 years or so. (I’ll dig out the references later today perhaps.)

    Moreover, the aging population problem is starting to hit almost every country now. How does China deal with its aging population: take young immigrants from Britain?

  7. Very, very good comments.
    I will reiterate the point that Ms. Cheryl Lans and Mr.Mark Pack - most individuals are harmed by the competition. The fact that overall GDP goes up means nothing to them
    I could fill the page with references to declining real income.
    I listened to economists during the NAFTA debate, and I listened to politicians, about how there would be winners and losers, but we could mitigate the downside.
    That was the last time I listen and believe a politician or an economist...

  8. All studies about immigration effects on growth, always assume full employment. They never, ever, consider the effects of immigration when unemployment is a huge problem.

    Immigration DOES have a positive impact on growth assuming full employment, but we DO NOT have full employment. Why does macro-economists, like Simon, expect that dumping more workers on an already over-supplied labor-market would do any good for the average worker?

    Most people, intuitively, understands that more competition for their jobs will not lead to higher wages. So, in this case, as far as the general public is concerned, the immigration debate just demonstrates the "clueness" of economists.

    1. This isn't true. Take a look at the work from Giovanni Peri on the differential impact of immigrants on the labor market at different points in the business cycle. His work is on the US, but there is almost certainly some relevance for the argument in the UK.

      And he finds that there are differences in the rates of absorption of immigrants, and the adjustment times. But most of those tend to wash out after a 5-7 year period. Which is a decent amount of time, but not an eternity by any means.

      The important issues are not the temporal ones, but the distributional ones -- that is, how does immigration impact different segments of the labor market. And that depends on the characteristics of the supply of labor already in place; the demand for labor in that place; and the characteristics of the new labor supply embodied in the immigrants. So there isn't "an answer" to that question.

  9. It would be interesting to see Miliband attempt to change the terms of the debate on this one - unfortunately there are few signs of this ( It seems to me that the most successful election policies (/campaigns) do set the terms of the debate. Austerity in 2010, blair on education, Thatcher on Europe. Boldness could payoff, it will be interesting to see what happens later this year

  10. The fiscal impacts of immigration tend to be a wash, frankly. On the one hand, migrants tend to be younger, so they absorb less in health care than the population as a whole. Also, they contribute to public goods like defense without any real increase in those expenditures (nobody really believes that national defense expenditures increase because of the presence of immigrants). On the other hand, migrants tend to be poorer than the native born population, so their tax contributions are more modest than the native born. For a good, broad look at the mountain of data on this (and this is something that really needs more than just a few analyses to get a full picture), you might want to look at:

    Rowthorn, Robert. 2008. The fiscal impact of immigration on the advanced economies. Oxford Review of Economic Policy. 24(3): 560–580

    Take away line: "there is no strong fiscal case for or against sustained large-scale immigration. The desirability or otherwise of mass immigration should be decided on other grounds."

    I think that's probably about right...

  11. A fairly recent (last January) summary of of UK studies, that I have found very useful, is It was linked in the John Harris article mentioned in the blog. The summary is by researchers at the same university as the author of this blog, but is more cautious and nuanced in its summary of the evidence than some of those he quotes. It is particularly careful in summarising the evidence on how migration affects different groups of workers, in different places, over different time horizons and in different macroeconomic conditions. These differences could be important in linking the politics to the economics of immigration.

  12. I do not think people like Portes give us a balanced picture on immigration. The policy challenge is to employ Britain's large amount of youth and long term structurally unemployed. This is key, not only to dealing with Britain's deepest social problem, but to getting out of this deflationary spiral. I always think that for economics, historical case studies are more useful than models- particularly the type that have been drummed into the political elite. The interwar period was instructive: in many countries large movements of workers had a deflationary impact on wages, particularly at the lower end, which made recovery very problematic. China controls labour flows between regions because of the simultaneous severe inflationary and deflationary effects that arise from completely free movement of workers from rural to urban areas (eg high housing costs, intense competition among unskilled workers for low paying jobs). The high costs of housing and low wages in London, and sorry this is partly caused by high rates of immigration, makes moving from the depressed north to the affluent south a totally uneconomic proposition for a low skilled worker. The other point is, no study has definitely found how high rates of low skilled immigration, even on the scale of that from Eastern Europe, has increased the welfare of the EXISTING population (ie led to a real PER CAPITA GNP increase). For low income, low skilled workers, this is even more the case.

  13. Most of the important points have been made, but it's worth noting that economists are really bad at isolating sectoral or regional or city or town level effects of immigration. And yet, most of the "adjustment tensions" - (this might be the time to remind ourselves of Keynes - in the long run we are all dead) - most of the "adjustment tensions" occur within specific sectors of work, or in specific towns. The education and health services in certain Lincolnshire towns came under huge pressure with the first wave of EU expansion immigrants. Government funding is based on the census at heart, it was way too slow to catch up with the reality. Economists assume away these problems - but on the doorstep they impact people's opinions greatly. Economists would sound less ignorant if they could either measure these problems or at least acknowledge them.

  14. Worth adding of course that many migrant populations are used to get around minimum wage legislation (and health and safety, and more…) Polly Toynbee's recent column points to some good sources on this issue. GDP is far from the only measure of how things are affecting people.

  15. The danger too is the growing opposition to the European community; the immigration debate has become a kind of proxy for the EU exit for some Right Wingers. A real danger because the anti-Europe brigade including their right wing press could win through because the Lib-Dems, Labour, the quality press ( except FT) , BBC , all of UK business, intelligentsia/academics/economists , trade union movement... have become enfeebled & are ignoring this nasty elephant in the room. There is in other words no organised counter arguments being articulated or even debated in the main media.

  16. Shorter SpinningHugo @1: Politically it doesn't matter if immigration increases GDP per capita because it worsens the DISTRIBUTION of that GDP.

    And he has a point - aggregate GDP is a poor measure of aggregate welfare for a variety of reasons, ranging from diminishing marginal utility to the many components of full income not captured in GDP measurement.

    Though it is of course the deformation profesionelle of macroeconomists to assume away distribution, as not being their department.

  17. Nice debate on immigrants and economic politicians. In my point of view, illegally entered persons should anyway get deported, but jobholders mainly suffers a lot than economics. Right to think properly as of now situations there...

  18. While I have heard it a lot that migrants give more to the society by working more, at the same time, would these jobs otherwise not be filled by domestic citizens who are as a result pushed into unemployment because of immigration?

    What figures or research has there been to show that the jobs that migrant workers do would not be done by domestic workers? Surely, in your experience you have noticed, like me, that most immigrant workers work in lower paid, unskilled jobs?

    In my opinion, as seen in cases of mass migration such as from Syria for example, the open border policy can do nothing to prevent a tide of people from entering the UK. This is a fact. Such a tide could put critical strain on social and civil services. Therefore, I believe that there should be autonomy in his regard being returned to national governments.

    Without this, you have fragmentation of society into culturally diverse and threatened localities of deprivation, crime, rioting and disintegration of areas ino areas of hopelessness.

    People need to grow up and realize that immigration is a potential cause of crisis as well as a force for good to some degree. People need to detach immigration from cheap judgement on racism. It is not racist to call for border controls to protect the economy or society from a force that can destroy an economy and cause societal tension leading to a rise in the extremism on all sides.

    *From someone against what the EU has become.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.