Immigration and the macroeconomy

Eric Crampton, the strongly pro-immigration head of research at the strongly pro-immigration New Zealand Initiative drew attention to a nice summary (on a high profile portal for the wider dissemination of economic research) of a new empirical paper by a couple of researchers at the Norwegian central bank looking at the macroeconomic implications of immigration in Norway.    It looks as though a draft of the paper may have been presented in New Zealand, and at very least the authors thank a couple of Reserve Bank researchers for comments, and cite a couple of  Reserve Bank research papers.

Unlike most European countries (apparently) but like New Zealand, Norway has good quarterly immigration data.  Historically, Norway had quite low rates of immigration, but over the last 15 years or so there has been a substantial increase, mostly stemming from the liberalisation of migration flows within the EU (and closely associated countries, including Norway).  By New Zealand standards, the net inflows are still modest, and annual population growth peaked at around 1.3 per cent, compared with a 2.1 per cent peak recent annual increase in New Zealand.

Figure 1 Annual change in Norway’s population (%)furlanettofig1

One of the challenges in assessing the economic impact of immigration is assessing which bits are a response to domestic economic developments, and which are more genuinely exogenous.  There aren’t easy or foolproof ways of doing that (after all, even when there are discrete policy changes –  as with the recent change in New Zealand’s target level of residence approvals –  those changes aren’t made in a vacuum, and may be influenced by the actual or perceived state of the economy).

The latest paper uses a particular empirical “identification strategy” to try to distinguish endogenous from exogenous changes in the immigration  flows.  It looks a plausible enough approach, but it is hardly the last word on the subject.   There are two other features of the paper to note, both of which should make it more likely that the authors will find positive results.

The first is that the authors look only at immigration from “western countries” (EU/EFTA, North America, Australasia and Eastern Europe).   As they note, in Norway “immigrants from non-western countries exhibited an employment rate substantially lower than natives, and migrated to Norway mainly because of family reunification or as asylum seekers”.  The second is that the analysis uses “mainland GDP” –  which excludes oil production and associated transport.  Doing so is common in short-term macroeconomic analysis in Norway, but tends to skew the results towards finding positive results from immigration –  since the oil resource is a fixed factor, and a major contributor to Norway’s overall economic prosperity.  Every increase in the population spreads that particular fixed resource across more people, lowering the average per capita output from that sector per person.

So what do the authors’ find as a result of their modelling?  Using quarterly data, they examine the response of a number of variables over periods up to 36 quarters (9 years) after the initial immigration shock.  Here is their chart.

Figure 2 Responses to an exogenous increase in immigration


When the grey shaded area encompasses the red (zero) line, there is no statistically significant effect found.

The results largely rang true to me, and are consistent with the longstanding approach of New Zealand economists to the short to medium term impact of immigration.    Unsurprisingly, an immigration shock boosts GDP –  to a statistically significant extent –  over the first year or so.  The new migrants (particularly these western migrants) are both workers and consumers, and generate a need to additional private and public infrastructure.  But interestingly, the effect doesn’t last.  On this modelling, there is no permanent increase in GDP –  let alone GDP per capita.

And consistent with the New Zealand stylised facts –  over decades, but including Reserve Bank research in the last few years –  a positive shock to immigration typically lowers the unemployment rate in the short-term (in this case two to three years), and in the long-term immigration makes no difference to the unemployment rate. That all makes sense –  typically the demand effects of additional immigration exceed the supply effects in the short-term –  after all, new arrivals need to eat (and do all or most of the normal consumption spending natives do), but they also create a demand for new houses –  which might last 100 years –  and other public and private additions to the capital stock.   In the longer-term, of course, the basic institutions of the labour market (regulations, unemployment benefits etc) determine the unemployment rate.

What about the fiscal impact?  On this modelling, it was found that additional immigration –  and recall that this study focused on western work-oriented migrants, not refugees or people coming on family reunification –  made no difference to public finances in the long-run.  There was a “slightly positive” effect in the short-term, which again shouldn’t be surprising, since –  as we saw earlier-   demand effects tend to exceed supply effects in the short-term, and strong demand tends to boost government finances.

These Norges Bank authors seemed to have started from a position of being relaxed about the economic impact of immigration to Norway.  They focus on the widely-heard arguments that increased immigration will raise unemployment and put additional strain on the public finances, and note that

Our research did not find any support for the macroeconomic arguments that have recently been used against immigration. In our model, immigrants do not limit job opportunities for native workers, and an increase in immigration has no negative effects on the fiscal balance (if anything, we find a small positive effect). It is important to stress, however, that our results refer only to immigration from western countries, and so largely capture job-related immigration.

However, reassuring (and unsurprising) as those results might be, the authors did find some more concerning results.  Specifically, GDP per capita (even just the mainland measure) appears to fall as a result of positive immigration shocks –  recall that in the long-term GDP didn’t rise and the population did –  and GDP per hour  worked also fell.   In their results, this arises from

This decline in productivity was mainly driven by a strong drop in capital intensity reflecting the adoption of less capital-intensive and more unskilled efficient technologies

As they note, laconically, in concluding their note

This may be a worry for long-term growth.

I wouldn’t want to make too much of these results. It is just one paper, and there are plenty of other papers with different research strategies and modelling techniques, that claim to find more positive results for other countries.    I’m not even sure this is the best approach to trying to sort out the long-term effects.  Then again, by focusing on only western immigration, and by looking only at mainland GDP (ignoring the fixed quantity of oil/gas), this paper gave itself the best possible shot at finding economic gains from immigration and –  at least on this methodology –  didn’t.  And these are results for a small country that – while not exactly located as favourably as Belgium or the Netherlands –  suffers nothing like the penalties of distance New Zealand does.

But there is a tendency in the New Zealand debate to (a) wipe from the memory banks all record of the longstanding scepticism of New Zealand economists about the value –  in gains in GDP per capita to New Zealanders –  of large immigration flows in the post-war decades (I wrote about one leading, and early, example of such scepticism here), and (b) discount any scepticism about the economic value of immigration to New Zealand as flying in the face of all serious literature.

The modelling approach in the Norwegian paper doesn’t really allow the authors to offer much insight on why mainland real GDP per capita, and real GDP per hour worked, in Norway might be being adversely affected by immigration, even though –  as in New Zealand –  the short-term effects on GDP and unemployment are typically positive.

As a reminder, while I am somewhat sceptical of the likelihood of large benefits to natives from immigration pretty much anywhere in the world –  unless people from a clearly more productive economic culture/set of institutions move to, and largely swamp the people of, a less successful place (the uncomfortable, but not seriously inaccurate summary of the 19th European migration to New Zealand).  Generally, I suspect immigration doesn’t make that much (economic) difference to natives.  But in some cases, it is likely to be materially harmful.  Give Wales, Nebraska, Tasmania or Southland control of their own immigration policies, and it is very unlikely that large scale immigration would leave the people of those places better off.  I suspect that New Zealand as a whole –  remote islands with control of its own immigration policy –  is also such an example.

I suspect there are two main channels through which this effect occurs here:

  • the first mechanism is the pressure that persistent large immigration inflows put on real interest rates and the real exchange rate.  The economy is skewed towards meeting the physical needs and demands that arise from a rapidly growing population, and away from outward-oriented business investment.
  • the second mechanism is through the dilution of natural resource endowment.  New Zealand’s exports remain overwhelmingly natural resource oriented (be it dairy, or coal, or wine or tourism), and there has been little sign of any dramatic transformation of that picture.  There are individual firms that succeed here simply on the quality of their people and ideas, not tied to any particular location-specific resources, but there simply aren’t many of them.   The persistent pressures on the real exchange rate make it less attractive than otherwise to develop them here, but in any case, New Zealand just doesn’t look like a natural location for lots of such businesses.  If so, we will remain very reliant on the fixed natural resources –  and ever more people, induced by immigration policy –  just makes it ever harder to keep up with, let alone catch up to, incomes and productivity in other more economically fortuiutously located countries.

It is a narrative that fits a lot of the stylised facts about New Zealand.  Of course, it doesn’t fit the prevailing “ideology” or mindset of our economic and political establishment – past and present Prime Ministers, key economic government departments, or (heavily non-tradables oriented) think tanks like the New Zealand Initiative.  But decades on there is still no evidence that their approach –  favouring lots of immigration to New Zealand, as some sort of “critical economic enabler” –  is producing economic gains for New Zealanders, and making New Zealand a more productive place.