Why are NZ economists polarised on the economic effects of immigration?

When the empirical literature is reasonably clear

There is a large empirical literature on the economic effects of immigration

The literature covers the labour market effects of migration in many countries, experiencing different sorts of migration (e.g. skilled and unskilled), from different sources and at different rates, and as a proportion of the destination country population.

The literature finds that that impact of immigration on the average wage and employment of local workers is nil or slightly positive

A relatively recent review was conducted by Edo (2019). However, because adjustments take time, the immediate labour market effects of unexpected (as opposed to expected) migration episodes can be detrimental. Immigration also can have distributional consequences. In particular, the skill composition of immigrants matters in determining their impact on local labour market outcomes. And by affecting the skill composition of the workforce, immigration can change the wage structure of the economy. High levels of immigration can also increase the overall demand for labour as migrants are also consumers of locally produced goods and services. A short useful summary (focused on the UK labour market) of labour market effects can be found at The Migration Observatory.

The impacts of immigration on the labour market depend not only on the skills of migrants and the skills of existing workers but on the characteristics of the firms that hire them and whether low-skilled migrants are substitutes or complements for the existing workforce.

The black swan, an immigrant from Australia, is now considered a NZ native. Photo: Dave Heatley

The NZ Productivity Commission recently completed an inquiry into New Zealand’s immigration settings. Using IDI data between 2004 and 2019 the Commission found that shorter-term migrants who were not on a “skills visa” were more akin to low-skilled New Zealand workers, from which I conclude that these workers are competitors for jobs. Interestingly, migrants were also more concentrated in industries where New Zealand born workers were more likely to be low-skilled. However, the study also found tentative evidence that high skilled New Zealand born workers made an even stronger contribution to output and are paid more when they work in firms with higher proportions of migrants, suggesting that the two groups have complementary skills or, at least, positive mutual sorting of these groups into higher productivity firms. The full paper was presented at this year’s NZAE conference.

But NZAE members, according to the recent survey, think differently

The 1st NAZE members survey attracted 61 respondents. Their views on the economic effects of migration were polarised.

Polarized on migration, less so on unemployment

When asked whether the average NZ citizen would be better off if a larger number of low-skilled foreign workers were allowed into the country:

• 44% agreed or strongly agreed (with more people strongly agreed than agreed); and

• nearly 40% disagreed or strongly disagreed (with three times as many people strongly disagreed than disagreed).

There was less polarisation about whether low-skilled NZ workers would be substantially worse off with the arrival of larger number of low-skilled foreigners, but divergent views nonetheless:

• nearly 64% agreed or strongly agreed that low-skilled workers would be substantially worse off with the arrival of larger numbers of low-skilled foreigners; and

• nearly 30% disagreed or strongly disagreed.

Given the literature, I am surprised by the survey results

I was surprised by both the polarisation and the strength of views – on the impact of low-skilled foreign workers on low-skilled locals or the average citizen. Unfortunately, “no significant impact” wasn’t available as an option for those who were familiar with the literature. I certainly expected more economists to say they were “uncertain”, reflecting an acknowledgement of a nuanced story, of second round effects, and differences between short run and long run effects.