Friday 19 August 2011

Youth unemployment and beating up on migrants

Colin James wonders whether more migrant workers on temporary work visas may have worsened outcomes for youths:
That ignored that thousands of foreigners are brought in to do low-skill jobs beneficiaries cannot or will not do, because they lack the skill even to turn up for work and stick at it. Labour itself recognised this by starting the temporary work visa system when in government.
Sir Roger quoted Canterbury University economist Eric Crampton's estimate that "up to 13,100 young people are out of work as consequence of the youth wage being abolished; in 2008. Sir Roger, too, seemed oblivious to the temporary visa phenomenon.
First, it's likely the case that workers on temporary visas are complements to domestic labour rather than substitutes for it. See the work of Giovanni Peri for starters. Then read this: massive unskilled migration from the Chinese hinterland to the cities has, if anything, positive employment effects for urban workers. Cowen talks here about how immigration creates more jobs. Here's Dustmann, Hatton and Preston in the EJ, summarizing articles in a feature issue:
His [David Card] conclusions on employment and wage impact are in line with much of the previous literature: although immigration has strong effects on relative supplies of different skill groups, local labour market outcomes for low skilled natives are not much affected by these relative supply shocks.
Then also see this nice NBER survey:
The large majority of studies suggest that immigration does not exert significant effects on native labor market outcomes. Even large, sudden inflows of immigrants were not found to reduce native wages or employment significantly. Effects that do exist tend to be relatively small and concentrated among natives or past immigrants that are close substitutes (e.g., Okkerse 2008). Overall, the limited substitutability of immigrants for natives in many European economies would suggest that displacement effects are likely to be small.
For example, Angrist and Kugler (2003) found that European labor market rigidities exacerbated the negative impact of immigration on native employment. These labor market rigidities include, for example, centralized wage setting that does not allow for downward wage adjustment and restrictive employment protection laws.
So where we get employment displacement effects, small though they may be, it's through interaction with labour market rigidities that prevent downwards wage adjustment. What's a high youth minimum wage? A labour market rigidity that prevents downwards wage adjustment. So even if James is right that temporary workers are a factor, it's likely that they're only a factor because of the crazy high $13/hr youth minimum wage.

Ok. Now let's assume that all that empirical lit is wrong and that immigrants have strong negative effects on native employment outcomes. Now, correct me if I'm wrong here, but I'm pretty sure that New Zealand admits more temporary migrant workers when the economy's booming and tightens up when we're in recession. In that case, youths are now facing less competition from unskilled migrant workers than they were facing in the mid 2000s. I'd expect that this attenuates my results rather than leading to their being overstated, but I'd need the time series to check things properly. You could imagine things going the other way if the flow of migrants dries up but there's a big stock. But these temporary visas are only good for two years. So if they started being clawed back when the recession really hit end-2008, that stock would be pretty low by end-2010. And youth unemployment outcomes are worse now relative to adult unemployment than they were end-2008.

It's not particularly plausible that temporary migrant workers are the cause of any substantial portion of current youth unemployment outcomes.

Colin James has one interesting quote from the PM though:
Key discounts (though does not dismiss) the youth wage factor and notes his offsetting policy that employers can hire young workers for 90 days without risking an unfair dismissal claim.
Again, that attenuates the effects. In the absence of the 90-day bill, outcomes for risky workers (like youths) would have been even worse.

I'm more than open to that there are plenty of other things that could have caused the massive ramp-up in youth unemployment relative to adult unemployment. Something really weird happened starting end 2008. I've suggested the abolition of the differential lower youth minimum wage as being the most likely culprit. Labour's been pushing changes in apprenticeship and training funding as alternative explanation, but there's no way the effects there are big enough to explain things and the time pattern doesn't fit well. I don't think James's blame-the-migrants explanation works either.


  1. Hi Eric,

    Immigration New Zealand keeps statistics on work visas. If you look at W1 (PDF)—which I understand includes seasonal work—you will see that the total number of visas peaked in 2007/08 at 190,040 and there has been a small but constant decline since then, reaching 171,898 in 2010/11 (see page 5).

    There is also information for W3 visas (I'm not sure what is the difference) but they also peaked in 2007/08.

  2. Luis bringing it once again.

    Angling for a deputy-poster/house-researcher role? I can foresee only good things coming from that.

  3. @Kimble I was angling for house-researcher, but Eric is still offering less than the minimum wage ;-)

  4. @Luis: Thanks! I'd had a quick flip through the Immigration site but hadn't seen it.

    So there are 20k fewer competitors for young workers. That's consistent with effects I'm finding not being contaminated by changes in immigration visas; it's always possible through that youth unemployment outcomes would have been even better had that dropped further (ignoring all the empirics on potential complementarity...)