6. Conclusions

In this paper, we use data from the 1997-2007 New Zealand Income Survey (NZIS) to examine how employment rates, hourly wages, annual income and occupations for immigrants compare to those for the New Zealand-born. Extending previously literature in this area, we examine how outcomes for immigrants change with years spent in New Zealand in a semi-parametric manner that makes no assumptions about the time pattern of labour market outcomes as more host country experience is acquired and consider the role that occupational choice plays in explaining differences in outcomes between immigrants and the New Zealand-born.

Our preferred regression specification shows that newly arriving immigrants experience, on average, employment rates that are 20 percentage points lower than comparable New Zealand-born people, and annual incomes that are ten to fifteen thousand dollars lower. For immigrants who gain employment, occupational rank is 5 to 8 percent lower, and hourly wages are 10 to 15 percent lower than for comparable New Zealand-born workers. However, after around 15 years in New Zealand, relative outcomes have improved to the point where employment rates for immigrants are about the same level or slightly below those of their New Zealand-born counterparts, and the income difference is halved for men and eliminated for women. For employed immigrants, occupational rank is about the same level or slightly below that of comparable New Zealand-born workers after 15 years in New Zealand. The relative wage disadvantage for immigrant men remains more or less unchanged at about 10 to 15 percent lower for many years after arrival and for immigrant women has closed to within 5 percent of comparable New Zealand born women workers after 15 years.

We examined whether the wage disadvantage experienced by immigrants reflects a low return to qualifications gained outside New Zealand and found some evidence that university qualified immigrants receive a smaller wage premium for their qualifications than do New Zealand-born university graduates. However, immigrants with vocational qualifications receive a higher premium for their qualifications. Overall, the size of these effects is relatively small and allowing for different returns to qualifications does not change the implied pattern of wage disadvantage and non-convergence.

Not all immigrants experience the same adjustment over time in relative labour market outcomes. The pattern of entry disadvantage followed by subsequent improvement is particularly pronounced for immigrants from the Asian region and, to a lesser extent, for those from the non-classified regions, which consist of non-United Kingdom Europe, Africa and the Middle East (mainly South Africa) and the Americas (mainly United States and Canada). Immigrants from the Pacific region have poor relative outcomes at the time of arrival, with no improvement as they spend more years in New Zealand. University qualified immigrants recover their entry disadvantage relatively quickly, within around 10 years, whereas immigrant men without qualifications have a much slower improvement, taking around 20 years. These findings are perhaps unsurprising, since less qualified immigrants, who are not admitted under the skill migration categories and include refugees and other humanitarian migrants, and immigrants from the Pacific Islands may benefit greatly from immigration to New Zealand, even if their labour market outcomes lag behind similarly qualified New Zealanders, because the labour market opportunities in their origin country are much worse than those in New Zealand.

Overall, there is much stronger evidence of adaptation for employment rates than for wage or occupational rank. The dominance of quantity adjustment over price adjustment in the pattern of adaptation of New Zealand immigrants makes New Zealand more similar to Australia than to the United States. Antecol et al. (2003) attribute the dominance of quantity adjustment in the Australian case to relatively inflexible wages and generous unemployment insurance. The summary indicators in Table 7 show that New Zealand has labour market institutions that are closer to those of Australia than to those of the United States. In fact, New Zealand's earnings dispersion is smaller than that of the other countries listed, suggesting more limited scope for relative wage adjustments.

Table 7: Indicators of labour market institutions (2001)

There are a number of related questions that this line of research could pursue. For example, it would be interesting to examine whether the initial entry disadvantage experienced by immigrants vary with macroeconomic conditions in New Zealand, and whether this affects the patterns of subsequent improvement? (eg, as in Barth et al. 2004; Aslund and Rooth 2007; Chiswick et al. 1997). Future work could also examine whether average outcomes of immigrants arriving in different years reflect changes over time in immigration selection policies, or whether the rate of subsequent improvement is related to settlement policy settings? (eg, Cobb-Clark 2004; Edin et al. 2004).