4  Labour Utilisation

Migrants can affect labour utilisation through three channels: the participation rate, the unemployment rate, and average hours worked. During the first half of the 1990s New Zealand’s economic growth, measured in GDP per capita, came predominantly from an increase in labour utilisation caused by a decline in the unemployment rate and an increase in labour force participation.[8] Participation in the workforce includes not only those who are in employment, but also those who are actively seeking work. Those who are not participating include students, stay-at-home parents and those who choose not to work. Those who are economically inactive, but of working age, may be facing barriers to their employment, ultimately leading to unemployment which deprives the labour market of additional resource as well as possibly increasing welfare costs to New Zealand. These barriers to employment are effectively a brake on economic growth.

Migrants may have a positive impact on GDP per capita if they have higher participation and lower unemployment rates than natives. Migrants with lower participation rates than the native average will have the effect of lowering GDP per capita with a consequential negative impact for the native population. This section looks at the comparative rates of employment and participation for migrants, and at barriers to employment.

4.1  Rates of participation and employment

This section looks at the employment rates and workforce participation rates of migrants and compares them to the native population. Secondly, it discusses evidence on the difference between skilled and family migrant stream outcomes.

4.1.1  Employment rates

The employment outcomes for migrants with different levels of skills tend to vary although there is a tendency to converge with native employment rates for groups with similar skills over time. Winkelmann (1999) shows that employment rates for immigrants appear to converge with those of New Zealand-born residents within 5-15 years for most groups.

The differentials can persist for certain migrant groups. Poot, Nana and Philpott (1988) show that in 1981, whereas recent migrants from the UK, Australia and North America had similar employment outcomes to the New Zealand born, unemployment rates for recent Pacific Island migrants were considerably higher than New Zealand-born workers and other immigrant groups. Both Poot, Nana and Philpott (1988) and Winkelmann (1999) find that male immigrants from UK and Australia had higher unemployment at the outset of their residency but that they converged to New Zealand-born levels and then fell below native rates within three years of residence. These findings suggest that male immigrants from UK and Australia face lower barriers to employment than other migrant groups.

Analysis of the 2001 census compared to 1996 census demonstrated that the employment rates of recent working aged migrants had improved from 46 percent in 1996 to 50.4 percent in 2001 (Boyd 2003). The remainder were either unemployed (8.7 percent) or not in the labour force (40.9 percent). The greatest improvements in employment rates were for the recent prime working aged migrants which had increased by over 7 percent for both males and females. The male migrant employment rate was 72.5 percent and the female migrant rate was 52.7 percent. This is still below the levels of native born with similar characteristics.

The employment rate gap between recent prime working aged migrants and the New Zealand born had narrowed between census, although the gap was still significant at 17.8 percent in 2001. From 1996 to 2001 there was also an increase in employment rates for the native population which sit at around 80 percent. For example, the 45-49 year-old age group had the highest rate of employment at 82.4 percent (Boyd 2003).

Analysis of the 2001 census data carried out by the Department of Labour showed that employment rates were lower for migrants from non-English speaking countries.[9] Prime working age migrants from non-English speaking countries had consistently lower employment rates than prime working aged migrants from English speaking countries or the New Zealand born (New Zealand Department of Labour 2004).

The LisNZ pilot survey collected information on the labour force activities of migrants (New Zealand Department of Labour 2004).[10] The employment rate for Wave 2 in the pilot group was 62 percent, with 84 percent of skilled/business migrants employed or self employed. This preliminary data suggests that later cohorts of migrants may be experiencing improved employment outcomes. The figure of 84 percent is commensurate with the 45-49 year old New Zealand employment rate stated above.

4.1.2  Participation rates

Analysis of the 2001 census data showed that young and older recent migrants were less likely to be in the labour force than other age groups. Participation rates for recent migrants in 2001 were 39 percent. Migrants aged 15-24 were more likely to be studying than to be in the labour force, although the proportion of migrants studying had declined in comparison to the 1996 census results. Recent, older migrants had labour force participation rates which were significantly lower than those of long-term migrants and the New Zealand born. Their participation rates were 28 percent in comparison to 64 percent for long-term migrants and 66percent for the New Zealand born.

Similarly migrants from non-English speaking countries had lower participation rates than English speaking migrants or the New Zealand born. Analysis of the 2001 census data showed that North East Asian migrants had the highest non-participation rates at 51.5 percent, compared to South East Asian migrants at 33.6 percent and migrants from the Pacific at 30.4 percent. In 2001 the non-participation rate for the New Zealand born was 16 percent. These results show that a large section of the migrant population is not achieving native rates of participation and therefore may be having a negative impact on GDP per capita growth. While this would give policy makers cause for concern, conclusions cannot be drawn in isolation from other factors.

New Zealand’s overall participation rates are high relative to the OECD. But the evidence outlined above shows that there is potential to increase the participation rates of migrants. Assuming no displacement of native workers (discussed below), increasing the number of people employed in the economy raises gross domestic product by the equivalent of their production. This has the effect of increasing total output. Unemployment of migrants has a high opportunity cost in terms of total output for the New Zealand economy. There are two main levers through which New Zealand can attempt to improve migrants’ participation rates. First, the entry requirements need to be set so that those people entering New Zealand have the skills and characteristics that are in demand in the labour market, thereby increasing their chance of being employed. Secondly, there are barriers to employment that exist within the host community which may be able to be partially addressed through government policies. These are discussed further under the policy implications section.

4.1.3  Outcome differences by category

It is generally assumed in the literature that skilled migrants have more significant and enduring positive outcomes than less-skilled or family-based migrants. The United States has never had immigration flows that are focused on skills whereas this has been the focus of immigration flows to New Zealand since the 1970s. In analysing the economic benefits from immigration experienced by the United States, Borjas (1995, p6) suggests that the current small economic benefits could be increased considerably if the United States were to attract a more skilled immigration flow.

However, the extent to which greater economic benefits would be experienced from a shift towards a more skilled immigration flow might not be as significant as is often assumed. There is evidence to suggest that long-run differences in employment outcomes for skilled versus family-based migrants are not as significant as might be expected.

Cobb-Clark (2000), writing about family based and skilled migrants in Australia, finds that employment rates more closely match visa categories than participation rates do. Six months after arrival employment rates for immigrants in all visa categories were significantly lower than those who had arrived under the Australian Business Skills/Employer Nomination Scheme programs. Skill-based migrants seemed to have an initial head start in obtaining employment which dissipates over time, but does not eliminate the relative gaps in employment outcomes between different categories even 18 months after arrival. She concludes that while visa category seems to be quite important in the short-run (initial 18 months) for employment outcomes, the evidence in Australia indicates that this is unlikely to persist over the long run.

Jasso and Rosenzweig (1995) examine changes in occupation status after entry for US male immigrants selected under two different groups in the United States; family based migrants (marriage to a US citizen) and employment-based migrants (having an occupation or skill deemed to be in scarce supply in the United States).[11] As expected, they find that initially the employment-based migrants had significantly higher skills and earnings levels than the family-based migrants, even when age, employment experience and country of origin are controlled for. Their work indicated that employment-based migrants suffered substantial downward mobility in terms of change in post-immigration occupational status, and the family-based migrants experienced substantial upward occupational mobility particularly in the first five years after immigration (Jasso and Rosenzweig (1995, p108). They suggest that for the employment-based immigrants this is partly because they have limited opportunity for upward mobility due to their high skills and so it is not surprising that they experience a ‘regression to the mean’. Similarly, the family-based migrants are usually in lower-paying occupations initially so their upward mobility reflects a natural movement towards the mean.

Jasso and Rosenzweig (1995, p86) suggest that the difference in overall productivity between these groups is usually overstated and may be “small or nonexistent”. The two main reasons are first that family-based migrants have immediate access into family networks in the United States and second that family members may be adept at screening for long-term productivity as opposed to employers who may have a short-term focus. Their results indicate that in the long-run (ten years after immigration) the differential in average occupational earnings between the employment-based and family-based migrants was 60 percent of what it was when the groups became immigrants and this was a result of the different directions of their occupational mobility after admission (Jasso and Rosenzweig 1995, p88). In other words, the occupational disparity between skill-based and other immigrants in the US tends to diminish over time, although it still remains considerable. This is an important finding as in New Zealand it is sometimes assumed that the family category is a necessary, but unproductive element of the NZIP. This is because to get skilled migrants, they will often want reassurances that they can bring their family in later, and if they cannot they will be less inclined to migrate in the first place.

Similarly, Canadian research (Duleep and Regets (1992) as cited in Cobb-Clark (2000)) concludes that Canada’s relatively greater emphasis on skills in selecting immigrants does not appear to give immigrants in Canada a long-term earnings advantage.

The LisNZ data indicates significant differences by category in New Zealand. The results of that pilot indicated a significant difference between the employment rates of skilled/business as opposed to family/international migrants (86 percent labour force activity rate versus 56 percent). That data, however, includes refugee members as well which may result in an unrepresentative result for the family category labour force activity rate (New Zealand Department of Labour 2004). The data is short-term but in the future will allow greater analysis of whether the findings of Cobb-Clark and others apply in New Zealand.