Migrants and Labour Market Outcomes
MIGRANTS AND WORKING-AGE POPULATION 1981-2006
Data have been obtained from the censuses of 1981, 1996, 2001 and 2006. Because of changes in questions and classifications, all data are not comparable across all censuses, but sufficient comparability has been obtained to allow analyses of migrants and their labour market behaviour. In the first instance, we shall explore details of the patterns of migrants as a component of the working-age population from 1981-2006. The people available to participate in the labour force are the working-age population, namely those 15 years old and over.
The data from each census are a 'snapshot' at a given point in time and mainly describe whether the number of migrants of various characteristics has increased or declined. It is possible to explore the outmigration phenomenon by using age group data and data on years since arrival to find figures for gross and net migrant inflows - the difference being the migrants who left between censuses. In later sections, we shall analyse the changes in employment, unemployment and other labour market behaviours.
As with all datasets, there are some weaknesses, and the main one with this census data is that there are significant numbers of people that either have no birthplace recorded, or who are known to be overseas born but have an unknown overseas birthplace. The incidence of non-reporting of birthplace varies across censuses.
The working-age population 1981-2006
The numbers and characteristics of the working-age population from 1981-2006 are explored in relation to migrant influences on change.
Changes in numbers of the working-age population 1981-2006
Over the period 1981-2006, the number of working-age people in New Zealand increased from 2.30 million to 3.16 million, an increase of 864,000 people (or 37.6%). This is a compound growth rate of 1.29% per annum over this 25-year period.
The pattern of changes in the working-age population (WAP) between censuses from 1981-2006 has fluctuated, particularly over the last decade.
In the first 15 years (1981-1996), there was an average increase in the WAP of 163,000 in each five-year period. In the period 1996-2001, the increase in the WAP was only 103,000 over the five years, but in the following period (2001-2006), it rebounded. In the latter five-year period, there was an increase of 270,000 in the WAP. Even though this increase is much higher than the previous periods, it mainly compensates for the below-average growth in the preceding period (1996-2001).
These increases can be given another perspective by considering the impact they will have on the existing population. It is likely that the level of impact on an existing population of any increase in population will be dependent on the increase per 1,000 of the existing population. To provide this perspective, we have taken the above increases in the WAP over five years and calculated what these increases would be, expressed as the number of people increase per year for every 1,000 people in the existing population at that time.
These figures are that there was an increase by 12.8 people per year for every 1,000 WAP from 1981-1996. The impact figure dropped to an increase by 7.3 people per year for every 1,000 WAP from 1996-2001. Finally, the impact lifted to an increase by 17.9 people per year for every 1,000 WAP from 2001-2006. This analysis does put the recent trough and boom over the period 1996-2006 into medium-term perspective. The average of 7.3 and 17.9 is 12.6 and so the average annual rate of increase between 1996 and 2006 was 12.6 people for every 1,000 WAP. This is similar to and very slightly below the average increase by 12.8 people per year for every 1,000 WAP for the 15 years 1981-1996.
The 2001-2006 increase is a compound growth rate of 1.8% per annum over the five years. This major change in population growth rate was driven by increases in numbers of both New Zealand born people as well as migrants. The future profile of working-age population change depends on whether the future impact is similar to the 1996-2001 annual impact of a 7.3 increase per 1,000 WAP, or a continuation of the 2001-2006 annual impact of a 17.9 increase per 1,000 WAP. The conservative assumption would probably be that the impact would lie between the average for the periods 1981 to 1996 to 2006, i.e. between 12.8 and 12.6 per annum per 1,000 WAP.
Ignoring those with unknown birthplace, the New Zealand born population had been increasing by about 78,000 per census from 1981-1996, then declined to just 31,600 from 1996-2001. The number has now increased back to about 78,000 between 2001 and 2006. Over the same period, the migrant (overseas born) population increased by about 45,000 per census from 1981-1996. This increased to over 80,000 from 1996-2001 and to over 160,000 from 2001-2006.
(As with all census data, there is a problem in apportioning accurately the increases from each because of those who did not supply a legible census return - in this case, a legible birthplace. There is a significant number in this category, and also the behaviour is erratic, which makes it difficult to postulate which class they may belong to.)
Profile of the working-age population 1981-2006
The changes in the inter-census increases of New Zealand and overseas born people have changed the profile of the working-age population over the period, from a situation in 1981 where, of each 100 people, 82 were New Zealand born and 18 overseas born, to 2006, where 70 were New Zealand born, 25 were overseas born and 5 were of unknown birthplace.
The extreme contrast is that, of those with known birthplace, the inter-census increase from 1981-1996 had 28 of every 100 overseas born, and the 2001-2006 increase had 60 of every 100 overseas born. The comparable figures for New Zealand born were 48 per 100 from 1981-1996 and only 29 for every 100 increase from 2001-2006. (Inclusion of those of unknown birthplace moderates the swing.) Over the whole period 1981-2006, of every 100 increase in the working-age population, 44 were known to be overseas born and 40 were known to be New Zealand born, with 16 having an unknown birthplace.
This change in the profile of increase of working-age population shows clearly the need to understand the effect of migration on the labour market and particularly the behaviour of these overseas born who enter the labour market.
3.1.3 Origin of overseas born working-age population 1981-2006
The number of overseas born working-age people increased from 409,000 in 1981 to 789,000 in 2006, an increase by 380,000. Of these 380,000, there was a 162,000 increase in the latest five years between 2001 and 2006.
The origin of the various groups of these people is likely to have some impact on their behaviour within the labour market. This section, therefore, notes that, of the 789,000 overseas born aged 15 and over in New Zealand in 2006, the region of birthplace of 232,000 was United Kingdom and Ireland, 225,000 was Asia, 123,000 was Pacific Islands, 86,000 was Europe and North America, 69,500 was from countries classified 'Other' and 49,000 was Australia. This pattern had changed significantly over the period 1981-2006 (see Table 3.3).
The region with the largest increase over the period was Asia, with an additional 203,000. This figure included an increase of 79,000 between 2001 and 2006. Next largest increase was from the Pacific Islands, with 73,000, increasing at a reasonably steady rate over the period. The Other category increased by 60,000 and Europe and North America by 36,000, both increasing their contributions in the 2001-06 period. Those from the United Kingdom decreased until the 2001-06 period, when there was a reversal to a steady increase.
When we look at the profile of the inter-census increases in overseas born, we find that the Asian share of the increase has actually declined a little from 56-57% of the increase from 1981-2001, down to 49% of the increase from 2001-2006.
The other main change in composition is that, over the period, there has been a relative shift from Pacific Islands born to United Kingdom and Other birthplaces. The Pacific Islands fell from 21-29% of the increase in 1981-2001 to just 10% from 2001-2006. Other places increased to 17%, and United Kingdom and Ireland increased to 11% of the increase in overseas born from 2001-2006.
In absolute terms, as shown in Table 3.3, the working-age population in 2006 had 79,000 more Asian born than in 2001, 28,000 more people from Other birthplaces, 18,000 more United Kingdom and Ireland born and smaller numbers from other birthplace regions. This is in addition to 79,000 more New Zealand born. For the labour market to deliver to the economy, it is important to know the labour market status and behaviour of these various groups, especially the overseas born.
Scale of outmigration: gross and net increases by birthplace
The figures in the preceding sections are the net changes in number from one census to the next. However, it is possible that people from some birthplaces may be in New Zealand at one census and have left by the next census. In this case, the gross inflow over the inter-census period would have been greater than the net increase between the censuses. The gross inflow from any source must first compensate for any outflow of people from that source (referred to here as outmigration). Any further gross inflow then generates a net increase from that source. The purpose of the present analysis is to ascertain whether or not there has been a significant level of outmigration and, if so, the general nature of it.
Accurate estimates of outmigration would require a full reconciliation of each cohort tracked from one census to the next. This would enable us to track the working-age population (with a given region of birth) at one census and to estimate the expected number in these cohorts resident at the next census. To estimate this, we would add the number turning 15 in the next five years and subtract the number of deaths of those aged 15 and over.
However, the base numbers have significant uncertainties because of the variation in the number with birthplace 'not elsewhere included'. Therefore, in this section, we take an approximation of the net inter-census change in the working-age population of overseas born. This is compared with a measure of the gross inflow of working-age migrants who have been in New Zealand for less than five years as at the subsequent census.
Outmigration by migrant birthplace
In our database of census information, we have the numbers of people in the working-age population from each birthplace region recorded as having resided in New Zealand for less than five years. This implies that they arrived after the preceding census, and these are called recent migrants. By comparing this number of migrants who had arrived over the previous five years with the net increase over the inter-census period, we can obtain an approximate estimate of the number of existing overseas born in New Zealand at the first census who had left by the time of the second census of that inter-census period. This number is called the approximate outmigration of this group of migrants.
Table 3.5 shows that, between 1996 and 2001, there was a net increase in overseas born of 81,700. There were a number of migrants entering New Zealand between 1996 and 2001, and as at Census 2001, there were 139,000 of these recent migrants remaining resident in New Zealand. Since there was a gross inter-census increase by 139,000 and a net inter-census increase of only 81,700, the difference (57,300) could be primarily accounted for by 57,300 overseas born who were migrants resident in New Zealand in 1996 but had left by 2001. Additionally, there could be some migrants who arrived and left all within the inter-census period.
Between 2001 and 2006, the comparable figures are that there was a net increase of 162,300 working-age overseas born, and as at 2006, there were 213,100 recent migrants of working age resident in New Zealand. This implies that 50,900 overseas born working-age migrants who were in New Zealand in 2001 and had left by 2006.
Looking at the birthplace regions, these numbers show that, in both of the inter-census periods, the largest number of existing overseas born working-age migrants who left came from the United Kingdom and Ireland, 24,800 of whom left from 1996-2001 and 19,300 of whom left in the period 2001-2006. The second largest numbers of outmigrants were born in Asia, with 14,700 and 16,400 in the respective periods.
Outmigration per 100 arrivals
The relativity between the number of existing migrants who become outmigrants compared with the recent migrants who arrive could be called the rate of outmigration. For all birthplaces, for the 139,000 recent arrivals remaining in 2001, there were departures of 57,300 existing residents. This means that, for every 100 of the recent overseas born migrants remaining in New Zealand, there were 41 overseas born in New Zealand in 1996 who had become outmigrants by 2001. This is how we have estimated outmigration per 100 arrivals.
The overall outmigration rate of overseas born of working age dropped, from 41 over the 1996-2001 period to 24 from 2001-2006. This, perhaps, indicates that New Zealand increased its ability to retain overseas born working-age migrants even after a period of relatively high inflow during 1996-2001. In this context, the prolonged period of economic growth in New Zealand may have assisted in this outcome, as well as newer policies focusing on matching skills to employment needs.
Another interesting aspect of this outmigration estimate is that the regions of origin with the highest rates in 1996-2001 were United Kingdom and Ireland (134), Europe and North America (60) and Australia (56). These factors had all decreased significantly by the 2001-2006 period. The outmigration of Asian migrants was low, at 24, for 1996-2001 and declined further to 17 for 2001-2006. The migrants born in other countries had an even lower rate of outmigration, at 6 and 4 respectively, perhaps reflecting the ability of South Americans, African born and others to settle in New Zealand and stay. This may reflect the conditions in countries of origin and the increased migration options of English-speaking migrants.
Age profile of the working-age population
Labour market behaviour of people born in different jurisdictions will differ according to whether they have lived in New Zealand for a period, or have just arrived from their birthplace. This pattern of change is measured by the three main classes of migrants described in later sections, namely recent migrants (0 to less than 5 years resident), intermediate migrants (5-15 years resident) and earlier migrants (over 15 years resident).
Inter-census change in migrants by age group
It is also useful to find whether there are numbers of people in the various age groups who become resident or leave residence in New Zealand between censuses. We are able to find these net flows by comparing the number of people in one five-year age group at one census with the number of people in the next-following five-year age group at the time of the next census. The positive change is the net arrivals in that class, and the negative change is the net departures in that class. (Note that these are the net flows and ignore any outmigration and deaths.)
It will be recalled from section 3.1 that there was an increase in the working-age population by 103,000 between 1996 and 2001 and by 271,000 between 2001 and 2006. The component profile of those changes is repeated, and the breakdown of these changes by main age group is shown in the table below.
The inter-census net change in the 15-19 year old age group includes all residents who have turned 15 in that period, plus the net number of 15-19 year olds at the time of the second census who took up residence during the inter-census period. By the same token, the inter-census net age group decrease for the 70 years plus age group includes those who have died as well as the net number who have ceased residence by going overseas during the inter-census period.
The figures of most interest are, therefore, the 20-24 and 25-69 year old age groups, as these represent strictly net migrant flows, with (presumed) relatively small numbers of deaths. Table 3.6 shows that, in 1996-2001, the total working-age population increased by 103,000, and this increase was due solely to the number of people turning 15 years old and/or 15-19 year old inward migrants (265,000) being greater than the total of the net migrant outflow and deaths.
The overall pattern of flow between 2001 and 2006 was remarkably similar in some respects:
- The net increase in 15-19 year olds was 300,000, similar to the 265,000 in 1996-2001, and the birthplace profile was reasonably similar.
- The net decrease in the 70 years plus age group was 103,000, similar to the 100,000 in 1996-2001, and the birthplace profile was similar.
- The net decrease in New Zealand born 20-24 year olds was 25,000, similar to the 29,000 in 1996-2001.
The main difference was that, whereas there was a net decrease of 39,000 people aged 25-69 years in 1996-2001, there was a net increase of 68,000 in this age group in 2001-2006. The cause of this turnaround was both a reduction in the net outflow of New Zealand born, from 74,000 in 1996-2001 to 48,000 in 2001-2006, and an increase in the net inflow of overseas born in this age group from 48,000 in 1996-2001 to 103,000 in 2001-2006.
There was a lesser turnaround also in the net flows of those aged 20-24 years, from an outflow of 23,000 in 1996-2001 to an inflow of 6,000 in 2001-2006. This turnaround was caused mainly by known overseas born net inflow increasing from 8,000-25,000.
The very relevant finding from these analyses is that a significant component of the inter-census increase in working-age population is a net inflow of overseas born people aged 25-69 years. This net increase in 2001-2006 was 103,000.
Delving within that group, our data indicate that 64,200 were in the high-participation age groups of 30-49 years. This is in contrast to a net outflow of just 1,200 New Zealand born in this 30-49 year old age group.
Migration has clearly resulted in a significant increase in the population of people aged 30-49 years old, which should have significant labour market consequences.
Age profiles by source birthplaces
The similarity between the behaviour in 1996-2001 and 2001-2006 in the overall flows follows through to the pattern of behaviour of people from the respective overseas birthplaces. The 2001-2006 figures showed:
- the net increase in 15-19 year olds was 56,000, similar to the 47,000 in 1996-2001, and the birthplace profile was reasonably similar
- the net decrease in those 70 years plus was 22,000, the same as the 22,000 in 1996-2001, and the birthplace profile was reasonably similar.
As with the overall pattern, the big change was in the age group 25-69 years, where the net increase of 48,000 in 1996-2001 lifted to an increase of 103,000 in 2001-2006. This lift of 55,000 came from the United Kingdom and Ireland (+20,000), Asia (+18,000), Europe and North America (+6,000), Other (+6,000) and even Australia born (+3,000).
This analysis shows that the contribution of overseas born to the 25-69 age group comes from a broad range of birthplaces - some English-speaking, some not.
Again, we have delved into the 30-49 year age group for the birthplace regions. The composition of the 64,200 net inflow from 2001-2006 of 30-49 year olds reported above was:
- 24,200 from Asia
- 16,800 from United Kingdom and Ireland
- 10,300 from Other
- 6,200 from Europe and North America
- 4,400 from Pacific Islands
- 1,700 from Australia.
The conclusion is that a very significant core of inward migrants contributing to increasing the working-age population is migrants in the high-participation 30-49 year age group, coming from a broad range of birthplaces.
Working-age migrants - conclusion
In 1981, 18% of New Zealand's working-age population were overseas born. This proportion has increased to 25% in 2006. The working-age population increased strongly between 2001 and 2006, with contributions from both the New Zealand born and the overseas born. In the last five years, at least 60% of the increase in the working-age population were overseas born. This 60% numbered 162,000, of which 79,000 were born in Asia.
Outmigration migrants (migrants who subsequently left New Zealand) numbered at least 50,000 between 2001 and 2006. While relatively high, this implies a rate of outmigration of only 24 per 100 arrivals over the 2001-2006 period. This is less than the 42 per 100 arrivals outmigration experienced over the 1996-2001 period.
The labour market benefits from migration are shown to increase for migrants who remain in New Zealand for longer than five years. The decline in the rate of outmigration shows that New Zealand is increasing its ability to attract people to settle and stay here. It is interesting that the highest outmigration is for countries with mainly European ethnicity. This group presumably have few language and other constraints on their global mobility.
Numbers in the main economically-active age group of 25-69 years old decreased by 39,000 over the 1996-2001 period. However, it increased by 68,000 people over the 2001-2006 period. These changes resulted, in part, from a reduction in the net outflow of New Zealand born - from 74,000 over 1996-2001, to 48,000 over 2001-2006. Conversely, the net inflow of overseas born increased from 48,000 over 1996-2001, to 103,000 over 2001-2006.
Focusing closer on the high-participation age group of 30-49 year olds, over 2001-2006, there was a net inflow of 64,200 overseas born and a net outflow of just 1,200 New Zealand born. Further, the 64,200 net inflow of migrants was balanced across the whole range of birthplace regions.
 Of these people, a number will be unavailable to participate in employment because they are not in the labour force for a number of reasons and, consequently, are not available for paid employment for even one hour per week. The main categories of these people are those full-time studying, fully retired, at home (either looking after children or otherwise) and a number for unspecified reasons.
 The percentages of the total working-age population reporting an unknown birthplace in each of the censuses were 0.4%, 4.6%, 4.1% and 4.6% in 1981, 1996, 2001 and 2006 respectively.
 It may be thought that the number of people of unknown birthplace are most likely to be overseas born and perhaps with English language difficulties. We have not studied the characteristics of the unknown categories throughout the censuses, but note that a change in ethnicity questions and the Aotearoa/New Zealand dichotomy could also have caused some differences among the New Zealand born as between censuses.
 The people with their stated birthplace classified as 'Other' are those from South America, Africa and the Middle East. Those who are known to be overseas born, but for whom birthplace was illegible, are classified as 'Unknown but Overseas Born'.