Who Left, Who Returned and Who Was Still Away? Migration patterns of 2003 graduates, 2004-2010

Who Left, Who Returned and Who Was Still Away? Mig…
01 Nov 2012

This report presents new research on the extent that domestic graduates from tertiary education institutes (TEIs) leave, and then return to, New Zealand. It examines the migration patterns through to 2010, of New Zealand students who completed a tertiary qualification in 2003. We exclude international students and those domestic students who spent an extended period overseas using a non-New Zealand passport before study.

This research improves our understanding of the extent that graduates from TEIs are attracted overseas and therefore do not contribute, at least initially, to New Zealand’s stock of human capital. Of particular interest is improving our knowledge of which type of graduate is more likely to leave New Zealand permanently.

Key Results

Table 1 summarises our key results in terms of which 2003 graduates left New Zealand, which returned and which were still away at the end of the study period. These results control for the differences between graduates that we could observe, such as age.

More than a quarter (25.9 percent) of 2003 domestic tertiary graduates left New Zealand between 2004 and 2010 for a year or more. Of those who left in 2004 or 2005, around a quarter (25.6 percent) returned to New Zealand four years later. Of all 2003 graduates, 15.1 percent were abroad in 2010 and had been abroad for at least three years.

Table 1: Summary of main results, by level of qualification

Level of study

Number of graduates

% of all graduates who left

% of leaving graduates who returned

% of all graduates that were still away

Level 1–3 certificates





Level 4 certificates





Level 5–7 diplomas





Level 7 bachelors/grad





Level 8 honours/postgrad





Level 9 masters





Level 10 doctorate




All levels 40,623 25.9 25.6 15.1

Notes: Figures have been extracted from the IDI prototype managed by Statistics NZ. They have been adjusted to take account of differences between graduates in terms of age, young completer status, sex, ethnicity, leaving student loan amount; return rates were also adjusted for whether the plane disembarked in Australia or not. (1) Of all 2003 graduates, the proportion that left New Zealand between 2004 and 2010. (2) Of all 2003 graduates that left in 2004 or 2005, the proportion that were back in New Zealand in years four and five after leaving. (3) Of all 2003 graduates, the proportion that were abroad in 2010 and had been abroad since at least 2008.

Level of qualification

The likelihood that a 2003 graduate left New Zealand over the following seven years is strongly associated with the level of their qualification. The probability of leaving increases with level, from 18.7 percent for those graduating with a level 1–3 certificate to 48.1 percent of those graduating with a doctorate.

The relationship between qualification level and the likelihood of return after four years was weaker. This may be partly due to data limitations, primarily not being able to observe graduate migration patterns beyond 2010. From what we could observe, the probability of return was above average for those with level 5–7 diplomas and below average for those with doctorates.

The proportion of 2003 graduates who were abroad in 2010, and had been for at least three years, is strongly related to level, even after controlling for other differences between graduates that we could observe. The proportion of graduates who were abroad, through 2008 to 2010, increases from 10.6 percent for those with level 1–3 certificates to over one third of those with doctorates.

Graduates in bachelor and postgraduate qualifications tended to go overseas straight after graduation or after a few years in the workforce. In comparison, graduates in lower level qualifications tended to leave at a more consistent rate over the seven year period. In doing so they are behaving more like the overall New Zealand population.

Field of study

There was not much variation by field of study (after controlling for other differences) in the proportion of 2003 graduates abroad through 2008–10 especially for those with a level 1–3 certificate or a level 4–7 certificate or diploma. There was more variation for those graduates with bachelors or postgraduate qualifications, with graduates in architecture and building being more likely to be abroad through 2008–10, and graduates in agriculture, environmental and related studies being less likely to be abroad. Those with bachelors degrees in education were also less likely to be abroad.

Looking at a more detailed breakdown of qualifications (without controlling for other differences), there were some qualifications where between a third and 40 percent of 2003 graduates were abroad through 2008–10, many in science, technology, engineering and mathematics (STEM). A number of these specialised qualifications were held by small numbers of graduates, which means that we cannot be confident that future graduates in these qualifications will be abroad at similar rates.

Other characteristics

There are other characteristics that affect the migration patterns of graduates, even after controlling for other differences. The likelihood of being abroad decreases with age, with those aged between 20 and 24 years when completing their qualification being more likely to be abroad seven years later than any other  age group. Being female significantly reduced the probability of being abroad through 2008–10, although the effect was relatively small. Asians and those in the ‘Other’ ethnic group were significantly more likely to be abroad than Europeans. The student loan balance at the time of graduation was positively associated with the likelihood of being abroad, although this effect was relatively small.

We could not identify the final destination for departures from New Zealand, only where their plane had landed. We simplified this information into an ‘Australia’ or ‘rest of the world’ variable. Although this is limiting, this variable still had a large association with the probability of returning to New Zealand. Departing graduates whose plane landed in Australia were significantly less likely to return to New Zealand after four years than those whose plane landed elsewhere. We would expect this type of effect given that it is easier for New Zealanders to stay for extended periods in Australia than other countries.

Around 60 percent of graduate departures landed in Australia, compared to 70 percent of all New Zealand departures aged 17 to 59 years. Across all age groups, with the exception of the 17–20 year olds, graduate departures were less likely to land in Australia than New Zealand departures in general. The likelihood of graduate departures leaving for Australia generally decreases with the level of their qualification, from 79 percent for level 1–3 certificates to 42 percent for doctorates.

Page last modified: 15 Mar 2018