● Research · NZ labour market

The first-job gap: a 20 to 24 year old in New Zealand is almost twice as likely to be unemployed as someone 25 to 29, and the gap is near a forty-year high.

People say a first job is harder to get than a second. We went looking for the New Zealand number. In the June 2026 quarter the unemployment rate was 25.3% at 15 to 19, 12.0% at 20 to 24, 6.5% at 25 to 29 and 4.1% at 30 to 34: a staircase in which each five-year step down to the early thirties cuts the rate by between a third and a half. The whole of the past year’s rise in unemployment, and more, landed on people under 30. Against forty years of the same series, the gap between young and prime-age workers is the second widest on record. And it did not coincide with a rise in graduate numbers: the number of domestic bachelor’s degrees completed in 2025 was within fifty of the number in 2016.

Rolebird researchPublished 11 October 2026Stats NZ, MSD, Education Counts and OECD data files, plus named surveysofficial figures read from the agencies' own files and releases

12.0% vs 6.5%
Unemployment rate at 20 to 24 against 25 to 29, June 2026 quarter (a year earlier: 9.8% vs 5.1%)
Stats NZ, HLFS Table 4
+16,200 / −2,800
Change in unemployed people under 30, and aged 30 and over, in the year to June 2026
Stats NZ, HLFS Table 4 (summed)
4.5×
Youth (15 to 24) unemployment as a multiple of prime-age (25 to 54): the second-highest of 162 quarters since 1986
OECD series via FRED, our ratio
24,815
Domestic bachelor’s degrees completed in 2025. In 2016 it was 24,865.
Education Counts, table COM.9

What we are claiming, and what we are not

Four things, each checkable against a file you can download yourself. One: in New Zealand the unemployment rate falls steeply with age until the mid-thirties, and from 20 to 24 into 25 to 29 it nearly halves. Two: in the year to June 2026 the number of unemployed people under 30 rose by about 16,200 while the number aged 30 and over fell, so the year’s entire increase sits with the young. Three: measured the way the OECD measures it, the ratio of youth to prime-age unemployment in New Zealand is close to the highest it has been in the forty years the series has run, and well above Australia’s. Four: the supply of new graduates has not grown. Bachelor’s completions have been flat for a decade.

Here is what we are not claiming, and the first one matters most. Stats NZ does not ask anyone whether a job is their first. Age is a proxy. A 23 year old may be on a third job and a 31 year old on a first; “first-job gap” is our name for the age gradient, not a measurement of first jobs. We do not know why the gap has widened, and we say so rather than guess; the candidate explanations are listed near the end, each labelled untested. We are not saying a degree is a bad decision: the staircase shows the gap closes with age, and this article measures unemployment, not what a qualification is worth over a working life. And nothing here claims that any product, ours included, moves anyone up these stairs.

One more limit. Figures by age from the Household Labour Force Survey are not seasonally adjusted, and Stats NZ says to compare them year to year, which is what this article does. Each rate carries a sampling error, which we print beside it. A single band moving a point or two may be noise. The pattern across all eleven bands is the evidence this article rests on, and we have not run a significance test on it.

The staircase

The unemployment rate by five-year age band, June 2026 quarter, with the same quarter a year earlier in brackets and the sampling error on the 2026 figure. Source: Stats NZ, Household Labour Force Survey, Table 4, not seasonally adjusted.

15–1925.3% (23.0%) ±2.8
20–2412.0% (9.8%) ±2.1
25–296.5% (5.1%) ±1.4
30–344.1% (3.6%) ±0.9
35–393.1% (3.8%) ±0.6
40–443.9% (3.9%) ±0.9
45–492.9% (3.0%) ±0.8
50–542.9% (3.4%) ±0.9
55–593.0% (2.9%) ±0.9
60–642.7% (3.2%) ±0.8
65+1.4% (1.6%) ±0.6

Read down the column and the shape is the finding. From 20 to 24 (12.0%) to 25 to 29 (6.5%) the rate nearly halves; from 25 to 29 to 30 to 34 (4.1%) it drops by more than a third again; from the mid-thirties to retirement it sits between 2.7% and 3.9% and barely moves. The 20 to 24 rate is 1.85 times the 25 to 29 rate and 2.9 times the 30 to 34 rate. For 15 to 19 year olds, a band in which fewer than half are in the labour force at all (participation 47.6%, same table), it is 25.3%.

Underutilisation, the broader measure that adds people who want more hours and people who want work but are not currently searching, tells the same story louder: 53.8% at 15 to 19, 24.4% at 20 to 24, 13.3% at 25 to 29, 10.1% at 30 to 34, against 13.4% for everyone (same table, same quarter). Stats NZ’s own release puts the 15 to 24 group at 180,200 underutilised people, up from 158,600 a year earlier.

Where the year’s rise landed

Nationally there were 13,400 more unemployed people in June 2026 than in June 2025 (166,500 against 153,100, not seasonally adjusted). The age bands below are the change in the number of unemployed people in each, in thousands, from the same table.

15–19+6.3k
20–24+5.9k
25–29+4.0k
30–34+1.6k
35–39−2.2k
40–440.0k
45–49−0.1k
50–54−1.5k
55–59+0.5k
60–64−1.1k
65+0.0k

Add the three bands under 30 and the increase is about 16,200. Add the eight bands from 30 up and the total fell by about 2,800. The year’s whole national increase, and a little more, is people under 30. Put another way: people under 30 are 23.7% of the labour force and were 56.5% of the unemployed in June 2026, up from 50.9% a year earlier.

Two cautions, both honest. The change in any one band is within or near its sampling error: 35 to 39 falling by 2,200 could be noise, and so could 30 to 34 rising by 1,600. What the table does show, consistently, is the direction: every band under 30 rose, and nearly every band above it was flat or fell. That is descriptive evidence, not a significance test. And the staircase has always sloped this way: in all 162 quarters of the forty-year series used in the next section, youth unemployment has been at least twice the prime-age rate. So the question that matters is not whether the young are worse off but by how much, compared with before. That needs a longer series than one year.

Forty years of the same measure

The OECD publishes harmonised, seasonally adjusted unemployment rates for New Zealand by age, quarterly since 1986. We downloaded the 15 to 24 and 25 to 54 series (via the St Louis Fed’s FRED mirror, which serves the OECD data as a plain CSV) and divided one by the other for every quarter: 162 quarters with both values. The seasonally adjusted youth rate for the June 2026 quarter is 17.9% against a prime-age rate of 3.9%, a ratio of 4.55. Only one quarter in the series is higher, the March 2024 quarter at 4.85, when youth unemployment jumped while the prime-age rate had barely moved.

2.3×
Early 1990s (annual means, 1991 to 1993): youth 15 to 24 unemployment as a multiple of prime-age 25 to 54
OECD via FRED, our ratio
3.9×
2007, the pre-crisis low point for unemployment overall
OECD via FRED, our ratio
3.7×
2019, the last full year before the pandemic
OECD via FRED, our ratio
4.4×
Mean of the March and June 2026 quarters, above any full-year mean in the series; the June quarter alone is 4.55
OECD via FRED, our ratio

Two things are worth separating here. The level of youth unemployment, 17.9%, is high but not unprecedented: it reached 17.7% at the end of 2009 and 18.0% at the end of 2012, during and after the global financial crisis. The ratio is what is unusual. In those earlier peaks prime-age unemployment was also high, around 4.6% to 5.4%, so the gap was roughly 3.7 to 4.0 to 1. Today prime-age unemployment is 3.9% and youth unemployment is where it was at the worst of the last downturn. The downturn, as measured by unemployment, is concentrated in the young in a way the series has rarely shown.

Australia, measured by the same OECD series, had a youth rate of 10.7% in the June 2026 quarter, and in the March quarter (the latest with both series) a ratio of about 3.2. New Zealand’s young people are unemployed at a higher rate than Australia’s and at a higher multiple of their own prime-age workers.

The caveat that belongs in the same paragraph: a rising youth-to-prime ratio is partly a composition effect, in New Zealand and everywhere else. As more 15 to 24 year olds stay in study, the ones in the labour force become a smaller group with a different mix, and their unemployment rate can rise without any individual’s chances changing. Some of the climb from 2.3 to 4.4 over thirty-five years is that. We have not separated the structural part from the cyclical part, and a reader who wants to argue the gap is mostly structural has a real argument. Whether composition alone can account for a move from 3.7 in 2019 to 4.55 in 2026 is a question we have not answered; it would need the share of 15 to 24 year olds in study over the same years, which we did not analyse.

There is no flood of new graduates

The lazy explanation is oversupply: too many degrees chasing too few graduate roles. The Ministry of Education’s Education Counts tables show no surge in completions. Domestic students completing a bachelor’s degree, all providers, by year:

201624,865
201724,035
201824,495
201923,920
202024,325
202124,605
202223,260
202324,325
202424,975
202524,815

Ten years, a range of 23,260 to 24,975, and 2025 fifty below 2016. Total domestic completions across all levels did rise in 2025, by 5.9% to 123,200 according to the Minister for Tertiary Education’s statement of 7 July 2026, but the statement attributes the growth to Level 1 and Level 4 certificates and says bachelor’s and master’s completions each fell by less than 1%. Whatever is happening to graduates, there are not suddenly more domestic bachelor’s graduates among them. Whether the supply is too large relative to demand is a different question, and a count of completions cannot answer it.

The demand side has moved. Three measurements, each from a named source and each partial:

  • Public service graduate roles: 485.5 to 192.5, down 66%. Responses to Written Parliamentary Questions from Green MP Francisco Hernandez, reported by Newstalk ZB on 7 April 2026, put graduate-specific roles across the public service at 192.5 in the year to November 2025, from 485.5 in the year to November 2023. The Public Service Commission’s response, in the same report, was that the figures “did not tell the full story, as graduates also enter the workforce through permanent jobs,” while acknowledging that “some individual graduate and intern programmes have been paused over the past two years.” Some agencies were excluded because their data could not be compared across years. We have not seen the underlying answers; we report the figures as attributed.
  • Nursing: places for about 60% of the graduates. Kaitiaki Nursing New Zealand reported on 24 July 2026 that Te Whatu Ora would offer 1,800 graduate nurse positions over the twelve months from 1 July 2026, against roughly 2,978 new graduates expected to seek work in that period. A regulated profession with one dominant employer, and still a gap of about 1,200 between graduates and the positions offered; the Chief Nurse’s response pointed to aged care and primary care.
  • Graduate ads are a sliver of the market. The Otago Daily Times, on 1 October 2026, reported SEEK’s figure that graduate-tagged advertisements are “just over 2% of overall jobs on Seek,” in a market where total online ads were up 7% over the year to June 2026 (MBIE Jobs Online, as cited in the same report) and SEEK’s August 2026 report had ads 9.7% above a year earlier. Ads overall were up on both measures. Whether the entry-level slice grew with them, neither measure shows.

Where graduate hires actually came from

The New Zealand Association of Graduate Employers surveys its member employers, their graduates and their interns each year. The 2026 Key Insights, from 42 employers, 736 graduates and 419 interns surveyed in 2025, is a self-selected sample of member employers, 92% of which run a formal graduate programme, which is exactly the segment in which a graduate role exists as a category. Within that segment:

60%
of employers saw more graduate applications in 2025; 13% saw volumes rise by more than half
NZAGE Key Insights 2026
77%
of employers filled some or all graduate roles with people who had already done work experience with them
NZAGE Key Insights 2026
63%
of surveyed graduates had worked for their employer before their graduate role began
NZAGE Key Insights 2026

Read those two figures together: in this segment, nearly two-thirds of surveyed graduates had already worked for their employer before the graduate role began, and three-quarters of employers filled at least some roles that way. The survey does not say how many roles were filled through open advertisement, so we do not. It also found 41% of graduates received two or more offers, so competition for some candidates and rising application volumes for employers sit side by side in the same market. Thirty-five per cent of surveyed graduates planned to work overseas before the end of 2027.

Students can see it. Talent Solutions’ Student Pulse survey of 1,338 tertiary students, reported by 1News on 1 September 2026, found 14% “very confident about securing a job after study,” the lowest in the survey’s eleven years. The same measure was 40% in 2017, 28% in 2023 and 19% in 2024.

Two more series, and why they disagree

Jobseeker Support. The Ministry of Social Development’s national tables put Jobseeker Support recipients aged 18 to 24 at 48,045 at the end of June 2026, 22% of everyone on the benefit. Three years earlier it was 33,720: a rise of 42%. But over the most recent year the number barely moved (48,207 in June 2025), while the HLFS count of unemployed 20 to 24 year olds rose by about 5,900. The two measures are not the same population. Work and Income’s own eligibility page says a full-time student cannot get Jobseeker Support, that a partner’s income counts, and that 18 and 19 year olds living with parents face a lower income limit; and someone can be on the benefit with a health condition without counting as unemployed in the survey. A flat benefit count beside a rising unemployment count is consistent with more young people looking for work who do not qualify for support. It is also consistent with other things. We note the divergence and do not resolve it.

NEET. The share of 15 to 24 year olds not in employment, education or training fell to 13.8% in the June 2026 quarter from 14.4% in March (seasonally adjusted, Stats NZ). The rate fell, and that points the other way from everything above, until you remember what NEET counts. A young person who gives up looking for work and enrols in a course leaves the NEET count and the unemployment count at once. Both numbers can be quoted honestly to tell opposite stories about young people; the label on the number decides the story. Among 20 to 24 year olds the NEET rate was 15.2% for men and 18.5% for women.

Student Job Search. The charity that has placed students in work since 1982 reported, for the year to June 2025, 57,697 student registrations (up 20%), 399,961 job applications (up 24%) and around 62,000 to 64,000 listed jobs (its annual report gives both figures): about six applications per listed job, with 106 jobs filled each working day. Its own words: “The demand for our services has never been higher.” Its new graduate service, SJS Recruit, took 18,000 applications in the 2024/25 year. One graduate’s account, reported by the Otago Daily Times on 1 October 2026, gives the texture the aggregates cannot: 661 applications, 53 interviews and two offers between July 2025 and July 2026, with a master’s degree. One person is not a statistic, and we do not present her as one; her account is here for the texture the aggregates cannot give.

What could explain it, none tested

A widening gap between young and older workers is consistent with several things, and this article has not tested any of them. Last in, first not hired: when organisations are told to cut, pausing a graduate intake cuts a cost without making anyone redundant, so entry roles disappear first and reappear last; the public-service figures above are consistent with this and do not prove it. Experience as the filter: if 77% of surveyed employers fill graduate roles from people who have already worked for them, the open market a graduate without that history faces is far smaller than the ad count suggests. Composition: more young people in study leaves a smaller, different group in the labour force, which lifts the youth rate mechanically. Where the public sector sits: if public-service graduate programmes made up a large share of formal graduate intakes nationally, their contraction would read as a national story; we have not measured that share. Departures: young New Zealanders leaving for Australia change who is left to be counted, in a direction we have not estimated.

What would settle it: unemployment by age and highest qualification over several years, which the quarterly release’s summary tables do not provide; the Integrated Data Infrastructure’s post-study outcomes for the 2024 and 2025 graduate cohorts, if and when they are published; and graduate-programme intake numbers across private employers, which we did not find published as a series. If we do that work, we will publish what it shows, whichever story it supports. The next HLFS release, for the September 2026 quarter, is due on 4 November 2026, and we will update this page when it is out.

A note for the next four weeks

New Zealand votes on Saturday 7 November 2026, and the figures on this page are the kind that get quoted in a campaign. We take no position on the outcome and name no party. If you hear youth unemployment cited between now and then, three things help you judge the claim.

One. The 12.0% and the 25.3% are June 2026 quarter figures, not seasonally adjusted, each carrying a sampling error of two to three points. A claim about change since some other date needs the same Stats NZ table for that date, compared June to June.

Two. A speaker who wants the number to sound worse can quote the ratio (4.55 times prime-age unemployment, near a forty-year high). A speaker who wants it to sound better can quote the level (17.9%, below the peaks of 1991 and 2012). Both are true. This article thinks the ratio is the one that matters, for the reasons given above, and you can decide for yourself.

Three. The last labour market release before the vote is the September 2026 quarter, due from Stats NZ on Wednesday 4 November 2026, three days before election day. We will update this page that day, whichever way the numbers move.

Why we are publishing this

Rolebird sells job-application and interview preparation, and the people in the 12.0% are among the people we sell to. Read the article knowing that. We have tried to earn trust the only way that works: every Stats NZ, MSD, Education Counts and OECD figure above was read from the agency’s own file or release, is dated, and is checkable without taking our word for it; figures from news reports and surveys are labelled as such. Where we did arithmetic, the working is in the evidence ledger linked below. We claim no cause, and we do not claim that any product, ours included, changes these numbers for the people in them.

If we got a figure wrong, we will correct it and say what changed; minor wording fixes get a dated note. That is the same corrections practice as the rest of our research.

Common questions

The plain versions of the questions this article is about, each answered from figures it already cites.

Is it harder to get a first job than a second in New Zealand?
The survey cannot see first jobs, but it can see age. In June 2026 the unemployment rate was 12.0% at 20 to 24, 6.5% at 25 to 29 and 4.1% at 30 to 34 (Stats NZ, Household Labour Force Survey). The 20 to 24 rate is about 1.85 times the next band and nearly three times the one after.
What is the youth unemployment rate in New Zealand?
For 15 to 24 year olds, 17.9% seasonally adjusted in the June 2026 quarter (OECD series) or 17.2% unadjusted, derived from Stats NZ’s Table 4. Within that, 25.3% for 15 to 19 and 12.0% for 20 to 24. The national rate was 5.6%.
Is the gap between young and older workers unusual?
Yes. Youth unemployment is 4.55 times prime-age (25 to 54) unemployment, the second-highest ratio in 162 quarters since 1986; it was about 2.3 in the early 1990s and 3.7 in 2019. Australia’s equivalent ratio is about 3.2. Part of the long-run rise is because more young people are in study; we have not measured how much of the change since 2019 that explains.
Are there more graduates than before?
No. Domestic students completed 24,815 bachelor’s degrees in 2025 against 24,865 in 2016 (Education Counts). Total completions rose 5.9% in 2025, but the Minister’s statement attributes that to lower-level certificates.
Are there fewer graduate jobs?
In the public service, graduate-specific roles fell from 485.5 to 192.5 between the years to November 2023 and November 2025, according to answers to Written Parliamentary Questions reported by Newstalk ZB; the Public Service Commission says graduates also enter through permanent roles. Te Whatu Ora offered 1,800 graduate nurse positions for about 2,978 graduates. Graduate-tagged ads are just over 2% of SEEK listings. No all-employer series exists.
How many young people are on Jobseeker Support?
48,045 aged 18 to 24 at the end of June 2026, 22% of all recipients and 42% more than three years earlier, but almost unchanged over the latest year (Ministry of Social Development). The benefit count and the unemployment count measure different groups.
So why is it happening?
This article names five candidates (entry roles cut first, experience as the filter, composition, the public sector’s weight, and departures) and has not tested any of them. It claims no cause.

Related guides

How to write a cover letter with little or no work experience (NZ graduates), how long a New Zealand job search really takes and how to stand out when 100 people apply for the same job. The regional picture is in the Northland divergence.

Sources and method

All accessed 7 and 11 October 2026. Official figures were read from the agencies’ own downloadable files, not from coverage of them; where a figure comes from a news report or a survey publisher, the text says so. Figures by age from the HLFS are not seasonally adjusted and are compared June to June, as Stats NZ recommends. The forty-year ratio uses the OECD’s seasonally adjusted series and is our arithmetic on published values. The claim-by-claim ledger, including what was checked and not used, is in the repository as docs/drafts/nz-first-job-gap-evidence-ledger.md.

  1. Stats NZ, Labour market statistics: June 2026 quarter (released 5 August 2026): the release text, and Household labour force survey: June 2026 quarter (Excel), Table 4, key labour market measures by age group, including sampling errors.
  2. OECD, Infra-Annual Labor Statistics: unemployment rate, 15 to 24 years and 25 to 54 years, New Zealand and Australia, quarterly, seasonally adjusted, as served by FRED (series LRUN24TTNZQ156S, LRUN25TTNZQ156S, LRUN24TTAUQ156S, LRUN25TTAUQ156S).
  3. Ministry of Social Development, Benefit Fact Sheets: June 2026 quarter, national level data tables (Excel), “JS - last 5 years” and “Summary table-current”; the Snapshot PDF; and Work and Income’s Jobseeker Support eligibility page.
  4. Ministry of Education, Education Counts, Students gaining qualifications from tertiary education providers 2016–2025 (Excel, updated July 2026), table COM.9; Minister for Tertiary Education, media statement, 7 July 2026.
  5. New Zealand Association of Graduate Employers, Key Insights 2026 (PDF, April 2026), in partnership with Talent Solutions.
  6. Student Job Search, Annual Report 2024/25 (PDF, September 2025).
  7. Newstalk ZB, Azaria Howell, “Public sector graduate jobs down 66% as Green MP warns of exodus to Australia”, 7 April 2026.
  8. Kaitiaki Nursing New Zealand, “‘It’s really disheartening’: Te Whatu Ora graduate job offers fall short”, 24 July 2026.
  9. Otago Daily Times, “Graduate applies for more than 600 jobs in very tough market”, 1 October 2026; SEEK NZ, Employment Report, August 2026, 17 September 2026.
  10. 1News, “Student job confidence slumps to its lowest in 11-year survey”, 1 September 2026; HRD New Zealand, “Students’ job confidence plunges to lowest levels”, 7 September 2024 (Talent Solutions Student Pulse).
  11. interest.co.nz, Ella Somers, “Rising underutilisation and unemployment weigh heavily on young New Zealanders”, 5 August 2026 (for the 15 to 24 underutilisation rate as a percentage; the counts are in source 1).