AKA: inequality, equality
What this graph shows
- Material hardship % of New Zealand's total population
- Severe material hardship % of New Zealand's total population
- 2013 onwards, as that is when this time series bega7n
Key notes
- Issues polling, e.g., from Curia or Ipsos, does not distinguish between child or adult poverty so we don't know the level of concern respondents have towards each group. As such, we have shown a poverty metric for all Kiwis, not just children. We also believe this is the most responsible way to measure poverty which isn't always just a child issue, of course, despite the way it is often portrayed by politicians and media.
- There is no single measure of poverty, however, the DEP-17 index and corresponding 'material hardship' and 'severe material hardship' classifications are good indicators that cover a wide range of poverty outcomes.
- We couldn't use 7 of the 9 measures of child poverty that were created by the previous Labour-led Government, or their all-ages equivalent, due to them being measures of income equity, not poverty. More detail below.
- There is a significant 7-month delay in the release of poverty data, e.g. 2024/25 data was published on 26 February 2026. 2025/26 data isn't out until February 2027.
- External factors can also influence these results, and context is important, but governments are ultimately accountable for New Zealand's performance.
- Remember that each Labour-led or National-led government has had coalition or support partners that share in their successes and failures. A full list of governments for this time range is at the bottom of this page.
Data sources
Data timing
- Frequency: Annual
- Update delay: 7 months
- Last update: 2025 = 26 February 2026
- Next update: 2026 = February 2027
Why this KPI?
- In 2018, the Labour-led Government created the Child Poverty Reduction Act. This included 9 child poverty metrics, and these were the most commonly referenced poverty measures by the Labour-led Government of the time, and still are by many politicians, media, and bureaucrats. However, we haven’t used these because:
- They only measure child poverty, not all poverty. Poverty also affects adults, especially the elderly or vulnerable, and an increasing % of households don’t have children.
- The 9 measures only really contain 2 poverty measures.
- The 7 metrics containing ‘X% living below Y% of the median’ for differing financial metrics are arbitrary measures of income equity, not poverty. They can't measure if all of society is struggling or thriving more, just a comparison to the median. For example:
- It’s possible, if not likely, during tougher economic times, that a growing number of individuals fall into poverty, even if the bottom incomes are growing faster than the median via minimum wage or benefits increases.
- If the cost of living decreases relative to incomes, which it did for food for 20 years when we look at the Household Incomes vs the Food Index up to March 2022, or there is state intervention like free school lunches, which is not recorded as income, then that can reduce poverty levels even if the incomes of the lowest earners aren’t growing as fast as the median.
- These same limitations apply to the all-ages data set too.
- The DEP-17 scores, shown as 'material hardship' and 'severe material hardship' are a good poverty measure for children and adults alike; based on 17 indicators of material deprivation and financial stress: protein every second day, two pairs of good shoes, special clothes, contents insurance, gifts, fresh fruit and veges, decent meat, doctor visits, dentist visits, local trips, heating, appliance repairs, utility bills, car expenses, borrowing for living costs, clothing choices, and an unexpected $500 expense. It is important to note some weaknesses, however:
- The full data showing the % of households against each DEP-17 score from 0 to 17 is not published anywhere, only the % that scored 6+ (defined as 'material hardship') or 9+ (defined as 'severe material hardship').
- Some questions have a degree of choice and relate to values rather than poverty, e.g. buying gifts for others, or wanting to pay for home contents insurance versus choosing 'self-insurance.'
- Table 12 mentions 'Material wellbeing index', but this has not been well defined.
- The amount of discretionary income in (inflation-adjusted) $ terms could be a great measure. Table 3 looks to have this information, but it hasn’t been inflation-adjusted. We have asked Stats NZ if they've done these calculations. If not, we've asked them if we can use the RBNZ inflation calculator (CPI), as a fair and accurate way to calculate this.
- Homelessness and people living in cars might be in the Census data, although the last data set available is 2023. We have asked for more information on these measures.
- Given all of these limitations above, we turned to other leading authorities:
- Oxford definition: poverty = the state of being poor
- When experts speak about the 'poverty trap' or 'poverty cycle,' they closely correlate this to inter-generational social welfare dependence on the state (hence why the % on benefits used elsewhere on KPI.nz is also a useful measure, but not as good as the DEP-17 measures.).
Discarded
The 7 poverty reduction metrics with ‘X% living below X% of the median’ for differing financial measures.
- For the reasons mentioned above, getting closer to the median value doesn’t mean you’re being lifted out of poverty, especially if everyone in society is struggling more under cost of living pressures.
Official unemployment rate.
Related facts
Current
Older
Wishlist
With support, we’ll be able to add multiple KPIs for each issue:
- True unemployment rate
- Updates on many of the KPIs listed above.
- DEP-17 scores for all ages
- % of people living in cars
- Homelessness
- Emergency food grants
- Child-specific:
- % of dependent children for working-age people on main benefits (we have this data)
- Hardship and severe hardship as measured by DEP-17 scores
- % of children living in cars.
Governments over the timeframe shown
- 2011 = National + ACT, United, Māori
2014 = National + ACT, United, Māori
- 2017 = Labour/NZ First + Green
2020 = Labour+Green
- 2023 = National/ACT/NZ First
/ = Coalition partners
+ = Other formal governing agreements, including confidence and supply, support and cooperation.
Sources:1,2 (see table)
© Copyright
Data published by Stats NZ
© Crown Copyright
Licensed for use under the creative commons attribution licence (BY) 4.0
Analysis and graphics © KPI.nz 2026. Please share widely! You're welcome to reproduce our work with attribution and a link to the original post.
Corrections & Suggestions
If you spot a typo, mistake, or improvement opportunity for this page, please let us know.
