Lost in the Margins: How Incomplete Data Undermines Child Poverty Solutions

21 February 2025

by Yu Shi AKA Harry, Research & Programme Officer of Child Poverty Action Group

The latest child poverty statistics have reignited frustration over the government’s failure to meet reduction targets, a shortcoming that has now occurred twice in six years. While this failure deserves condemnation, an even greater concern is the reliability of the data that informs policy decisions.

At the Wellington lockup, statisticians repeatedly mentioned "costings." The latest figures primarily reflect household conditions under the previous government, yet the current government's financial constraints imposed on Statistics New Zealand are shaping data quality. Expenditure cuts and broader defunding of the public sector undermine the accuracy and reliability of child poverty statistics, making it harder to develop and implement effective policies.

The Household Economic Survey, which serves as the foundation for child poverty statistics, has faced persistent challenges in securing adequate participation. In the first two years of reporting (2017/18–2018/19), the survey reached its intended sample size of 20,000 households. Since then, response rates have steadily declined. In 2019/20 and 2020/21, the sample size dropped to 16,000, only 80% of the target. By 2021/22, participation rates plummeted to an alarming 44.5%, with just 8,900 households surveyed. Although response rates have improved in the last two years – 70.5% in 2022/23 and 95.7% in 2023/24 – this instability has introduced significant uncertainty into year-to-year fluctuations in key child poverty measures. Without a robust and consistent dataset, tracking progress and responding to worsening poverty trends becomes increasingly difficult.

Beyond these fluctuations, there is a deeper issue: data silencing. Marginalised communities, particularly Māori and Pacific households, continue to be underrepresented in the survey, with participation rates lagging at just 67%. This underrepresentation is not a mere technical issue; it actively distorts the statistical picture, producing larger margins of error, greater uncertainty, and an unreliable measure of ethnic disparities in child poverty. The very groups most affected by economic deprivation are being rendered statistically invisible. When data collection systematically excludes these communities, their struggles are not just undercounted – they are erased.

The consequences of this erasure are profound. Without reliable ethnic and disability-specific breakdowns, policymakers can deprioritise targeted interventions, citing statistical uncertainty as justification for inaction. High margins of error become an excuse for delay, reinforcing a cycle of systemic neglect. This follows a broader pattern in which official statistics, often perceived as neutral, fail to capture the full extent of structural harm. The statistical invisibility of Māori, Pacific, and disabled households does not just obscure disparities; it enables them to persist unchecked, compounding slow violence through policy failures.

If New Zealand is serious about reducing and eliminating child poverty, it must confront weaknesses in data collection head-on. Instead of just tracking poverty rates, statistics must be a tool for policy change. This requires:

·         Investment in robust, representative, and community-informed data collection to capture the full extent of child poverty.

·         Ensuring targeted interventions for Māori, Pacific, and disabled children based on accurate, disaggregated data that accounts for their specific experiences.

·         A government commitment to using poverty data as an active tool for policy reform, ensuring action is taken based on evidence rather than statistical uncertainty.

·         Recognising data as a public good that requires sustained investment rather than being treated as a cost to be minimised.

The defunding of public services like Statistics New Zealand is not merely a budgetary decision; it is a political act that directly limits the visibility of poverty and structural inequalities. When funding for high-quality data collection is cut, the most marginalised communities, Māori, Pacific, and disabled households, are the first to disappear from statistical measurement. These cuts exacerbate statistical biases, ensuring that official data remains skewed toward dominant, more accessible populations while underrepresenting those most affected by economic deprivation. Without investment in culturally responsive and accessible survey methodologies, the reality of poverty within these communities will remain increasingly invisible.

The repeated emphasis on “costings” by statisticians reflects a shift in priorities, from ensuring data integrity to justifying financial constraints. Budget cuts shape data collection methodologies, influencing decisions about sample sizes, and community engagement. When financial feasibility, rather than social necessity, dictates how child poverty is measured, the result is weaker, less representative data. Higher margins of error and lower response rates among Māori, Pacific, and disabled households create statistical uncertainty that, in turn, makes it easier for policymakers to justify inaction. When statistics become unstable, policy responses become delayed or diluted, allowing deprivation to persist unchallenged.

These methodological compromises are not neutral; they perpetuate slow violence by ensuring that poverty remains obscured, debated, or deprioritised. If the government is serious about addressing child poverty, it must prioritise investment in robust, disaggregated, and community-informed data collection. Otherwise, the cycle of statistical invisibility, weak policy interventions, and entrenched poverty will continue. The policymaking process must move away from treating official statistics as a budgetary burden and instead recognise them as a critical public good, one that demands sustained investment and meaningful collaboration with community-led data initiatives.

There is a glimmer of hope in the fact that Statistics New Zealand recognises the low participation rates among marginalised communities and has expressed a commitment to improving them. However, addressing this issue requires more than just technical adjustments; it demands a coordinated effort to urge the government to allocate resources toward strengthening data collection in underserved communities. Beyond this, a broader re-evaluation of political participation and the government’s relationship with marginalised groups is necessary. Ensuring that all voices are counted in our statistics is not just a methodological challenge but a fundamental step toward equity and justice in policymaking.

Every child deserves a life free from poverty. Let’s make sure our data and our policies reflect that commitment.