Administrative Burden: The Burden of Being Poor

24/04/2026

How Deservingness Shapes Who We Are Willing to Help

This is the third post in a series about what it actually costs to navigate the state: whether you are a regulated party or an individual and whānau trying to access a publicly funded service. Last month, I went back to the foundational 2015 article by Moynihan, Herd, and Harvey and sat with its most unsettling argument: that the friction regulated parties and citizens experience when dealing with government is not always an accident. Sometimes, it is a design choice. I also proposed a fourth cost the framework does not name: the burden of impermanence, of a state that keeps forgetting who you are because it keeps restructuring itself.

So far, the series has moved between regulatory systems and public services. Today’s post narrows that focus. The research I draw on today is about ordinary publicly funded service delivery: health systems, social security, and the places where citizens encounter the state not as regulated parties but as people seeking help. The examples come from the United States and from Aotearoa, and they point in the same direction. In the post below, I want to do two things. First, look at where burdens actually fall, because the research is clear that they are not distributed equally. And second, sit with a recent piece of research about who the public is willing to help, and ask whether what it describes is also happening here.

A Tax That Falls Upside Down

Administrative burdens appear to be regressive. They function like a tax that falls hardest on those with the fewest resources to pay it.

Think about what it takes to apply for a benefit. You need to find out whether the benefit exists. You need to work out whether you and your whānau are eligible. You need to assemble documents, fill in forms, attend appointments, and follow up when things go wrong. For a salaried professional with savings, a computer, and flexible working arrangements, this is an annoyance. For a person working two insecure jobs, without reliable internet, with children to collect from school at three o’clock, every one of those steps is a barrier. The fifty-dollar fee for a certified document is not the same fifty dollars in both households. The half-day off work to sit in a waiting room is not the same half-day.

The inequality runs deeper than time and money. Poverty itself consumes cognitive bandwidth. The constant calculation of which bills to pay, which meals to skip, which risks to take, uses up the same mental resources that navigating a complex government system demands (Gennetian & Shafir, 2015). The people who face the most burdensome processes are also those with the least capacity to spare to deal with them. The form is the same. The person filling it in is not. And, for all the co-design many of us have been subjected to for the past six years, it is still a surprise to me how poorly understood this is.

We do not need to look to the international literature to see this. The Welfare Expert Advisory Group, in its 2019 report Whakamana Tāngata, consulted over three thousand New Zealanders and found people on benefits living what it described as “desperate lives” on “seriously inadequate incomes.” The consultation process documented, repeatedly, people describing the experience of seeking state support as degrading: a process that required you to prove your need in terms that diminished your dignity. That is not a compliance cost. It is a psychological cost, borne disproportionately by those with the least.

Who Deserves an Easier Time?

If burdens fall hardest on the least resourced, you might expect the public to favour reducing them. The research suggests it is not that simple.

A recent large-scale survey experiment with over four thousand respondents in the United States tested something deceptively simple (Haeder, 2025). Participants were presented with women seeking enrolment in Medicaid under four different life circumstances: a woman with disabilities, a single mother, an able-bodied single woman, and a woman with an opioid addiction. All were legally eligible. The question was whether respondents supported reducing the administrative burdens these women would face.

The public was most willing to reduce burdens for the woman with disabilities. Somewhat willing for the single mother. And markedly less willing for the able-bodied woman and the woman with addiction. Legal entitlement was identical. Willingness to make access easier was not.

The pattern held across political lines: conservatives and liberals made the same broad distinctions. It held across racial groups. The woman’s race or ethnicity in the vignette made almost no difference; life circumstances overwhelmed everything else. The hierarchy of deservingness, it would appear, is not a partisan artefact. It is a deeply embedded cultural judgement about what kind of suffering counts.

What This Looks Like Here

I want to be careful about importing American findings into Aotearoa. Our welfare state has a different history, and Te Tiriti introduces dimensions of obligation and partnership that the American literature does not contemplate. But there is a recent episode in our own system that is difficult to read as anything other than the deservingness dynamic at work.

In May 2020, when COVID-19 hit, the government created the Income Relief Payment, which provided $490 per week to workers newly unemployed due to the pandemic. At the same time, the existing Jobseeker Support sat at roughly $251 per week. Same unemployment. Same need to pay rent, feed children, keep the lights on. But if you had lost your job because of COVID, you received nearly twice as much as the person who had been on a benefit before the pandemic arrived.

The government did not set out to illustrate the deservingness thesis. But the effect was plain. It created a two-tier system that distinguished between the newly unemployed, who were sympathetic, and the existing beneficiaries, who were not treated with the same regard. As Louise Humpage (2020) observed, the welfare system had long operated on an “us or them” logic, and the pandemic response made it visible. The Income Relief Payment was, in effect, an acknowledgement that the existing benefit rate was inadequate, but only for people the system was prepared to see as deserving of something better.

This was not an isolated signal. The Whakamana Tāngata report had already found, a year earlier, that benefit recipients were routinely treated as undeserving of either assistance or respect, both by Work and Income and the broader public, with Māori, Pacific peoples, people with disabilities, and young people especially adversely affected. The Ministry of Social Development’s own rapid evidence reviews, prepared for the Welfare Expert Advisory Group, found little evidence that sanctions move people into work. What they documented instead was counterproductive compliance, disengagement from the benefit system, increased poverty, and worsened health. The conditionality system, in the consultation findings, diminishes trust, causes anger and resentment, and contributes to what the report described as toxic levels of stress.

And the burdens do not fall equally across communities. Successive nationwide surveys of health providers have found that Māori health providers face discriminatory differences in contract length, monitoring intensity, compliance costs, and auditing frequency compared to generic providers: shorter contracts, heavier scrutiny, and higher administrative overhead, all of which place additional strain on already under-resourced organisations. The most recent child poverty data shows material hardship at roughly 31 per cent for Pacific children and 25 per cent for Māori children, against 14 per cent overall. These are not small differences. Indeed, the need to equalise the burden was at the heart of the Pae Ora Reforms, which are currently either on hold or being undone: it is hard to tell.

The Moral Architecture of Access

A person’s entitlement to a benefit is established by law. The process by which they access that benefit is established by regulation. But the public’s willingness to see that process made less burdensome appears to be governed by neither. It is governed, the research suggests, by a moral judgement about the claimant’s character.

The social construction of target populations as deserving or undeserving becomes embedded in policy design (Schneider & Ingram, 1993). Programmes for populations perceived as deserving tend to have lower barriers, more generous provisions, and less surveillance. Programmes for populations perceived as undeserving tend to have higher burdens, more intrusive verification, and more punitive consequences for non-compliance.

So when the Ministry for Regulation, or for that matter any policy shop, talks about cutting compliance costs or administrative burden, it is worth asking: for whom? When we design a simpler process, whose simplicity are we prioritising? And when we leave a process complex and punishing, are we doing so because it is genuinely necessary for programme integrity, or because we have made a quiet judgement about the people who bear the cost?

After Whakamana Tāngata, after the two-tier COVID payments, after the evidence on sanctions, I do not think we can treat these as purely technical questions.

Next month: We turn to the gendered dimension of administrative burden. The unpaid labour of navigating the state falls disproportionately on women, and it remains almost entirely invisible to the policymakers who design the systems that demand it. The month after that, we explore the role of Weavers and Whānau Ora navigation in reducing administrative burden: one of the most exciting and innovative responses we have seen in Aotearoa for some time.

References

Gennetian, L. A., & Shafir, E. (2015). The persistence of poverty in the context of financial instability: A behavioral perspective. Journal of Policy Analysis and Management, 34(4), 904–926.

Haeder, S. F. (2025). Assessing the effect of deservingness cues on tolerance for administrative burdens. Policy Studies Journal, 54, e70043. https://doi.org/10.1111/psj.70043

Humpage, L. (2020). In our welfare system, it’s ‘us or them.’ Newsroom, 31 May 2020.

Moynihan, D., Herd, P., & Harvey, H. (2015). Administrative burden: Learning, psychological, and compliance costs in citizen-state interactions. Journal of Public Administration Research and Theory, 25(1), 43–69.

Schneider, A., & Ingram, H. (1993). Social construction of target populations: Implications for politics and policy. American Political Science Review, 87(2), 334–347.

Welfare Expert Advisory Group. (2019). Whakamana Tāngata: Restoring dignity to social security in New Zealand. New Zealand Government.