The 'New Zealand Listener' conversation

As authors of Child Poverty in New Zealand, we are obviously happy that our book is being discussed across the political spectrum. However, there is much to disagree with in Lindsay Mitchell’s recent consideration of the issue.

Mitchell’s basic argument is simple: child poverty in New Zealand today is a problem created by past government intervention. Specifically, she argues child poverty is the cumulative malign effect of the introduction of the Domestic Purposes Benefit (DPB) by the 1972-75 Labour government. She also claims the DPB encouraged an increasing number of young single women to have children and bring them up reliant on the state. These children then grow up poor.

To sustain her mono-causal hypothesis, Mitchell asserts that child poverty was not an issue in the pre-DPB days. But this strong claim lacks reliable empirical support, as we have no comparable data on child poverty rates in New Zealand until after 1982.

The only information we have on child poverty rates around the time of the Royal Commission on Social Security of 1972 (not the Royal Commission on Social Policy as Mitchell writes – that Royal Commission was in 1987) was provided in research done by long-term NZ Listener columnist Brian Easton. Easton concluded that in 1973/4 – before the DPB could take any effect on the two-parent family form – 29 per cent of New Zealand children lived in families below the Royal Commission’s poverty line. As indicated, Easton’s rate is not comparable with the 1982-2012 child poverty data from the Ministry of Social Development on which we draw heavily in our book. What his analysis does suggest, however, is that Mitchell’s belief that child poverty was not a serious issue in New Zealand, prior to the claimed negative effects of the DPB, is incorrect.

If child poverty was a significant problem in New Zealand before the days of the DPB, as the evidence suggests, then in part the DPB needs to be seen as a response to an existing and possibly worsening problem, rather than the unilateral cause. Seeing the DPB as a response to poverty does not, of course, preclude it having a secondary downstream cumulative negative effect on the two-parent family form. But we need to dig a little deeper before allowing that story to stand.

New Zealand has shared the phenomenon of rising numbers of sole-parent families over the last two generations with virtually all other rich countries. Although direct comparability through time and across countries is not easy because of measurement issues, the growth of sole parenthood was probably the largest, and certainly reached the highest levels, in rich Anglophone countries. For example, the OECD Family Database reports comparable data on the proportion of children aged 0-14 years old growing up in sole-parent families by country. By their most recent data, New Zealand’s rate is 21.4 per cent. The rate in Australia is somewhat lower at 16.8 per cent. The rates in the United Kingdom at 22.9 per cent and in Canada at 22.1 per cent are virtually identical to New Zealand. None of these other countries has a DPB, but they do provide reasonable support for children in sole-parent families. The remaining rich Anglophone country is the United States. Their rate of sole parent children is 25.8 per cent – higher than in New Zealand and other Anglophone countries – and yet the United States is well known internationally for providing only very modest support for sole-parent families without work. These international comparisons again raise serious questions about mono-causal stories attributing high sole parenthood in New Zealand to a generous sole-parent benefit system. The comparisons also suggests there are likely to have been common social factors across rich countries – not a policy shift – behind the changes that affected all developed countries.

Data on the proportion of ex-nuptial New Zealand births, shown below, indicate a rising trend both before and after the DPB watershed, consistent with a process of gradual social change leading more women and men to choose to have a child outside marriage. The biggest acceleration in the share occurs over the 1986 to 1996 period. The data offer little evidence for the claim that the introduction of the DPB in 1973 caused any accelerated flight from marriage, leading to large numbers of women having babies by themselves, knowing that they could rely on government for an income.

Ex-nuptial births in New Zealand, 1964–2008

This item is licensed under a Creative Commons Attribution-NonCommercial 3.0 New Zealand Licence from

In fact, there is another important factor, in addition to changing social values and attitudes to partnering, behind the rise in sole parenting in countries like New Zealand. It is a factor, oddly, that both sides of the political spectrum are reluctant to acknowledge – namely economic trends for low skilled males. This is the other side of the DPB equation and, with the exception of the work of social researcher Paul Callister, it has not been given sufficient attention. That trend has been, in the last two generations, the progressive market elimination of relatively well-paid blue collar male jobs with a low educational entry requirement. The main reason for this trend has been technological shifts in jobs away from manufacturing where such jobs were relatively plentiful. The men who would have had these jobs, but now don’t, no longer have the financial security that can help establish and maintain long-term sustainable partnerships. In other words, changes in the balance of market forces have greatly contributed to causing serious social problems for raising children in New Zealand. Child poverty is not valued when employers and employees make wage bargains. Therefore it is only by very happy accident that pattern of wages and jobs arising out of these bargains result in low rates of child poverty.

Several years ago when the OECD considered solutions to child poverty solutions across its member states, including New Zealand, it concluded that a strategy to address both income transfers for children and the employment of parents with poor children is necessary. Both blades of the scissors are needed to obtain leverage on a complex problem. Mitchell steps well outside this international consensus in her consideration of solutions, and advocates scissoring with one blade. She rejects any rise in income transfers to children as part of a solution to child poverty, on the basis that such use will discourage parents from taking work.

Even taken at face value her point on benefits discouraging work is only logically correct for certain forms of income transfers for supporting children, not all forms. Clearly work is discouraged if core benefits for those people with children and without work rise, all other things being equal. However, if income transfers for children rise, and these are paid to children in families both in and out of work, there is no change at the margin on incentives to work, and thus no relative discouragement of work. Furthermore, in-work income transfers can be made to poor working parents – this both reduces child poverty of the working poor and increases incentives to move into work, especially of sole parents, improving both equity and efficiency in one fell swoop.

Additionally, Mitchell doesn’t acknowledge that it is possible to reduce child poverty by raising welfare benefits while simultaneously adapting other policy instruments, as part of a package, to offset any resulting negative incentives on work. Stronger work testing, better provision of child care, and better coordination of active employment policies are all examples of such offsetting policies which could be used as a part of such a package.

Mitchell seems particularly concerned with young teenage mothers. There is good evidence that New Zealand has a relatively high rate of teen fertility. But we note that, according to the OECD Family database, the most recent comparable data for 2010 suggests that at 22 per thousand, our rate of teen births is actually lower than both the United Kingdom at 24 per thousand and the US at 35 per thousand. We also note that there are countries, like Italy, with exceptionally low teen births (5 per thousand) and very high rates of child poverty (18 per cent), so the two phenomena are scarcely synonymous. Teen parenting is a relatively minor component of the overall problem of child poverty – most poor children are not born to single teen mothers, and if these people delayed their fertility a few years it is highly unlikely that their economic circumstances would improve significantly. We accept, however, that more needs to be done to give young people worthwhile life-paths, focussing on both potential teen mothers and the oft-invisible fathers, and better supporting those teenagers who end up as young parents.

Lastly, Mitchell seems to pigeonhole our book as advocating a simple tax, spend and forget policy. Our position is more complex. We emphasize the importance of evaluation: high quality, politically independent consideration of whether policy changes actually work. We also recognise that child poverty can and should be addressed in part by reallocating resources, and partly by using existing resources more effectively. Furthermore, additional resources to address child poverty can also be acquired by increasing taxation. There are current opportunities to raise taxation by broadening the tax base at low or no cost to efficiency, capital gains and estate taxes being two good examples. Additionally, the current system of collecting child support from non-custodial parents to offset benefit payments is a glaring example of a highly inefficient and inequitable tax and we are surprised that Mitchell does not support its reform on these impeccably liberal grounds.

As our book argues, there are significant future economic and social gains from reducing child poverty today. Under such circumstances, spending more on children to capture these gains makes economic and social sense. But the balance of funding between higher spending and reallocating spending and spending more efficiently, obviously, involves political as well as objective technical considerations, and judgments based on the weight of evidence, where subjectivity can always play a role. Nevertheless, if we are committed to reducing child poverty then relying solely on more efficient spending of existing government resources is unlikely to be enough. Ideology should not blind us to the real possibility that more spending overall on children may be needed.

Jonathan Boston and Simon Chapple