"Child poverty is worse than we thought, Prime Minister."
That was the news from the statisticians which greeted Jacinda Ardern on her return from Beijing this week.
By taking ministerial responsibility for the reduction of child poverty, she has signalled its central importance to the Coalition Government.
And if all its talk about inclusive growth and wellbeing budgets is to amount to anything more than pious waffle, it has to be.
The numbers released by Statistics NZ on Tuesday make grim reading. They are for the 2017/18 year and form the baseline to judge progress towards the three-year and 10-year targets the Government announced early last year.
Nearly one in six of the country's 1.1 million children live in households below a poverty line which is relative to incomes across the population and which takes no account of housing costs.
When housing cost are factored in, nearly one in four kids is below the line.
And one in seven is counted as experiencing material hardship, nearly half of them severe material hardship.
This is an area where metrics matter because — as businesspeople know — what gets measured, gets managed.
But while it is relatively straightforward to count how many dwellings get built or how many seedlings get inserted into hillsides, measuring poverty rates is much more difficult.
The Child Poverty Reduction Act, passed with cross-party support last December, specifies three primary measures, which governments have to set targets for, and another six supplementary measures, some more and some less stringent than the primary ones.
The first primary measure sets the poverty or "low income" line as a household income, after tax and transfers, of 50 per cent of the national median.
Incomes are "equivalised", or adjusted for differences in household composition.
So, for example, a solo parent with one child is deemed to need 1.4 times the income of a single person on her own.
Statistics NZ's best estimate of the number of children in households below that line in the 2017/18 year is 183,000, or 16.5 per cent of all children, give or take 1.1 percentage points.
When Ardern early last year announced the target reduction from "roughly 15 per cent to 5 per cent" over the next 10 years, they thought this would be a drop from 160,000 to 60,000 children.
On the latest numbers, which draw on data from the household economic survey, the (larger) household labour force survey and administrative data for Inland Revenue and the Ministry of Social Development, the starting point is significantly higher at 183,000.
The three-year target is to reduce the rate by 6 percentage points. That would still leave more than 100,000 kids below the poverty line.
This is a measure of relative poverty. What counts as low income will rise as the median level when all households are ranked by income rises.
The rate has wobbled around a range of 14 to 16.5 per cent over the past 10 years, but because of the margin of error around the numbers, it is best described as fairly flat, at least over the past five years, said Statistics NZ's Diane Ramsay.
Apologists for the National Government seized on the fact that the latest year was a statistically significant 2.3 per cent higher than the year before.
But they overlook the fact that the June year concerned straddled a change of Government and entirely preceded the Coalition Government's families package which came into effect on July 1 last year.
So it reflects National's legacy, not the current Government's first year, the effects of which will not be known until the results of the current, and much larger, household economic survey are produced next year.
This measure of the poverty line suffers from being before housing costs which, especially in Auckland, gobble up a large amount of households' incomes.
Research by Auckland University economist Susan St John and researcher Yun So found that solo parents spend between 33 per cent and 44 per cent of their disposable incomes on housing costs.
And these are nationwide numbers. Statistics NZ expects that the enlarged household economic survey now in the field and its successors will allow it to provide regional breakdowns of incomes after housing costs.
So the second primary measure adjusts for housing costs. By that measure, the 2017/18 baseline has 254,000 children or 22.8 per cent of the total living in households whose income after housing costs is below 50 per cent of the median.
The targets the Government announced last year are to reduce that from "roughly 20 per cent to 10 per cent" over the next 10 years, and by 4 percentage points within three years.
Unlike the before-housing costs measure, this metric is relative to a fixed historic level (last year's baseline) which will be adjusted for inflation.
So it will record whether a household's income has improved, in real terms, irrespective of what has happened to the incomes of other households.
Of the 254,000 kids in this group, 174,000 are in households with incomes less than 40 per cent of the national median. The level of these "lowest income" kids has been there or thereabouts for the past 10 years.
Susan St John and Yun So argue that the measures in the families package which came into effect last July are seriously inadequate for the task of addressing their needs.
So it will be telling to see what next month's "Wellbeing Budget" does about this.
Finally, there are the measures of material hardship.
These are based on people's responses in a face-to-face survey about their ability to afford things: A meal with meat, fish or chicken (or a vegetarian equivalent) at least every second day? Or two pairs of shoes? Or suitable clothes for a special occasion? Or home contents insurance? Or to give birthday or Christmas presents?
They are also asked if in the past 12 months they found they could not pay electricity, gas, rates or water bills on time. Or put off visiting the doctor or dentist? Or had to put up with being cold? Or borrowed from friends or family to meet everyday living costs? Or, if they had an unexpected and unavoidable expense of $500, could they pay it within a month without borrowing?
If they tick at least six of 17 such boxes they count as in material hardship; nine and it counts as severe.
Last year 148,000 New Zealand children were in material hardship by that measure, or just under one in seven, of whom 65,000 were in severe material hardship.