Lindsay Mitchell: Stuff says, "Little evidence of child poverty coming down"

Stuff is writing a series called 'The Whole Truth' running up to the election. It aims to fact-check campaign claims, party policies and achievements. 

Today they feature a piece on child poverty which also appears in the Dominion Post. I've included the entire short piece (because it is the simplest summation of the measures I've read) and added  a few of my own comments:

"There is little evidence, on the Government’s own measures, of child poverty coming down.

As one of her Government’s earliest acts, Prime Minister Jacinda Ardern brought in the Child Poverty Act in 2018. It established ways of measuring poverty and laid out a series of targets against them.

There are three primary measures of poverty in the Act, two of which are income-based.

For these income measures, a mathematical model is used to “equivalise”, or flatten, the incomes of different households.

This allows a consistent measure to be applied across households of all different sizes. Larger households need more income to manage than smaller households do.As a result, there is no specific income figure under which all households are said to be living in poverty.

However, the government’s first measure for child poverty does say that a child is living in poverty if they’re in a home with an income less than half the average, equivalised disposable household income, before housing costs are deducted. 

(The average disposable household income, before housing costs, is currently $46,700 a year, according to Stats NZ.)

What do the numbers look like using this measure?

Ardern set a 2020/21 target of lifting 70,000 children above this line. That would mean shrinking the 2018 measure of 16.5 per cent of children to 10 per cent.

As of June 2019, the latest available figures, this measure was down by only 1.6 percentage points to 14.9 per cent. Instead of 183,500 children in poverty, there were 168,500.

There is another, similar sort of measure, which has recorded similar results to date.

Again, a child is considered in poverty if they’re in a home with an income less than half the average, equivalised disposable household income, but this time after housing costs are deducted.

(The average disposable housing income after housing costs is $35,800 a year.)

The target for this measure is a 4 per cent shift, or lifting 40,000 children from poverty. 

The 253,800 impoverished children first counted by this measure in 2018 reduced 2 percentage points in the year to June 2019, to 235,400 children.

A third measure goes more directly at material hardship. It defines a child as impoverished if they live in a home that lacks six or more key indicators, such as a home that lacks shoes, fresh fruit or vegetables, the ability to see a doctor, or the ability to pay power bills.  

This measure is favoured by Children’s Commissioner Judge Andrew Becroft, who says it’s a better measure of a child’s life at home. 

On this measure, the number of children in poverty has increased by 4100 in the year to June 2019, up 0.4 percentage points to 13.4 per cent. 

The shifts on all three measures are so small that they fall within the margins of error for each. 

And due to the lag in the data, it’s hard to know how recent events have affected New Zealand’s poorest children.

Despite this, Ardern has frequently cited a Treasury estimate that the number of children in poverty would reduce between 10 per cent and 12 per cent due to policies like the 2018 Families Package."

She also claims that poverty has fallen on 7 of the 9 measures. However only one of the small reductions is outside of the margin of error (-2% vs + or -1.9%). The margin of error exists because the results come from a survey of just 20,000 households. 

StatsNZ itself says, "...most of the changes likely reflect the expected uncertainties present in all sample surveys."

The Minister for Child Poverty Reduction also ignores that more children are now on benefits. That was the case BEFORE Covid. At December 2019 there were 12,000 more children on benefits than at December 2017. The number of children on welfare should actually represent one of the measures.

The writer, Thomas Manch, has unusually mentioned the equivalisation method. This means a large family with a relatively high income can appear in the poverty data. A household income gets equivalised down progressively based on household members.

Finally, the data is only till June 2019 - almost 15 months old. If the June 2020 data is released pre-election the PM will undoubtedly blame her lack of progress on Covid.

And that will represent yet another obfuscation.

Lindsay Mitchell is a welfare commentator who blogs HERE.