"Fell into crime" because of childhood poverty?

Labour's candidate for Whangarei. Lawyer, Kelly Ellis, says:

"If one addresses poverty then everything else falls into place," she said, referring to the "so many files" on her legal shelves of people who fell into crime because of issues related to childhood poverty.

 Serious criminals end up in prison in which case prison statistics should match child poverty stats.

From the Household Income Survey table H5 I can construct a comparable pie chart (bar Asian) for ethnic percentages of poor children 2010-2012:

Similar.

But a pie chart which shows the ethnicity of parents/caregivers on a benefit (2009) is a closer fit:

And if I'd depicted children rather than parents the fit would get even closer due to Maori/Pacific families being larger (but the Child Health Monitor source is currently under revision.)

On the other hand, the prison population is around 8-9,000 whereas the number of children living 'in poverty' is between 205,000 and 285,000. More generally there were 194,000 apprehensions for crime in 2012 of which NZ European made up 43 percent, Maori 42 percent, and Pacific 10 percent.

So I suppose I come back to the same sticking point. Maori are over-represented in child poverty statistics, the benefit system and in crime. Pacific people are over-represented in child poverty statistics but not particularly in the benefit system or in crime.

(All not withstanding there are ethnic categorising issues.)

Update

Berend suggested a bar chart as a better method of making my point. He's right (though I still have a slight problem with depicting caregiver instead of children.) The first two colums relate to children, the last to caregivers.

And here's another way of looking at it. As before the first two columns relate to children, the second to caregivers and the final to the prison population: