Published on the 20/02/2014 | Written by Newsdesk

The Government’s calculation of the baseline valuation of the New Zealand welfare system was a world first and now it is taking analytics even further in to the realms of predictive risk modelling…

Speaking at the SAS business analytics software users conference this week in Wellington, Cabinet Minister for Social Development Paula Bennett, revealed the importance of data analytics in achieving the Government’s welfare reforms. Without the analytics to back up the facts, Bennett compared spending money on welfare to putting our finger in the air and throwing a whole lot of money at it and hoping.

As a result of the welfare reforms and use of data analytics, Bennett revealed that New Zealand now has the lowest number of youth not in employment, education or training, since 2008 and in the last 12 months alone, there has been a 9.4 percent drop in sole parents on the benefit – the largest drop in 20 years apart from in the mid-2000s when working for families came in.

“I’ve got Treasury wanting to give us money – that’s unheard of – because they can see the results of it. We spent $330 million less on welfare in the last 12 months.

That’s the kind of difference this stuff can make.”

The reason we needed those reforms so much, Bennett told a packed auditorium of 250-plus delegates, is that 13 percent of the working age population was reliant on some sort of welfare in December 2010. 161,000 people had received a benefit for at least half of the previous ten years and 139,000 had spent more than a decade on benefit since 1993.

In a world first in 2011, Bennett and her team of analysts was able to calculate a baseline valuation of the welfare system by analysing, from a macro right down to an individual level, the information they already had. The result was a lifetime welfare liability of $78 billion.

They are now using analytics techniques such as segmenting and predictive risk modelling to identify and target those in most need. The results of the analysis have led to a radical reshape of Ministry assumptions. For example, 95 percent of Minister Bennett’s budget was being spent on the unemployment benefit and support programmes for them, but analysis revealed that they only constituted five percent of the long-term liability. Conversely, there was a projected long-term liability of $1 billion for just 4000 16 and 17 year-olds. In the case of abuse and neglect, the Ministry has used predictive risk modelling to identify the children most at risk, revealing that 70 percent of the babies it expects to be working with are not even born yet.

“In a world where I have a limited budget, what analytics have absolutely given us is the means to invest more where it makes the biggest difference,” Minister Bennett said. “It’s pretty exciting when you can target and know that the money you are investing has the greatest likelihood of a positive result.

“I could not present a case to Treasury and to Government that we should be spending more earlier [for better outcomes], without the use of data analytics and information sharing. And I can tell you the change it’s made is significant.

“I want you to know that government is catching up with business. We see the value in data analytics and the informations that we can get and the way it can be processed.”