Post sponsored by

MIL OSI – Source: Green Party – Press Release/Statement

Headline: Social Investment: Community is more than just dollars

The Government has a lot riding on its Social Investment approach. Last week they launched a new Social Investment Ministry. The Government has decided that they are going to take a data driven approach to government spending – well social spending at least. There are many things that this may entail.

We’ve already seen changes to social security legislation. This allows the Government to use a predictive risk modelling tool to label certain young people, needing income support, up to the age of 20 as at-risk-of-long-term-benefit-dependency. Once deemed a risk, the Government can then limit their financial autonomy and impose more obligations on them, that if not met could result in a benefit sanction. All of this while their friends who don’t meet the Government’s predictive risk profile have financial autonomy and far fewer obligations.

I have written previously about some of the reasons an actuarial model is problematic. More recently I’ve been mulling over some of the other things that really worry me about this actuarial approach to Government spending. My vision of Government goes beyond Government acting like an insurance company. I’m not sure where Te Tiriti obligations fit into the actuarial model. I don’t know where the concept of compassion, community or responsibility for each other fit into the actuarial model. I don’t know how we hold a model to account for neglect.

I am very worried that we will end it with a very blunt tool that ignores the morality of decisions and makes it more, if not very, difficult to question the grounds of any particular decision. Over the weekend I was listening to a Reith lecture BBC podcast with Professor Michael Sandel. He was making a case for limiting the role of market models in governance; particularly raising concerns about the use of Cost Benefit Analyses for directing political decision making.  In some ways the actuarial model is like a predict risk model to inform a Cost Benefit Analysis (CBA) for government action. He mentioned several examples that raise hard questions for what our Government is doing in being so narrowly focused on a social investment approach.

The Czech Republic decided to put a tax on tobacco as a means of reducing high levels of tobacco use. In response Phillip Morris, who were selling a lot of cigarettes in the Czech Republic, did a CBA. They found that while smoking undoubtedly had its costs the government was actually saving $147m a year from reduced social security and health costs as people died earlier. While I can’t see a recommendation as plainly absurd as that coming out of the social investment approach – we have to remember that a key identified purpose of the social investment approach is to reduce long term costs to the government.

So what is a social investment approach likely to say about investing in people who are may never be ‘fixed’? What if the modelling showed people in need of income support stopped getting it and relied on already poor family to avoid humiliation/jumping through the hoops? Would that be a good outcome for society even if it clearly reduced costs to government in the short/medium term? Is there anything in what we’ve heard so far that tells us what the Government thinks or are we all assuming either way depending on our trust in the Government?

In the Czech example, some people critiqued the modelling for not properly evaluating the value of human life. So other people did the figures again factoring in a $3.7m per life for those under 70 and $2.3m per life for those over 70. This was something that advocates for the elderly weren’t so enamoured with. Again this demonstrates seemingly rational inputs may not be consistent with the values of a community – but again because most of the actuarial process is technical we may never actually know what the underlying values are. These values are not being debated in parliament.

Another example that Professor Sandel discussed was on a cost benefit analysis on whether to make using a cellphone while driving illegal. A CBA resolved out to a $43bn cost and $43bn benefit. Looking further into the numbers though Professor Sandel found they had assumed and thus costed 2600 deaths per year. The best information they had though was that there were between 800 and 8000 deaths a year. So the market mimicking calculus was really no better than a stab in the dark. With what we know so far of the social investment approach the identified risk factors have more strength than this example but could still fairly be described as weak. Yet the Government is willing to use them to decide who will get help and who won’t.

A scientist once did a cost benefit analysis for a particle accelerator that had they estimated a 1 in 10 million chance of destroying the world. He decided it wasn’t worth going ahead because the 600 trillion dollar cost of total global destruction outweighed the benefits of the accelerator by $100m. This surely demonstrates how ludicrous it is to quantify the financial value of existence.

What this story highlights for me is the essential amorality of putting decisions about people’s health and wellbeing and the functioning of our communities into a market model. CBAs and actuarial modelling have their place but we should be very wary of introducing them into the governance of our country.

Surely good political leadership would focus our resources on creating the environment for everyone to thrive, including focusing attention on rebuilding our communities so we can support each other when bad things happen. We will all be better off when government policy and systems are informed by the wisdom of people using the systems and Government has a clearly articulated goal to engage with (not decide) identified community needs. Government should clearly articulate priorities in enough detail for everyone to understand the consequences and values statement that they represent. These priorities should then be open to democratic debate.