All policy is biased, so what can we do?
Jess Berentson-Shaw from the Morgan Foundation says all policy is infused with personal and cultural bias. What can policy-makers do to overcome it?
I had an interesting discussion the other day. It was in the context of a discussion of tax and welfare fraud, and the not insignificant body of evidence showing tax fraud is treated measurably differently from welfare fraud. This person was surprised, “fraud is fraud, it’s the law, and it all gets prosecuted surely?” They were genuinely surprised that law (and by extension policy) could be (and is) influenced by personal and structural bias. A rejection of bias in our systems is however pretty common.
Recently, when I raised the issue of the “breadwinner paradigm” in a group of senior tax and welfare people it was soundly rejected as an issue in New Zealand policy. The breadwinner paradigm assumes that (heterosexual Pākehā) nuclear families with one higher wage-earner are the norm, and policy is designed according to how it impacts upon this model. This family type might have been the norm in the 1970s, and while I know the past is, as they say, a different country, some apparently still live there. When I relayed the experience to a work colleague, she looked glum and said “I have been talking about that issue with a similar response since the 1990s.”
Why the rejection of the breadwinner paradigm or indeed that bias may be playing a role in policy? It is not for lack of evidence. Máire Dwyer demonstrated the breadwinner paradigm was at work in Working for Families some years back.
Working for Families was a package that was designed to incentivise more women into the workforce, help make work flexible for parents, and deal with the rising levels of in-work poverty. It actually resulted in 9300 fewer second earners in two-parent families being in paid employment as middle-income two parent families could now afford to work fewer hours. While there were increases in sole-parent engagement in paid employment upon its introduction (8100 more sole parents working 20 hours or more per week), following the Global Financial Crisis all the gains were lost. The policy works well for those in traditional family structures, but not sole parents. If we look at how the policy is constructed this is not surprising.
When we measure our growth and wellbeing simply using GDP we advantage a world view in which money is the most important metric of wellbeing – Māori culture does not view the world in such a way, and increasingly many other people don’t, so why do we continue with a myopic focus on GDP?
To receive the in-work tax credit a couple needs to work 30 hours between them - there is no requirement that both work: they have flexibility to share the hours however works for them. A sole parent on the other hand must work 20 hours to receive the credit. Sole parents are in the main women; their work tends to be more casual (and lower paid) in order to meet the obligations of parenting on their own. It is easy to slip beneath the minimum work requirement and lose the credit. Given the lower overall incomes women are on, the impact on their family economics is greater. Abatement rates on WFF and the 20 hours of subsidised childcare also compound the problem for sole parents especially. Professor Susan St John has done extensive analysis of how current effective marginal tax rates are especially damaging to low-income families.
Other policies where we see bias for or against certain groups are not hard to spot when you start looking. When we measure our growth and wellbeing simply using GDP we advantage a world view in which money is the most important metric of wellbeing – Māori culture does not view the world in such a way, and increasingly many other people don’t, so why do we continue with a myopic focus on GDP?
Fuel taxes, which government currently uses, and which were recently proposed by the Labour party as a mechanism to pay for public transport in Auckland are biased in favour of certain groups. For the well-off living close to their workplaces in the inner city, fuel taxes are not a big a deal. If you live on the outskirts of a city (where housing is cheaper), work more than one job (because your pay is low), and have little or no access to affordable public transport and hence have to drive frequently, then the policy has a proportionally greater impact on you.
Emma Espiner has recently argued that the policy that puts a cap on the number of years that a student loan can be accessed will leave us with a medical workforce comprised of well-off Pākehā doctors, because these are the people who can take such a policy on the chin financially. For anyone from a less well-off family (disproportionally Māori and Pacific people in New Zealand) it’s the point at which medical training is not achievable. The impact of a monocultural workforce on the health outcomes for the population is measurable and recognised.
Why is policy biased to favour the status quo?
Because people are biased – yes all of us tend to bias our thinking in favour of majority groups in society – often unconsciously.
Unconscious bias is an area that is increasingly recognised as a problem in the business world, and people in government are no less subject to the reality of being a human being.
New Zealander Dr. Carla Houkamau’s work helps us understand how and why people unconsciously privilege certain groups in society in their work through her research on stereotyping.
Stereotypes help us navigate the world, they are short cuts our brains make and have their uses, but they also work to support negative bias against certain groups often without us even realising it (that is the nature of short cuts - they become automatic).
We may have negative bias about specific ethnic groups or against people living on low incomes, sole women parents, because negative views of these groups were explicitly created in our society to serve the purposes of powerful people.
Stereotypes and negative bias are created within a culture, usually over a long period of time. We may have a bias that women are less capable in leadership roles because the history of our society has been to explicitly downplay the abilities of women (not that long ago women could not have their own bank accounts, and debate still rages about women’s ‘cognitive abilities’ compared to men – hopefully just in the mind of some Google engineers). We may have negative bias about specific ethnic groups or against people living on low incomes, sole women parents, because negative views of these groups were explicitly created in our society to serve the purposes of powerful people. Such views are deeply embedded in most cultures and even members of the stereotyped groups can be subject to such thinking (this is called horizontal bias).
These negative biases create a set of expectations about individuals, which we seek “evidence” in everyday life to confirm. How do we seek to confirm these biases? Often through our own behaviours, behaviours we are not even aware we are engaged in.
Research by Dr Houkamau and her colleagues have shown that teachers hold high expectations for all ethnic groups but not Māori (even though Māori were performing at the same level as other groups). Teachers' behaviour towards these students in the classroom was markedly different (e.g. they spend less time with these children), a difference that can affect children’s performance. Ultimately the teachers' expectations informed their behaviours, which in turn can confirm the bias teachers' hold. The point of all of this is not to blame teachers, but rather to point out we are all subject to implicit bias in our work because we are human, but if bias goes unchallenged then it affects people’s outcomes.
People who make policy are no different.
What can be done to overcome bias in policy-making?
There are many individual actions we can all take to overcome our unconscious bias, once we accept we all have it. Carla suggests some individual actions here.
But specifically, what can people who run organisations do?
Firstly, diversity in policy-making matters. We tend to favour what we ourselves have experience of, a singular experience of the world tends to (not always but often) lead to more of the same. Diversity of people and of life experience remains a problem in senior levels in all businesses, including government and politics. For organisations interested in the value that diversity brings there is plenty of help out there to call upon.
Secondly, in recognising that unconscious bias is an issue for everyone, there are tools that policy-makers can use to help overcome it. Many years back the Ministry of Health created the Health Equity Assessment Tool (HEAT). It is a series of questions to help people making policy consider exactly how a policy will work to create more equitable outcomes and experiences in society.
In my early days in health I would trot it out, but I wonder if junior policy-makers, or even senior ones use it, and do other government organisations have a similar tool: The Treasury, Social Development, IRD? The questions in HEAT would equally work to help overcome some of the unconscious bias that is at play in all other policy settings.
Of course it requires first that people in the system recognise that humans are all products of the society we live in, riddled with unconscious bias and in need of help to overcome it. As Carla Houkamau points out, it is more than possible to overcome our unconscious (and conscious bias), it just depends on how committed you are to it.
Your job as a scientist is to figure out how you’re fooling yourself.
— California Alumni, “Blown apart”,
Newsroom is powered by the generosity of readers like you, who support our mission to produce fearless, independent and provocative journalism.