Behavioural science offers the responsible business field the opportunity to test, trial and experiment. This would build a robust evidence-base to definitively know whether our interventions work, how best to optimise them and what to abandon entirely.
On the 'Behavioural Science in Practice' course that I attended Nick Chater, Professor of Behavourial Science at Warwick Business School, described an “experimental revolution” that began with plant biology and agriculture in the 1940s and 1950s, moved to bioscience and pharmaceuticals in the 1970s and to psychology in the 2000s. Before these revolutions these fields were “crafts” with knowledge being passed down through generations. The transition to evidence-based science was not seamless however: "when trials were first introduced in medicine, they were strongly resisted by some clinicians, many of whom believed that their personal expert judgement was sufficient to decide whether a particular treatment was effective"[1].
The experimental revolution has led to the gold standard of science – randomised controlled trials (RCTs) – that move away from anecdotal evidence (“I did X and Y happened”) to establish causality. This can only be achieved through RCTs which use control groups (receiving no intervention) and treatment groups (receiving the intervention) and measuring the results. Experiments that use before/after comparisons cannot establish causality because there are too many variable factors. We need to isolate our intervention and definitively prove whether X leads to Y. We expect this of the ‘hard’ sciences but the social sciences have a long way to go.
Behavioural Insights Team (2013), 'Test, Learn, Adapt: Developing Public Policy with Randomised Control Trials'
Why we must experiment
The current absence of evidence in responsible business and business and human rights is not a criticism of the field. This is the reality for much public policy work, the development field[2]and countless others. Only in the last decade have governments turned to evidence-based policy making – most notably in the UK with the launch of the Behavioural Insights Team or ‘Nudge Unit’ and there are now over 196 institutions using behavioural insights and experiments in public policy[3]. Without evidence backing up our interventions we simply don’t know whether what we are doing is effective or ineffective, a good use of time and resources or a bad use, leading to the changes we want to see in the world or having counterproductive effects.
Examples of why we should experiment abound. What sound like sensible solutions may not be so sensible. Take the issue of how to prevent young people from becoming criminals in adulthood. ‘Scared Straight’ was an American intervention that took children to visit adult prisons. The aim was to deter them from a life of crime by showing them what may lie ahead. The programme was the subject of a film by the same name which won Best Documentary at the 1978 Oscars. This led to the programme being scaled up and adopted in several more American states. The problem was that when the programme was measured for its effectiveness, not only did it not work, but it increased the future crime rates of those who visited the prison[4]. This shows why we should test our well-meaning interventions before we scale them (and commit public money to potentially ineffective programmes).
How can we run responsible business experiments?
So, we must run experiments to know whether our interventions work, and we must run them in our own contexts and not simply rely on academic literature and what has worked elsewhere (future blogs will elaborate on why this is). Experiments are generally underutilised in organisations[5], but companies are ripe with potential because they are used to experimenting (for example through A/B testing on websites to test which versions lead to greater engagement or purchases). This spirit of innovation and testing what works can readily translate into experiments to enable better behaviours that impact people.
I have identified a few areas where we could run experiments within the responsible business context:
First, focusing on what drives internal commitment and buy-in within the company. This would involve determining what messages resonate and lead to senior leaders dedicating increased resources and capacity to human rights issues. This could be conducted in lab settings to determine which aspects of the business case resonate and lead to action. Do executives respond to potential losses from getting things wrong more than the potential gains from getting things right? Does tapping into moral and ethical reasoning work? Are executives influenced by what competitors are doing?
Second, running experiments to test our problem-solving interventions on real-world human rights impacts at local levels. What are the specific behaviours that need to start or stop to contribute to some of the big challenges of our time (e.g. modern slavery or privacy rights online)? This requires more thinking and collaborative efforts to identify and measure the specific behaviours we are trying to change. But this will be critical in proving the efficacy of behavioural science to the responsible business field.
Third, experiments could focus on understanding how to get people (employees, contractors, communities, users) to report complaints, grievances and human rights impacts. Getting information on what is happening within the company and what impacts exist is difficult and yet crucial to the company being able to respond to issues and prevent adverse impacts. There are likely simple adjustments that companies can make to processes and operating environments that will yield improvements and demonstrate results.
The human rights community may feel unease at the idea of running experiments on control and intervention groups because, while in the long-term it may identify how best to prevent harms, in the short-term it may lead to harms playing out in control groups. Resolving this tension will require more discussion.
In the months ahead, I plan to have conversations with business practitioners, behavioural scientists, responsible business experts and academics about how we can run experiments and randomised control trials in the responsible business field. Please do get in touch with me if you have ideas at katrynwright[at]googlemail.com.
[1] Behavioural Insights Team (2013), 'Test, Learn, Adapt: Developing Public Policy with Randomised Control Trials' p.13
[2] You can read about how the experimental revolution is being applied to the development field in ‘Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty’ by Abhijit V. Banerjee and Esther Duflo and the work of the Poverty Action Lab at MIT
[3] The OECD maintains a list of institutions using behavioural insights in public policy
[4] Owain Service and Rory Gallagher (2017), ‘Think Small: The Surprisingly Simple Way to Reach Big Goals’, p.171-172
[5] O. Hauser, E. Linos, T. Rogers, (2017), ‘Innovations with field experiments: Studying organizational behaviors in actual organizations’, Research in Organizational Behavior