Policy Start-Up in a Skeptical Marketplace

Authors: Colin Dobson, Policy Horizons Canada
Document Type: Policy Insight
Published Date: Friday, June 1, 2012 - 4:00am
ISBN number: PH4-119/2012E-PDF, 978-1-100-20876-3
Alternative Format: 2012-0109-eng.pdf

Decision-makers are being challenged to rethink their approaches to improving public policy outcomes and there are new ideas being offered to them. While these alternatives are promising, translating them into practice across the whole population could be costly with the absence of evidence and public support. So, how can alternative interventions be introduced?

A Recipe from the UK

In a recent visit to Policy Horizons Canada (Horizons), Dr. David Halpern offered a straight-forward recipe for the introduction and scaling-up of a novel public policy intervention. His account was based on his experience as head of the Behavioural Insights Team (BIT) in the Cabinet Office of the government of the United Kingdom (UK). The strategy relates to the implementation of "behavioural" mechanisms, but might be generalized to the introduction of other novel instruments. It also appears to provide insight into the multiple benefits of effectively implemented small-scale policy experimentation.

What are behavioural interventions?

Based on insights from behavioural sciences, these interventions strive to find intelligent ways to encourage support and enable people to make better choices for themselves. This might include timely and personalised information at the point of decision making. It might also include "nudge" mechanisms that work by strategically altering environments in which decisions are made. If successful, they enhance the probability of certain decisions and in so doing contribute to public policy outcomes. They aim to be subtle and non-coercive. Please see the Horizons policy brief Nudge me baby one more time.

While Dr. Halpern's team had approval to proceed from the Prime Minister, they were facing healthy skepticism from the public when it came to the possibility of government employing such interventions. Despite the presentation of many benevolent possibilities, the public could still imagine controversial and even potentially inappropriate applications by government. There were also questions pertaining to impacts and efficacy over the short- and long-term.

In light of this, the BIT took the approach of "demonstration by experimentation". If the approach were to be broken down into its key ingredients, it would include an artful blend of experiments that are:

  1. Small in number, at least at first;
  2. Small in scale;
  3. Highly feasible to execute;
  4. Quite uncontroversial as policy interventions; and,
  5. Thoroughly measured and verifiable.

Willing and credible partners were also essential. For instance, they found certain courts whose bailiffs were open to trying new approaches to communicating with those having not paid court fines by sending personalized text messages prior to collection. This was done a day prior to a bailiff visit and produced quantifiably improved results – payments jumped from 5% to 33%.

Similarly, the BIT altered late payment letters to those owing taxes so as to include the rate of timely tax payment in individual debtors' immediate town. This intervention alone raised self-assessment tax payment to 98.3% from a baseline of 90%. (Image source: BIT Annual Update 2010-11.)

Diagram from Behavioural Insights Team

By measuring and documenting their interventions and their positive effects, the team demonstrated "wins" to an oversight committee of high-level public servants and political advisors, other government departments, as well as the public in the form of a committee of academics specializing in behavioural sciences. The experiments also served to show the competence of government officials in executing the interventions. Perhaps more importantly still, they served to display the spirit with which the novel intervention was being applied. It was being done in an evidence-based, transparent and step-wise fashion in accordance with public acceptance.

Food for thought in Canada

What might be some of the implications of demonstration by experimentation for the Government of Canada? In a certain sense, all policy is to some degree experimental in that it is monitored and evaluated for effectiveness and adapted to reflect what has been learned. Moreover, there are centres of policy experimentation regularly informing federal agencies, including the Social Research and Demonstration Corporation and the CIRANO research group. But is there a greater role for more deliberate, up-front and small-scale policy experimentation in federal processes around policy and program design? For example, would such policy experimentation be a useful approach to explore innovative approaches to regulation? In the absence of experimentation, how significant are the disadvantages of potential over-reliance on current models and assumptions?

Experimentation can often pay for itself when results point to more effective ways to implement policy, but it still requires resources and time that can be challenging to generate. There are also potential ethical issues related to policy experimentation, such as fairness when some individuals are placed in placebo (control) groups. These and other barriers can be addressed in part by ensuring transparency around the experiment and by learning from others. Inter-departmental and inter-governmental communication and coordination are potentially quite helpful in this regard. What insights could other governments learn from the "smart meter" implementation in Ontario and British Columbia, for example? There are also non-governmental sources of information. When it comes to behavioural interventions, there is now a growing database of nudge applications or case studies at a Danish website www.inudgeyou.com.

A Promising Approach

The demonstration by experimentation approach used in the UK to introduce a novel intervention into a skeptical marketplace would appear to offer useful lessons for us here on the other side of the pond. It shows that when used transparently, experiments can be an effective way to introduce new interventions while simultaneously building credibility and acceptability of new policy instruments and approaches, thereby enhancing public trust and collaboration.

References and Further Reading