This article is written by Kulani Abendroth-Dias, behavioural scientist and PhD candidate at the Graduate Institute of International and Development Studies Geneva

I was recently part of a workshop where a team of behavioural science consultants had been invited to give a presentation about using behavioural insights for public policy in developing countries.

During the course of the interactive workshop, I noticed the reticence of my fellow female participants, and the dominance of male dialogue in guiding the conversation on issues faced by public servants and possible behaviourally informed interventions to address these problems.

During a coffee break, I approached the team of behavioural science consultants (who all happened to be white men from developed countries) and mentioned that, as someone who had grown up in this particular developing country, I knew that many women did not feel comfortable speaking up in a room full of men, and to perhaps make some extra effort in including them in the conversation in a more personalised manner. I joked that as a behavioural scientist myself, I could suggest a few "localised nudges" in building a more inclusive environment for female participation.

“How are you a behavioural scientist?” One consultant asked me, raising an eyebrow.

Really? That’s your question? I explained to him that I had studied social psychology and economics during my undergraduate and graduate careers, and spent the next few minutes explaining my professional and institutional background. I wondered if they’d have asked me that question if I was a white male from a developed country myself.

They nodded after I finished outlining my resume and spent the rest of the workshop politely avoiding my questions on methodology and rigour of evidence. I wondered if I’d looked more like them, would we have been at the bar later, discussing behavioural science over a few beers instead?

The power of behaviour change

There is a systemic issue with the way behavioural science insights are being applied within developing countries.

We are building an availability heuristic — a bias whereby we make mental judgements about the likelihood of an event based on similar examples that quickly come to mind — of behavioural science consultants working in developing countries all being white (often male) individuals sent in from developed countries. This problem isn’t limited to the practice of behavioural science in development.

The notion of "development colonialism" is not new, and the politicised ways in which development aid is dispersed, along with Westerners telling people what to do with it in developing countries, is acknowledged and debated. To diagnose the core issue of how or why this type of availability heuristic has developed in the application of behavioural insights in developing countries (like any good behavioural scientist!), one has to only follow the increase of international development projects that use behavioural science in some form or another.

According to the OECD, more than 200 public entities around the world were applying behavioural insights to policies as of November 2018.

The big aid organisations with the capacity to create normative change have started using behavioural insights with important implications for the practice of international development at large

Behavioural science is built on the work of social psychologists and economists among others, and has been given new relevance by researchers such as Cass Sunstein and Richard Thaler, who wrote the pathbreaking Nudge in 2008 (see also Daniel Kahnemann’s Thinking Fast and Slow (2011) and The Behavioural Foundations of Public Policy edited by Eldar Shafir (2013) for further research and applications).

Since then, organisations have formed to systematise and mainstream behavioural insights in various ways for policy, from the Social and Behavioural Sciences Team under the Obama administration to the UK’s Behavioural Insights Team in 2010, to the World Bank’s Mind, Development and Behavior Unit (eMBeD) in 2015.

Organisations began working to leverage behavioural insights for public policy in many developed countries, including the UK, Denmark, the Netherlands, Canada, Australia and France, among others, and this interest in behavioural insights has also rubbed off on the international development community (and importantly so). The big aid organisations with the capacity to create normative change have started using behavioural insights with important implications for the practice of international development at large.

From interventions targeted at reducing school dropout rates and increasing learning outcomes in Peru to testing the effects of a universal basic income on quality of life and gender equality in Kenya, the increased use of behavioural science has exciting potential to influence the practice of international development. Behavioural science is about understanding the biases that govern human behaviour and designing ideally small investment high impact interventions to target these biases to influence the problematic behaviour.

Apart from the quick feedback surveys on lessons learned at the end of the workshop, there is often a lack of long-term behavioural outcome measurement of the effects of these relatively high-investment/low-impact "interventions"

In development practice, we talk about changing people’s behaviour to bring about social change. An understanding of the social psychological and economic principles, i.e. the human biases and environmental factors that govern these behaviours, can help us actually bring about these behaviour changes in very real ways.

As we understand the biases that govern the traditional practice of development, however, we must see that the proliferation of certain types of development practices moving unidirectionally from developed to developing countries are unsustainable and counterproductive to the outcomes pushed for by development practitioners themselves.

Traps of the well-meaning

One of the traditional traps of development aid is the superfluous transfer of knowledge via two/three-day workshops that allow for the short-term ticking of outcomes achieved on donor reports but with little thought into, or measurement of, long-term change. Workshops are often held in swanky hotels where highly paid consultants and practitioners are invited to give talks on the importance of certain topics and disciplines for local practitioners moving forward.

Apart from the quick feedback surveys on lessons learned at the end of the workshop, there is often a lack of long-term behavioural outcome measurement of the effects of these relatively high-investment/low-impact "interventions".

By behavioural outcome, I mean creative measurement of how individual behaviour was influenced six months, one year, or even two years after the workshop. Did any of the participants work to incorporate learnings from incentivisation or choice architecture in their workplace? Were there follow-up questions on designing a behaviourally informed intervention targeting GBV? Was there a measured increase or decrease in the types of behavioural change discussed at the workshop? Did anyone conduct a scoping mission to investigate a wicked problem using qualitative or quantitative analyses? Was there a follow-up call on how to conduct analyses on ethnographic data collected? Or were these done by more consultants called in — essentially making the first three-day workshop a consultant pitch for further (highly paid) projects?

The people with the most influence are often those with the least first-hand experience of the problems they’re there to solve

International organisations have invested thousands of dollars of development money to host workshops and invite behavioural scientists to conduct behaviourally informed analyses within their home countries.

Not surprisingly, most of these behavioural scientists that we invite to give their views on behaviourally informed development come from developed countries. The imminent danger here is that we are replicating the same issues with the traditional ways in which development aid is practised in how we are applying insights from behavioural science within developing countries. The people with the most influence are often those with the least first-hand experience of the problems they’re there to solve.

Behavioural science in theory and practice should be a key means of addressing this problem because practitioners are supposed to be aware of these psychological biases and develop more targeted, localised approaches to seemingly intractable social/economic/political issues. The unidirectional flow of superfluous knowledge of "applying" behavioural science from developed to developing countries, often in exchange for vast sums of development funds, only contributes to the problem.

An end to "going local"

The potential for behavioural insights to bring about low-investment/high-impact interventions for social change is clear — it is the implementation of these insights within developing countries that needs to be addressed.

Despite all the challenges, I still tend to be optimistic about the use of behavioural science to influence traditional development practices. Where traditional development practices have failed, behavioural science itself can be a way forward. Behavioural science at its core is an informed means of thinking about a problem — a means of breaking down seemingly intractable, socially entrenched issues via localised, targeted interventions.

Yes, the current knowledge bases for social psychological and economics research (among others) that contribute to behavioural science is concentrated in the West. However, the best practices for applying behavioural science for development are growing in developing countries, from Peru to Tanzania.

Knowledge-transfer should go beyond design thinking to involve in-depth training from problem analysis and scoping missions informed by guiding theoretical and practical principles

Behaviourally informed interventions need to take into account social psychological biases and failings as we see them, and apply tweaks with low investment for high impact. In this way, behavioural science consultants need to go beyond the traditional three-day workshop where they "go local" and build long-term outcome measurement into their interventions. Knowledge-transfer should go beyond design thinking to involve in-depth training from problem analysis and scoping missions informed by guiding theoretical and practical principles.

There is a misconception among practitioners that behavioural science is an innovation tool to be "tacked on" to an intervention, and the "behavioural science unit" is to be housed exclusively within the innovation department of an organisation.

A better understanding of the social psychological principles that underpin behavioural science insights make clear that it is a way of thinking about a problem that can be mainstreamed across departments and corporate functions — from the design and implementation of development interventions to how HR functions.

We need to think about the long-term process of collecting and analysing data to understand a problem. Instead of investing our money in flashy three-day workshops, we need to be building local teams of behaviourally informed research scientists with an excellent knowledge of the biases that influence human behaviour to guide international development practice.

Beyond design thinking workshops

There have already been important steps taken by aid organisations themselves to bring about normative changes in internalising the importance of behavioural science for development. The United Nations Development Programme SDG Accelerator Labs work with up and coming talents, which can lead to local actors being better trained in and engaged with the design and implementation of behaviourally informed interventions. The OECD BASIC toolkit illustrates the importance of mainstreaming behavioural insights across organisational functions.

Lessons can be learned from the MineduLab’s setup and work in Peru, as well as the work by the Behavioural Insights Team working with the Basel Institute on Governance, Utrecht University, the University of Dar ed Salaam, and the UK Department for International Development’s (DFID) Anti-Corruption Evidence Programme to leverage social networks to bring about normative change in fighting corruption within the health sector in Tanzania.

More long-term collaborative exchanges of knowledge, where research advisors are not afraid to be technical with those on the ground go a long way to transform development practice.

We now need to focus on how the knowledge transfer of behavioural insights takes place, with longer-term outcomes in mind. We need to go from short-term tick-box exercises — often in the form of design thinking workshops — to training on what a behavioural outcome actually is, and how we can build it into different stages in an intervention.

We need to talk more deeply about the social psychological principles that have underpinned our most famous Nudges and behaviourally informed interventions. We need to talk about choice architecture, naive incentivisation, and how to bring about normative change within the context of intervention that have been carried out across developing and developed countries, and ground these interventions in basic social psychological theory so that individuals from developing countries can think more comprehensively on how they can be replicated or adapted within their own contexts.

We need to use principles from behavioural science to nudge our current ways of thinking about using behavioural science in developing countries

We need to talk about the basics of designing experiments, randomised controlled trials, and creative means of measuring impact. We need to talk about how most research that has informed behavioural science has been conducted with participants from Western, educated, industrialised, rich democracies (often small groups of college students, some online!) We need to talk about interventions that failed (or didn’t reach statistical significance). We need to discuss how that failure is a learning in and of itself. We need to use principles from behavioural science to re-examine how we apply behavioural science in developing countries.

Most importantly, we need to create the space for locally driven experimentation and failure, focusing on long-term impact versus the short-term ticking of boxes for donors. Transparency for donors on measuring long-term outcomes, going beyond short-term objectives, can go a long way to bring about meaningful impact.

Making BI more than a cliché

Behavioural science is exciting because of its potential to influence behaviour and subsequently bring about normative change. This impact is precisely why we need to be thinking more deeply about its implementation as practitioners.

Instead of outsourcing the collection of behavioural insights to external consultants, international organisations should have more in-house behavioural scientists who can mainstream these insights across corporate and developmental functions — in both developing and developed countries, at both HQ and field offices. We need to be going to schools and talking to university students in developing countries about behavioural science as a discipline.

Consultants should be working with eager-to-learn younger associates and interns on how behavioural science can be mainstreamed across their organisations. We need to think about how we can comprehensively work to train practitioners in behavioural science to bring about more long-term change in the decentralisation (from a sole innovation unit with an organisation to across all if not most units in an organisation) and localisation of behavioural science across developing and developed countries. We need to use principles from behavioural science to nudge our current ways of thinking about using behavioural science in developing countries.

If not, we run the risk of behavioural science being another cliché in development organisations, instead of unleashing its capacity to bring about meaningful normative shifts for social good.

Many thanks to Benjamin Kumpf for his valuable comments on an early draft of this article. You can follow Ben on Twitter via @bkumpf. — Kulani Abendroth-Dias

(Picture Credit: Unsplash)

Kulani Abendroth-Dias is a behavioural scientist currently based in Brussels, Belgium. You can learn more about her at https://www.kulaniabendrothdias.com. You can follow her on Twitter via @kulaniadias.