Experiment and reason

The 2019 economics Nobel Prize for Banerjee, Duflo and Kremer (hereafter, BDK) offers much to celebrate for Indians, Bengalis and Frenchwomen among others. Designers, in their modern role as global problem solvers, should join in. They have much to be inspired by.

The practices of economics and design appear to have little in common. But they share common ground when it comes to intervening in real-world problems. Those, for instance, where poverty must be tackled (or hygiene improved, energy saved, or ever more cars parked).

Both these professions, along with businesses and governments, are one when their work interacts with the consumer, citizen or user — different terms for the same equally intractable and ultimately, central human being, especially when seen as a psychological creature, rather than just a bundle of physical or even rational wants.

This is territory traversed by previous columns like “Good Design: It’s All in the Mind”. which appeared a little over two years ago. It referred, not for the first time, to Daniel Kahnemann, who won the economics Nobel in 2001. And somewhat presciently, to Richard Thaler, who would win the big prize three weeks later.

BDK’s work is as inspiring for thinking about design as these earlier winners, because it affirms kindred ways of thinking, ideologically and practically.

BDK have been recognised for giving great force to experimental economics as it applies to tackling poverty. Their work champions the randomised controlled trial (RCT), a type of experiment regarded as the highest-quality evidence in medicine. Briefly, randomly selected patients are given the treatment that’s being investigated, and the results are compared to the untreated “controls”.

The aim is to discover which (medical, or in BDK’s case, anti-poverty) interventions work, against the background that many programmes fail. Incentives, the dominant tool proposed by economic theory, or subsidies, can underperform or backfire. Implementation isn’t always to blame, and indeed the behaviour of the intended beneficiaries can seem puzzling. Or irrational, as economists would have it. 

(Top, from left) Nobel laureates Abhijit Banerjee, Esther Duflo and Michael Kremer; deworming medications administered to Kenyan schoolchildren; (bottom, from left) randomised controlled trials; Banerjee and Duflo’s 2011 prize winning book on the effectiveness of solutions to global poverty using an evidence-based randomised controlled trial approach

It finds, perhaps intuitively to some, that the poor stay poor in part because they think differently from the rich. They are less likely to borrow money to make an upfront investment expense that would benefit them in the long run, for example. They are, apparently paradoxically, likelier to borrow to save. A theme that runs through the BDK work is that the poor have trouble thinking in terms of a future. For one, the rich, too, can make similarly silly decisions, but can survive them. The poor may find the effects of a bad call irreversible — no future.

The deep design of the BDK approach rests on human-centred practice, a phrase enthusiastically espoused by designers. By treating the poor as people to be understood, rather than mystifyingly stupid, the practice is inherently empathetic. The poor person is rational, once the peculiar circumstances are understood; the fault is in the intervention, not its target.

The deep designer will also approve of the BDK method’s deep pragmatism rather than its theoretical nature. It focuses on what it can change, rather than why it works (the ultimate reward for the theorist). It proceeds from a first-approximation observation about the inner workings of a problem and tests it via an RCT. If deworming Kenyan kids improves school outcomes, then subsidising the pills makes sense. (Interestingly, the already affordable deworming pills were effectively adopted only when entirely subsidised — why commit to an upfront expense with uncertain benefits?). This gives an unimaginably large return on investment — if it were a school performance improving drug, what would you pay?

This pragmatism leads to a type of cautious, reluctant theorising that should be dear to the design temperament. Proceeding from practice to theory is an underrated source of scientific knowledge. And ultimately, maybe the theory isn’t there? So be it.

Yet it’s a common criticism of the BDK way. The interventions don’t always transfer well; the Kenyan experience could fail to replicate. Local situations require local solutions, or they might not. Economic and medical RCTs have the same limitations: the difficulty of true randomisation, or that variations in individual outcomes might be large even if the average effect is favourable, and so on. Some raise ethical concerns (informed consent, and the withholding of benefits to the control arm, for example).

Thinking on poverty has typically rested either on the great themes of economics (inequality or inflation, for example) or on explicit welfarism. Yet their effect on an individual’s experience of poverty is indirect. Put another way, the average person does not experience the effects of GDP.

Like design, the BDK approach affirms the importance of the small, tangible and directly observable intervention, without challenging grand theory. The recognition that small things matter is food to the design soul. That an experience can be altered to great benefit; that perception is as real as physical reality.

If a poor person behaves more appropriately when wages are paid directly into his bank account, than paid in hand, then how we pay matters. This marginal detail could be the difference between a path that perpetuates poverty and one that leads, just maybe, to a transition to a non-poor state. A practice that becomes a new default may work where hours of explanation fail, because practice often changes belief more reliably than the other way around.

Understanding objective factors (those that everyone is subject to, all the time, like prices) in the light of a subjective ones (those that ultimately drive action, for varying reasons, like pride or risk) that results in a good system or well-designed objects. Designers, and problem solvers in the social domain — that’s all of us, to a degree — should delight in and internalise these ways of thinking.

Start with reading Poor Economics (it’s a breeze).
 

Itu Chaudhuri runs Itu Chaudhuri Design, a design and branding firm, in New Delhi; itu@icdindia.com