The systems development crowd has been busy. We have moved from facilitating to weaving, constraints to leverage points and relatively comprehensible diagrams to maps requiring their own legends. Systems are complex but recognising complexity and drowning in it are different things.
We’ve argued for going back to basics before, during and after the recent DCED seminar. But what are the basics? Given that we’ve already tried brokering, convening, dancing, ecosystem building and, most recently, weaving, surely there’s room for detecting too?
Detectives don’t start by mapping everyone who might conceivably have influenced the crime. They start with three basic questions: motive, means and opportunity. It turns out they’re not bad questions for approaching systemic change either.
Motive, means and opportunity won’t tell us everything about a system. Nor should they. They help focus the investigation: what behaviour are we trying to understand, what might explain it and where might there be scope to change it? From there, we can use whatever tools help answer those questions. Detectives have fingerprints, DNA, CCTV and phone records. We have our own analytical toolbox. The point is to start with the questions, then choose the tools that help answer them – not the other way around.
Motive: why would anyone do this?
Development is fond of assuming that people don’t do things because they don’t know how. Farmers need training. Businesses need the aptly named “business development services”. Government officials need technical assistance. MSD was supposed to make us more suspicious. Before asking whether someone can do something, ask why they would. What do they gain? What might they lose? What happens if they do absolutely nothing?
A bank might be capable of lending to small businesses and still prefer not to if it’s easier to lend to bigger ones. A retailer may understand poor consumers well enough and conclude that richer ones have more money. A government agency may appreciate the merits of a reform but still resist it if it doubles the workload with no additional budget. These aren’t necessarily capacity problems. More training may simply produce better-informed people with the same reasons not to change.
So, start with motive. Who has a reason to behave differently, and is that reason strong enough to survive our departure?
Means: can they do it?
Motive isn’t magic. A business may spot a great opportunity but lack the finance, skills or distribution network to pursue it. A government agency may genuinely want change but lack the authority to make it happen. Now we may be dealing with a means problem.
We’re back in development’s happy place. Information, skills, finance, technology, organisational capacity. Problems we can spend money on. Give us a budget and we can have a consultant mobilised, a toolkit drafted and a capacity-building workshop in the diary before anyone has had the poor judgement to ask whether any of it will change behaviour.
The trick is getting the order right. Starting with means gives us a world full of training needs. Starting with motive gives us a more useful question: what is stopping someone who already wants to act? And then comes the uncomfortable question: whose means are they using? With enough programme money, almost anything can be made to work for a while. We can pay the staff, subsidise the product, hire the consultant and organise the meetings. The activity happens. The indicators turn green. Everyone goes home delighted. The detective might remain suspicious. If the actor has the motive but can only act because we’re providing the means indefinitely, we haven’t cracked the case. We’ve joined the gang.
Opportunity: what makes change possible now?
Motive and means are not enough. People act within systems that open up opportunities or close them down. Solar equipment suppliers may see a potentially profitable market and have suitable products, but banks don’t finance installations and feed-in tariffs have yet to be established. Agricultural processors may be keen to work with farmers to secure more reliable supplies of quality produce for export, but cooperatives resist allowing firms to work directly with their members. A government may want reform and have authority to pursue it but face political resistance.
The question is not simply what else is wrong in the system, but what makes change feasible now. Perhaps technology has lowered costs, consumer preferences are shifting or competitive pressure is mounting. Maybe a first mover is experimenting, investing or getting annoyed enough to try something different. Systems analysis should help us find the loose thread, not catalogue the entire sweater. An ecosystem map containing 83 actors may be an impressively accurate description of reality. It does not necessarily tell you what to do on Monday morning.
Follow the evidence
Of course, detectives don’t arrest someone because motive, means and opportunity looked convincing on a flipchart. They start with what they can observe, form hypotheses about what explains it, then test those hypotheses against further evidence. MSD should work the same way. We can analyse a system for months and still be wrong about why people behave as they do. Incentives are inferred. Constraints are hypotheses. Opportunities that look irresistible from the project office can look rather less compelling to the person risking their own money.
So, put our assumptions to the test. If we think a bank wants to lend but lacks information, what happens when it gets better data? And watch behaviour. Not whether people enjoyed the workshop, signed the MoU or nodded enthusiastically while a consultant explained the business case. Do they invest, adapt, keep going when we stop paying? Behaviour is evidence.
Case closed?
None of this means systems are simple. They aren’t. But our response to complexity doesn’t always need to be another layer of complexity. Asking simple questions isn’t an argument for thinking less. It’s an argument for being clearer about what we’re trying to find out. Who has a motive to change? Do they have the means to act? Where’s the opportunity that makes change possible? And what evidence – preferably something more than nodding heads in a workshop – would give us confidence that we’re right?
That’s enough to begin the investigation. We can reach for more elaborate methods and tools when the case demands it. But start with motive, means, opportunity. Elementary, really.