Policy implementation is hard. It feels more like an art than a science. There’s a difference between knowing what to do (the theory) and how to do it – which comes from practice. I’ve been thinking about both, and want to share with you what have been the most useful frameworks and tools from my experience. This blog is a first guide to approaching complex problems.
One of the most valuable things I got from my Masters at the Harvard Kennedy School was a structured approach to addressing problems. The curriculum was intentionally designed to give you a blend of theory (Economics, Econometrics, Politics and Governance) and practice. As a Minister I didn’t go in depth in the theory, but it was as if I had a structure in the back of my head that essentially helped me make better decisions.
My experience in government – from working at the President’s office to acting Vice President, and later as a consultant – have given me the opportunity to adapt those frameworks and apply them in the real world, beyond the classroom. This is a brief guide to help you start thinking about how to approach complex problems.
Understand the problem
Imagine you go the doctor with a headache and some pain in your chest. You are worried that it could be something serious. You arrive to the doctor and she has other patients; so, she goes as quickly as she can. She explains that based on your symptoms it’s probably nothing, and prescribes for you to start taking aspirin every day. You’ll probably go to another doctor.
The minimum you would expect is that the doctor actually checks your symptoms, examines you, runs some tests and gets a deeper understanding of your condition before treating you. Sounds reasonable, right?
Well, the process in public policy should be similar. You first need to understand what’s going on. The problem is that we are often under so much pressure to have a solution, to take action, that we don’t take the time to go deeper and understand what’s really going on. So, the next time you feel like jumping into conclusions, stop for a moment. Remember the busy doctor.
How can you start?
Act with the curiosity of a three-year old and start asking yourself why.
The 5-whys is a simple technique developed by Toyota to drill down to the root cause of a problem. I like the idea of thinking of your problem as a question, which is how Harvard’s Problem Driven Iterative Adaptation framework – commonly known as PDIA – starts. Based on that question, think of possible causes that could explain your problem. For each of the causes, ask yourself why, then why again, and again. Five times.
If you were a doctor you want to know the symptoms, but you need to find the root cause of the problem to fix it. This technique helps you to do that. You can also download the PDIA toolkit for more examples and a deeper understanding of this framework.
The myth of the broken system
Remember, “there is no such a thing as a broken system. Each system is perfectly aligned to produce the results it currently gets”. The system is serving a purpose, it might just be different to the one you had in mind. A deeper understanding of the problem will help you identify the purpose the system serves.
Identify the type of problem
Problems can be classified in three types: simple, complicated and complex. The table below summarizes the characteristics of each and an example.
In this framework, ‘transaction intensive’ refers to the extent to which the delivery of a task requires a large number of interactions, nearly always involving face-to-face contact. The more transaction intensive a task, the more people that are involved.
‘Discretionary’ refers to tasks that require that individuals make decisions using the, often imperfect, information they have. Providers use their professional expertise or informal context-specific knowledge to assess a problem. For example, every time a doctor treats a patient, she relies on her technical expertise to diagnose that particular case.
Simple problems are those that are discretionary but non-transaction intensive. With adequate knowledge of the topic the problem can be solved. In other words, these are problems that “ten smart people in a room can solve”. Like designing a school curriculum.
Complicated problems are those that require coordination of specialized expertise to be solved. These tasks are transaction intensive but not-discretionary. You can also think about this type of problems as a set of simple problems. The construction of a hospital is an example, where you need more skilled people solving a part of a bigger project.
Finally, complex problems require an understanding of the unique local conditions and the capacity to adapt as conditions change. This type of task is transaction intensive and discretionary. For example, malnutrition interventions. A great part of the success of programs to improve nutrition depend on the individual’s conditions and behaviour. The design could include micronutrients, but if families don’t use them correctly it won’t have an effect on the child’s health.
Final reflections
To solve a problem, you need to understand the root causes of the problem first. This is easier to explain than to do. It only gets better with practice. Once you comprehend the problem clearly, you can categorize it. While in real life you won’t go back to the table necessarily, it will be like a ‘blind spot detector’ when you think about potential solutions to your problem. Not all types of problems require the same type of solution. While best practices work for complicated problems, for example, they can be misguiding for complex problems.
Photo by Siora Photography on Unsplash
