THE first and most fundamental step in addressing any policy challenge is to accurately define the problem. This is because a nuanced understanding of the constraints at play will already give any policymaker a general sense of the range of interventions that could potentially address these constraints.

While in theory, having a problem-driven approach to policy challenges seems obvious, we do not always see it in practice. Instead, it is also quite common to see policymakers use solution-driven approaches, which of course often lead to suboptimal outcomes.

Let’s take a simple example to see these two contrasting approaches to illustrate this point.

Assume that a city is struggling with low student enrolment in high schools. In a problem-driven approach, we would not jump to any potential solutions and would instead try to understand the challenge better. In this approach, we would start by asking some basic questions: what are the differential enrolment rates in different parts of the city? What are the returns on education in the city? If the economic incentives are there for people to send their kids to school, what are some of the other constraints at play? Is it because parents do not know about the economic returns on education? Or because of norms around sending kids to school? Or are schools located far away from where people live, making accessibility the key constraint? If multiple constraints are at play at the same time, which ones are the most fundamental ones?

Defining the problem is a long and hard process.

Different answers to such questions would lead to a very different set of potential policy options. Say, for instance, the low enrolment rates were primarily due to lack of accessibility. In this world, we might want to focus on infrastructure/transport-based solutions (ie building schools in areas without schools or providing transport). If on the other hand, access wasn’t really an issue and parents simply did not want to send their kids to school, then the policy options would be very different (ie essentially aiming to target beliefs about schooling).

A wildly different (and incorrect) approach would be to start with a solution without trying to understand the problem in detail. Going with the same challenge of low high school enrolment, assume that we live in a world where parents did not want to send their kids to school because of incorrect beliefs about economic returns on education. If the city simply decides to build more schools in this scenario, the high school enrolment rate will likely remain stagnant, and we might see a lot of newly built empty schools.

Such solution-driven approaches to policy problems might even create more problems due to unintended consequences. Say, for instance, that now the city is struggling with poor learning outcomes of kids enrolled in schools. Again, agnostic to the constraints at play, the city now decides to allocate more non-earmarked resources to the education department where capital isn’t really the constraint. To add complexity to the matter, the education department does not have the capacity to absorb this additional inflow of capital because of capacity constraints. At the same time, the department needs to show that on paper they are using this additional money well without any understanding of the binding constraints at play.

The result? The education department tries to do several different things without any coherent strategy and partners with contractors for construction without having the capacity for proper procurement and oversight. This not only leads to stagnant learning outcomes but also to additional waste of resources (eg, by fostering wrong procurement practices within the department), making the problem more complex in the next round of the reform.

On the face of it, it might be obvious that for policy interventions to work, they need to be problem-driven and not solution-driven. At the same time, it is also quite common to see solution-driven approaches all around us. This is because defining the problem in good detail is a long and hard process. It requires asking the right questions and answering them in a systematic way while being open to any solution that it might point towards within the realm of political and administrative feasibility.

In the policy world, we often need quick solutions, which can make it hard to fully understand the problem. Sometimes, there is a particular temptation to being solution driven (eg. at a surface level, who would have an issue with the city building more schools?).

However, the only way to really solve policy problems is with a deep understanding of the constraints at play. Solving policy challenges in the absence of such an understanding is like shooting an arrow in the dark.

The writer has a doctorate from the University of Oxford and is a graduate of the Harvard Kennedy School of Government.

Twitter: @KhudadadChattha

Published in Dawn, December 27th, 2021