Iterative Adaptation - Policy Design and Managing Organizations : Ensuring a Learning India S4 E.007
[25th post in the Ensuring a Learning India series. 42 posts in total. One post per day. 17 more to go.]
[Note: This has personally been a difficult post to write to structure the ideas that I have been thinking since long. I believe there is still lot of scope to be refined. Hope the text below is clear.]
Iterative adaptation intuitively means adapting (to situations) by iterating. In other words, it can also be called ‘learning by doing’ or ‘prototype and reiterate’ as the situation suits. World Bank’s Social Observatory programme talks about ‘adaptive capacity’ of a programme. Lant Pritchett terms it as Problem Driven Iterative Adaptation (PDIA). Even in our everyday life, we learn most things by doing. One of my programmer friend says, an experienced programmer is the one who knows what not to do. This simple concept has significant implications for policy design and managing an organization.
I. Policy Design - Adaptive Capacity: ‘One size doesn’t fit all’ is a commonly used phrase. The idea is that while operating in diverse contexts, one needs to account for this factor and design the policy accordingly. But how does one know what’s the best for a particular scenario? May be it is only possible to know the challenges only while implementing the policy in that context, and hence the design has to be adjusted as per the feedback. Though intuitively it makes sense, it is surprising to see the number of policies which don’t do this, in fact do the opposite. Designing such adaptive policies throws up some challenges.
One, pre implementation vs. on the go. All aspects of the policy can’t be decided and be concretized during the planning stage. Some aspects can be designed from previous experience, some can be designed to ensure minimum damage to be on a safe side and some aspects can be left open for experimentation. Categorizing these aspects can be a challenging task.
Two, discretion vs. standardization. Even while being adaptive in nature, all policies that are operating in diverse contexts needn’t give discretion to frontline workers. For example, consider the McDonalds systems inside the shop. These shops in a country like India operate in fairly diverse contexts. The streamlined operations that one sees in the shop mightn’t have been designed perfectly in its first attempt. This would have evolved over time, incorporating feedback, thereby being adaptive. In such cases, the lessons from one context are transferable to other contexts. Hence, it is possible to build a single standardized policy through adaptation.
Same is the case with incentives for drivers of Uber and other app based taxi systems. The number of compulsory trips per driver per day and the incentive systems keep changing, as the management discovers new loopholes. Even in this case, same policy can be implemented across the board learning from diverse contexts. These are cases where adaptive capacity is inbuilt but not discretion.
On the other hand, there are cases where this mightn’t be possible. Having some pre specified rules and words of wisdom certainly helps, but at the end of the day, it depends much on the judgment of the frontline employee or the person at the cutting edge of execution. Sports is one of the best examples for this. There are certain dos and don’ts for batsman. But while playing on the field, it has to be up to the judgment of the batsman to decide the shot to be played, considering the ball, fielding, situation of the match and many other constraints. It may not be possible to program this and make it into a form of a standardized manual. Education and health have similar challenges. There is only so much that one can standardize. Each student and each patient can turn out to be a unique case. These are the cases where the policy should have both discretion and adaptive capacity inbuilt into it.
Three, autonomy vs. rules. One may need autonomy to deal with constantly changing situations but in contexts with weak state capacity, this may result in negative effects at least in short term. We already discussed this in previous posts.
Four, trade-offs between decentralization and centralization. In cases with weak state capacity, one might prefer to go for a rule based approach than giving autonomy. But even in some of those cases, it might actually be beneficial to devolve despite weak state capacity because that would be relatively much better than trying to operate centrally, again in the same situation of weak state capacity.
Five, resistance to change. Changing rules has its own challenges associated with it, resistance from people, investments in new approach and other things. Not everything is as easy as changing the rules for incentives, like the case of Uber. Constant reiteration without involving stakeholders can lead to experimentation fatigue and lead to lack of seriousness.
This is about the need for iterative adaption broadly and the challenges in pursuing the same. There is much more to it in terms of managing an organization.
II. Managing Organizations: Broadly speaking what do people in the organization do? They solve problems, though its nature may vary across. For the sake of our discussion, we can focus on real life problems and problems in managing organizations and not academic problems, though they also eventually end up solving real life problems. Whom can we learn better about solving real life problems? May be entrepreneurs.
Entrepreneurs around are solving many problems. Keeping aside the viability of business models, there are some clear and distinct features among those who execute successfully. One, motivation to solve the problem. The entrepreneur in-charge is motivated to solve the problem. Two, focus. Focus follows motivation. They are single mindedly focused on this. Three, skill. One needs to have a skill to solve the problem. This can be either own skill or by someone else. Four, resources to solve the problem. Five, power to take decisions. By the virtue of being motivated and also sometimes due to clients, there is an external pressure – incentive and penalty systems to perform, and are forced to adapt iteratively.
All these together may end up solving a problem. If one looks at companies like Flipkart and others, one can observe that it took years of dedicated focus and execution to solve the challenge of delivery and payment systems. Some say, there is still a lot of scope to be improved. In summary, successful problem solvers are skilled people working with focus, under systems of incentives and penalties solving the problem by adapting iteratively. There are structures inside with increasing focus and specialized skill as one moves down the organization chart. There is demarcation between those monitoring and those executing.
In the absence of this focus and demarcation, one person or a group of people are made responsible for multiple projects. This limits the focus that one can pay toward a particular function. When one is overburdened, then they act to ensure that worse doesn’t happen rather than working towards excelling in everything. In this process, that are low stakes go down the priority list. One is constantly trouble shooting and not acting with focus and vision.
This explains two paradoxes of the working of some government systems. There are some government programmes which exist only on papers but not much in action. Bureaucracy is constantly blamed for it. On the other hand, the same bureaucracy executes policies of huge magnitudes, taking everyone by surprise. Conducting elections in a country like India, ensuring that all the schools have toilets within the deadline set by the Prime Minister are no easy tasks. What explains this paradox?
In the existing government structures, there is not much place for specialized problem solvers (like the entrepreneurs discussed above). Same set of limited people are responsible for multiple tasks, which keep increasing with time. A private organization hires a new project manager and team when a new project starts, but does the bureaucracy do so? Often, someone in the organization is given the additional responsibility and over time these build up. In such situations, when there is so much to do, aspects that aren’t urgent and those that don’t necessarily result in pressure from public go down the priority list. This can partly explain the non-execution of some policies.
What makes the successful execution of policies by the same bureaucracy? Partly the reason is due to the nature of these tasks – building toilets and conducting elections, which are mostly operational and don’t need much technical skill. If they don’t need such skill, why weren’t the toilets built earlier? The reason is – focus. Earlier, bureaucracy had too many things on its plate and hence couldn’t focus on these. When the diktat comes from above, someone like the Prime Minister, there is a strong justification for them to keep other things aside and just focus on one aspect, which isn’t possible to do at normal times. This re-adjustment of focus alone can yield results in solving such problems. The problem is that this isn’t a sustainable approach. One can only keep other tasks aside for a certain amount of time. They can’t be ignored and hence the bureaucracy has to come back to multi-tasking mode.
Can all the problems be solved with just that mere focus? Possibly yes – Problems like construction of toilets don’t need technical skill hence can be solved within fixed deadlines. Problems like education and health can also be solved, though not in short time frame, when appropriate ‘specialized structures’ are in place and there is demarcation between monitoring and execution. What does this mean? It means that the generalists in bureaucracy responsible for monitoring shouldn’t be also involved in execution on ground. These should be done by ‘specialized structures’ monitored by bureaucracy. Each specialized structure, with the qualities of entrepreneur as discussed above, is involved continuously in fixing one problem. These specialized structures can be the ones built inside the bureaucracy itself, can be outsourced to either an NGO or a private organization depending on the context as long as it is a specialized structure with relentless focus on this problem, adapting iteratively till the problem is solved.. This alone can solve problems to a major extent.
In summary, we need specialized structures supporting the bureaucracy, working under their oversight, which iterate and adapt till the problem is solved. This eases burden on bureaucracy, bringing focus on to the problems, making it easier to track and ensure that it is solved. If not, even good ideas will end up not being implemented well, getting labeled as non-implementable ideas and failures. In reality, it may just be that, all they needed was focus, which they didn’t receive.
This ends the theme on ‘Governance’. The next eight posts in the ‘Ensuring a Learning India’ series will buildup on the ideas and principles discussed in the ‘governance’ theme to analyze some of the issues in school education.