July 17 2013
Lant Pritchett gave a seminar yesterday at the Turkey’s Ministry of Development on how to design and implement social policy in a better way. Over 60 key experts and representatives from social policy line ministries were present in the seminar. The event was organized jointly by Development Analytics and the Ministry of Development. It was the first seminar of the newly launched Development Seminar Series of Development Analytics and Social Policy Seminar Series of the Ministry.
Pritchett started his talk by describing accountability and continued with the concept of isomorphic mimicry. Isomorphic mimicry a very interesting natural phenomenon; it is the case where one animal has almost the same look of the other animal which is more advantageous in the nature to protect itself. In his talk Pritchett gave the example of two snakes which almost look the same but one is poisonous while the other is not. Hence the snake which is not poisonous gains all the advantage in the nature of looking as if it is the poisonous one. This situation is used as a metaphor for many systems & institutions e.g. line-ministries in government which are responsible from implementation of social policies. So that they “look” like they are what they are but in fact they are not “functioning” as they should.
Pritchett continued with another interesting story -this time using turtles. The turtles, he explained, cannot be beaten to move ahead since they have very strong shells. Thus they will just hide inside the shell until the beating stops. Whereas in order for the turtle to move forward he should feel safe enough to take his head out of the shell and start moving. The turtle is used as a metaphor for line-ministries which have very strong self-protection mechanisms to protect themselves from outside “attacks” that use force to make them go some way or punish them for not following a certain way. So the mechanisms used for these reasons won’t make them move forward, while on the contrary line-ministries will just end up hiding inside their strong bureaucracy shell and staying there as long as the “beating” continues. What should be done instead is making the “turtle” feel safe thus giving the line-ministries some authority in implementing plans instead of heavily planning them from outside and punishing them for not being in line with the imposed plans.
In order to overcome these kinds of problems in systems and institutions he proposes “Problem Driven Iterative Adaptation (PDIA)” as a possible mechanism. This mechanism (1) looks to solve particular local problems; as opposed to solving externally nominated problems –e.g. by World Bank- or problems basically defined as lack of a solution; (2) Solves problems via creation of an authorization environment that facilitates positive deviance and experimentation as opposed to a heavily planned implementation map where deviance is not allowed; (3) Uses monitoring, experiential learning and impact evaluation in tight feedback loops during the implementation processes as opposed to feedback loops that are not aligned with the timing of decisions and does not allow for experiential learning during implementation process; (4) Scales of learning through diffusion, meaning that agents in the implementation process should believe in the innovations and be willing to implement them as opposed to the leader imposing the project on the agents and they just follow.
In his paper [1] Pritchett gives some examples that are similar to PDIA such as “Cash on Delivery” aid promoted by Center of Global Development. With this mechanism aid is given to countries on condition of achieving some results, so the methods to achieve the results are left to the country. Creative destruction techniques implemented by Google were also mentioned as a similar approach in the presentation.
The full presentation can be downloaded at this link.
[1] Matt Andrews, Lant Pritchett and Michael Woolcock (2012) ‘Escaping capability traps through Problem-Driven Iterative Adaptation (PDIA)’ Working Paper No. 299, Center for Global Development (forthcoming in World Development)
