Where Does Charity Lie? A Day At A Project Site

I am cross-posting this piece I wrote for Frank Water from here

This week another water supply project goes live in a distant village in Eastern Ghats of India. This idyllic day of the monsoon month can easily fool a visitor into thinking that the households here are as abundant in water as the streams appear.

Walk to the source of spring for Debametla

Full streams or not, either way, the people must hike a kilometre at least to fetch water at any time of the year. Water quality isn’t even a distant concern. This is a situation that our new Gravity Water Flow System (GWFS) based water supply project changes as soon as it goes live.

Distribution tank where the spring water gets filtered

Gravity water flow system uses springs that are found in hills as the source of water supply. The community chooses the spring that will supply the water needs with technical support from a project members of our local partner after a selection procedure. The selection procedure is conducted to ascertain if the spring is suitable to build a GWFS using it as a source. That involves observation of spring’s perenniality, its yield, altitude of the spring and discussion with locals to know the history of spring.

Stand post from where villagers fetch water to their house. The farthest house is 50 meters from the post.

Once the source is decided, with the participation of community, expert from VJNNS (our local partner) looks at the route to get the water to the village and the location of distribution tank is decided. Sometimes the spring is as far as 3 or 4 kilometers from the distribution tank. This process is technically challenging in the case of GWFS because unlike most other supply systems GWFS doesn’t use any energy to draw water from the source to the distribution tank or from the distribution tank to the stand posts near the households. With zero energy consumption GWFS are the best for remote regions where electricity is limited in supply and unreliable.

A villager carrying water from the standpost.

But for such a project to work this requires cooperation in the community.  Like in the case of the first village Debametla that I visited today, the spring that was identified as a suitable source was located in a private property of an individual from another village. And if the spring was developed he or his village will not benefit anyway as his village was located upstream of the spring. But he still generously accepted to let the spring be developed and used to supply water to Debametla. Whereas in Vamegada Kotur village, the distribution tank had to be located in a particular elevation and that happened to belong to Samalingam and Krishnapadal. Both of them gave away their land for the purpose without a second thought.

Samalingam and Krishnapadal who donated their land for distribution tank, Vamgada Kottur .

As an inspiring day comes to an end, I begin to think where does real charity lie in this project. Is it an act done only by those who help finance it? Or is it also those two brothers who chose to donate a part of their land to set up the distribution tank for the supply system. The concern is not about whose act of charity is greater. Its about being able to see the unnoticed and often unacknowledged ones. They make a difference as much as any other act of charity.

Our work with the communities finishes that last mile which makes or breaks a project through these acts of charity. It is humbling and overwhelming to see the communities step in to lead the charge. Ours is perhaps a charity inspired by these million little charities happening in these distant corners of the world.


Thoughts on using Randomised Control Trials (RCT) to understand , evaluate and conclude how development program work

Its been many months I wrote any post here at TMN. Since the last post I have been busy traveling on personal and work trips. Many discussions, debates and conversations I thought were worth sharing here. But they failed to make it here sheerly because of my lack of discipline to write regularly. One particular subject which I wanted to write and still want to share remains to be the use of Randomised Control Trials (RCTs) by development economist to understand development programs and schemes rolled out by governments.

The pioneers and starts who made RCTs the vogue are Abhijit Banerjee and Esther Duflo. I have read their book in parts, read criticism on their approach (RCT) to development economics and many other articles and interviews by them. To not be judgemental about the approach I wanted to understand it further. That took me to  ‘IGC-ISI India Development Policy Conference’ 2014 in July. But the cases, experiments and presentations made in the conference did not change but made my opinions and doubts on the approach stronger. A recent interview of Esther Duflo, made me revisit RCT and compelled me to write this post.

What is RCT?

A randomised controlled trial (or randomised control trial;RCT) is a type of scientific (often medical) experiment, where the people being studied are randomly allocated one or other of the different treatments under study. The RCT is the gold standard for aclinical trial. RCTs are often used to test the efficacy or effectiveness of various types of medical intervention and may provide information about adverse effects, such as drug reactions. Random assignment of intervention is done after subjects have been assessed for eligibility and recruited, but before the intervention to be studied begins.

Source: Chalmers TC, Smith H Jr, Blackburn B, Silverman B, Schroeder B, Reitman D, Ambroz A (1981). “A method for assessing the quality of a randomized control trial”. Controlled Clinical Trials 2 (1): 31–49. doi:10.1016/0197-2456(81)90056-8. PMID 7261638

How can RCTs help development or alleviate poverty ?

Here is an apt description of what RCT can do to development according to the authors of Poor Economics:

The authors propose that although we do not know  “what works,” careful observation  of the poor to help design interventions, cemented by randomised trials to assess  these interventions, can help us identify what does. Those who have the power to intervene (governments, international organisations, NGOs, philanthropists, and the  global middle and upper classes) are assumed to be well motivated, so that once  the deficit in their knowledge is overcome (in part through the good offices of the  authors), they will act.

Source: Randomise This! On Poor Economics Author: Sanjay G. Reddy*

Here is what I feel about RCTs used to implement  development projects and evaluation of development projects:

  1. The method as a tool to evaluate how development program have been delivered, seems to do a good job at it, as the sample size are vast enough and wide  spread and well ‘randomised’. But everything that makes a program click or fail is not measurable.
  2. There are ethical issues when RCT is used to implement a development program and see how the program  pans out. But I do not feel so strongly about it, because implementing a program that will fail to a large set of population is worse than trying it out on a small population and learn from it.
  3. The outcomes and conclusions made after RCTs (especially the ones I heard about during the IGC conference and read in Duflo’s interview) are very strong. Most times RCTs are only trying to test a particular aspect of a program. For examples : how does teacher’s attendance have impact on learning and therefore how to improve it. I find that approach quite narrow. It assumes many other factors that  play an important role in the success or failure of a program to be constant. It neglects the context of these programs. There is very little of context taken into consideration, eg: social structure and norms where the program is implemented,which the conclusions are made. I find this over simplifies complex problems and addresses the reasons for their failure only superficially. There is very little ethnographic outline to any program that is evaluated with this method.
  4. Most of the RCTs are designed with an inherent belief that human beings respond to carrots and sticks and THATS ALL! There is very little time spent on dwelling upon deeper reasons for dysfunctioning of institutions, programs and individuals of a particular society.
  5. Importantly there is too much arrogance when any economist concludes based on their RCT experiment. Which only makes one want to dismiss it immediately.

Abijit Banerjee , Esther Duflo and cohort who practice RCT seem to be  in fashion. As much as 100 RCT programs are on in India only from The Abdul Latif Jameel Poverty Action Lab alone.  Many state government agencies  here are already engaging with them quite a bit. And in all 500 odd such experiments across the world is happening .RCTs are very expensive. If only they could just make it more holistic, their work will have the impact they desire and more.