"Poor Economics" - my review

I just finished reading “Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty” by Abhijit Banerjee and Esther Duflo. I have to admit that at the beginning I was skeptical because I knew that the two MIT economists are huge fans of randomized control trials (RCTs). I do not share their enthusiasm for RCTs and I believe that it is a methodology that belongs to pharmaceutical trials, not development work. I am a practitioner, more at ease in the field than behind a desk. My understanding of the sector comes from years of experience in the field, traveling from country to country, talking to people and comparing projects. I prefer a holistic approach to methodological silver bullets. However, I gave it a try and I was really surprised to find out that the authors are not methodological purists and in many cases we get to the same conclusions.

I was very impressed by their honest analysis of microcredit, a sector that has been either eulogized or vilified in the last 10 years. Their perspective is that most of the loans do not make people rich nor they send them burst. In the great majority of cases, microlending allows people to just get by. Which is already a great result in my opinion, but certainly not what the big actors in the field want to hear.

Another development myth debunked by “Poor Economics” is that we are all entrepreneurs. Or actually, that the poor are all entrepreneurs. In reality, poor people want secure jobs in the public sector, like most people do, everywhere. But since they understand all too well that their next meal is not guaranteed, the poor are also more risk-adverse than their wealthier counterparts. The less you have, the smaller the margin of error in your actions. But entrepreneurs have to take risks and be bold. It is difficult to find both characteristics in the same individual, let alone in the whole poor population.

Banerjee and Duflo will not make many friends among right-wingers with their pragmatic analysis of the limits of a private sector approach. But their views on education will not improve their reputation in left-wing circles either. In a nutshell, they say that school programs for the poor are too ambitious. Much better results could be achieved by focusing on the basics: reading and writing. As a son and brother of dedicated teachers, I am not completely comfortable with this idea. There are all kinds of students in one class, smart and dumb, interested and bored. Lowering the educational goals to get everyone at the same basic level at the end of the few years of schooling that the poor can afford seems to be unfair to the kids who could really get much more out of their school days.  On the other hand, I perfectly understand their point, having witnessed college graduates that could not read fluently a simple text.

But the part that I like the best is the final chapter where they present 5 key lessons that can help improve the lives of the poor, and therefore the effectiveness of development work:

1.     The poor often lack critical pieces of information and believe things that are not true

2.     The poor bear responsibility for too many aspects of their lives

3.     There are good reasons that some markets are missing for the poor, or that the poor face unfavorable prices in them

4.     Poor countries are not doomed to failure because they are poor, or because they have had an unfortunate history

5.     Expectations about what people are able or unable to do all too often end up turning into self-fulfilling prophecies

And finally, the two suggestions that we, development workers, should always bear in mind when designing strategies, programs or projects:

I.       Understand how people decide. Too often we let deep-rooted convictions and ideology guide our work.

II.      Devote attention to small changes in the way institutions work, because small tweaks in the vision and practice of institutions can have a huge impact (This is pure outcome mapping! I love it.)

Reading this book was quite an experience. I started with a heavy bias against the methodological approach and I tossed it aside for a few days after disagreeing with some of the earlier analysis. But I went back to it because it is also engaging and rich and full of interesting insights. Eventually, I saw that Poor Economics says things that are not that different from what I hear people say in the field. For the price of two caramel lattes (9 USD for the kindle edition), I think it is a must-read for both development workers and people who are interested in understanding poverty. 

Bamboo charcoal needs a make-over

No shortage of supply

Last week I was in Ghana, monitoring a project that promotes bamboo charcoal as an environment-friendly alternative to wood charcoal. The justification of the project is straightforward: 80% of primary energy consumption in Ghana is satisfied with firewood and wood charcoal causing high deforestation rates; switching from wood to the more sustainable bamboo, you can contribute to preserving the environment.

The advantages are quite impressive: bamboo grows quickly (3-4 year to maturity) and spontaneously; the supply far exceeds the current demand that comes mostly from the construction sector; lab analyses show that the calorific value of bamboo charcoal is as high as wood charcoal; last but not least, women prefer to cook with charcoal because it lasts longer and doesn’t produce nearly as much smoke as firewood. With these credentials, I expected high adoption rates. I was wrong. Households cannot be bothered to use bamboo charcoal instead of wood for a variety of reasons. I heard complaints about cooking time, amount of ashes, price, size of the chunks, etc. To sum it all up, bamboo charcoal is considered a grade B charcoal that sells for the same price as grade A. Among potential buyers, the decision is a no-brainer: they stick to wood charcoal.

However, of all the justifications I heard, the only one that seems to be supported by facts is that bamboo charcoal is being sold for the same price as charcoal made from wood, product that households have been using for generations. All the other issues (size, ashes, heat, etc.) are far less clear-cut as I heard both complaints and praises of the new type of charcoal. So, according to my highly un-scientific method, they cannot be taken seriously into consideration. Since there is no agreement on the performance of bamboo charcoal, I suspect that people are still strongly influenced by their preconceived ideas about it. 

Bamboo charcoal suffers because of its poor image among potential buyers. Bamboo is flexible and strong, while wood is hard. People know from experience that hard wood makes good charcoal; therefore in their mind bamboo doesn’t make good charcoal. End of the story. And pricing it on par with wood charcoal was the final mistake. Nobody buys a new/inferior product for the same price as an old-time favorite.

In this case, an oversimplified logic poorly served the organization involved in the project. They analyzed the technical and cultural components of the issue, but completely ignored the economic/marketing challenges posed by introducing an innovative technology in an established sector. A good marketing plan could have removed some obstacles to bamboo charcoal. 

Why actors matter

I received some very good questions on the outcome mapping discussion forum following my post on logical framework vs. outcome mapping. Another member in particular, Chew, asked me to present examples of results-based projects that need to focus more on institutional relationships to improve their performance. Here is my reply:

Example n.1 is a very ambitious food security project. Here is the logical framework:

OO: To reduce the incidence of poverty and improve the well-being of the population in XXX

PP: To improve food security, income generation and socio-economic conditions of small scale farmers in XXX and XXX District through the development of a sustainable and efficient agricultural sector.

R1: Appropriate agricultural practises developed, land productivity and farm yields enhanced for 4200 small scale farmers

R2: Management of post harvest improved, availability of storage facilities extended to serve 3,600 HH and selling of produce optimised.

R3: Diet diversification improved

R4: Capacity of operating conscious planning and management of business at HH level ameliorated

R5: Access to financial resources and productive inputs improved 

It would fit in the bottom-right corner of my table (hard project, small organization). The LF is pretty good: the logic is clear and coherent and the indicators are very specific (not included here). There is no confusion in my mind when I analyze the cause-effect relationships: with more and better inputs, training and infrastructure, farmers will be able to get more food for consumption and even sell some of the surplus. The timeframe is tight though (2 years) and it takes a huge effort to achieve all the outputs indicated in the LF: production, productivity, number of hectares, number of trainings, etc.  The project team dove headfirst in the implementation dedicating 100% of its attention to the activities. When the mid-term monitoring mission arrived, it represented the first opportunity for the project managers to analyze the situation from an outside perspective. What they saw wasn’t encouraging: in their frantic effort to keep up with their ambitious design, they had forgotten about the other actors working in the area and had promoted only a limited involvement of local institutions.

The result was that the outputs had been produced, but very few people were using them, the impact of the intervention was very small and there were very slim chances of achieving sustainability. The different stakeholders hadn’t modify their behavior and the relationships between them had not changed since the inception of the project:

•      beneficiaries’ organizations were brand new and mainly composed of illiterate farmers, living in remote areas. They had received trainings, but remained institutionally weak and had very limited technical capacity. 
•       Local development organizations showed very little appropriation of the intervention. They were getting paid to perform a task. Period.
•       Local government officials were satisfied to receive reports every semester and take credit for the distribution of tools.

A combination of results- and actor-based approaches would have contributed to a more balanced intervention, where productivity is linked to a change in behavior and relationship among stakeholders. Adopting an actors-based approach, the NGO would have set achievable goals for the groups of beneficiaries and promoted a bigger role for local institutions in order to enlist them as committed actors in the process of achieving food security in the area.

Example n.2 is a solid-waste management project aimed at increasing recycling in a big Latin-American capital. The intervention fits somewhere in between the two bottom quadrants of my table (small organization working on a mix of hard and soft issues). The NGO implementing the action had been working for a few years, mainly raising awareness and promoting the dialogue between the different actors involved in the solid-waste management sector through an institutional platform created and coordinated by the NGO itself. However, the new project required the NGO to focus on setting up recycling centers, organizing groups of citizens, starting cooperatives and coordinating garbage collection routes – a significant increase in the workload and the type of activities. Once again, all the effort was spent on achieving outputs, while the vision of institutional change got lost somewhere in the meanders of procurement and logistics. The NGO staff was so pressed to achieve output indicators, that there was no time to coordinate the solid-waste platform. As a consequence, the platform virtually stopped functioning. An opportunity for spreading lessons learnt and best practices was wasted. The impact of the project was much smaller than originally envisioned.

In this case, a combination of the two approaches could have helped to extend the results of the intervention to a bigger population. 

logframes vs outcome mapping

I have been giving some thought to two different approaches to project design, management and M&E: result-oriented and actor-oriented. My analysis doesn’t portend to be academically grounded. It is just the point of view of a practitioner. For a deeper comprehension of the different approaches, a good starting point is the documents section of outcome mapping website.

Logical frameworks are the perfect example of result-oriented approaches. They identify cause-effect relationships that are expected to produce the results you want to see during and after the implementation. Actor-oriented approaches focus instead on how stakeholders are supposed to behave in order to achieve the intended goal. Outcome mapping is an actor-oriented approach. If I had to explain the difference to a running buddy, I would say that result-based approaches are like a coach who creates a strict training regime. If you follow his training advice, he predicts that you will be able to run a sub-3 hour marathon.  The actor-based coach focuses instead to train excellent marathon runners, who live healthy lives, are engaged with their running community and develop a passion for an active lifestyle. How long it will take you to run 26.2 miles, he doesn’t know.

Personally, I find that both approaches are valuable. Logframes are great because they let you understand a project at a glance, while outlining a blueprint for action very much needed by dysfunctional organizations (there are quite a few out there). Outcome mapping does a fantastic job accounting for uncertainty in the project and highlighting the importance of institutional relationships. So I find quite extraordinary that the two approaches are generally presented as mutually exclusive.

If you ask me, finding a combination of the two approaches that retains the strengths of each one while shedding the respective weaknesses seems a very attractive proposition. Such a synergy of results- and process- oriented approaches is quite common in other disciplines. It is common practice in the business world to set specific sales and profit targets since the elaboration of the business plan and then periodically measure them to receive feedback on performance. At the same time, business plans also contain analyses of competitors, suppliers and customers and precise strategies to affect their behavior in a way that benefits the business. Mixing results-based and actor- or process-based approaches is the norm in the private sector. Going back to the running coach example, setting a PR in a marathon is not just a question of running 40 miles per week, it also takes an adequate diet and a healthy lifestyle. On the other hand, eating well and going to bed early is not enough to break 3 hours on a 26.2 miles course if you don’t run long and fast enough.

A good starting point for looking at integrated approaches could be this:

Size of implementing organization	Big	Results, although mostly intangible, produce observable impact due to relative size of the organization compared to the sector. Actor-oriented approaches can be helpful for project design and M&E.	Results are concrete and straightforward to measure. Impact is a consequence of sheer size more than of planned strategy. Impact on different stakeholders is likely to be overlooked. Integrating an actor-oriented approach is advisable to analyze full extent of project impact (potential negative impact).
	Small	Sector-wide impact is hard to measure/achieve. Results are mostly intangible. Actor-oriented approach is key to project design, management and M&E.	Although results can be easily measured, sector-wide impact eludes implementers. The focus is on activities/outputs. Integrating an actor-oriented approach to project design is necessary to scale up impact.
		Soft	Hard
		Type of project

  

Notes:

I tried to be consistent and use OECD-DAC definitions.

The categories “big”, “small”, “hard” and “soft” should be considered as extremes in a continuum that includes the full range of intermediate states. A good example of a “hard” project is an emergency food distribution. A “soft” project would be an advocacy campaign.

I consider a project “big” when its budget and/or actors represent a significant percentage of the sector of operations. Examples: line ministries, UN agencies or major NGOs are directly involved in the implementation and the budget is a consistent part of the annual government budget in that sector. 

Who needs our money?

I recently visited, in quick succession, Eritrea and Uruguay on M&E missions. The two countries are probably at the opposite ends of the development spectrum. Eritrea is a food-insecure country that is ranked 165 out of 179 countries in the 2009 Human Development Index. Uruguay has quickly rebounded from the 2001 crisis. Over the last few years, in the middle of the current financial turmoil, the Uruguayan economy has posted amazing growth rates (8,5% in 2010, 6% in 2011). However, what I observed during my visit is that the two countries do have something in common: neither one needs western aid, as it is currently delivered.

Eritrea says so explicitly. There are several reasons to explain such attitude, but the main one is simple: following donors' procedures is too cumbersome. Keep your money and stop bothering us with estimates, tenders and audits.

Uruguay is simply too well off: unemployment stands at 6%; investments flow to the country; the breach between rich and poor is closing. From many points of view, Uruguay is just way more advanced than many European countries. I had to monitor a EU funded project aimed at promoting transparency in local administrations. What I saw would have been considered amazing for most Italian municipalities. How can Europe afford to spend so much money on great projects abroad when the economy is so bad? How can Europeans preach transparency in Latin America when corruption is so pervasive in many regions of the old continent?

In my modest opinion, current aid delivery methods and strategies are as good as dead. Beneficiaries are out; it is time for partners now.

Presenting development goals

The story on measuring aid effectiveness in Haiti run by NPR this morning has plenty of material to stimulate a critical reflection on our work. Although I tend to become a little defensive when I hear criticism of NGO work, especially in such a difficult environment as Haiti after the quake, I have to admit that the report does a great job in highlighting some problematic areas. What interests me the most is the difficulty NGOs experience in presenting the impact of their work to the general public. M&E has made quantum leaps in analyzing and presenting information to institutional donors, but there is still a lot of work that needs to be done to improve accountability and communication to a wider audience.

Usually, what you find on websites, newsletters, etc. is a personal story of one of the beneficiaries that tells how the project changed her/his life. Nothing wrong with that: it allows the reader to establish a connection with the beneficiaries and to get a better understanding of how the project works. However, a personal story cannot tell the whole tale of how the project has been implemented. Success stories should be measured against costs, expected results, sustainability, etc.

A proper report should be able to present in layman’s terms the efficiency and effectiveness of your program and the wider impact on society. I remember seeing a couple of really cool software tools that combine maps, pictures and graphs to deliver a clear view of a project performance.

Here is one:

http://cartametrix.com/fata/demo_adv.php

Snake oil or miraculous tree?

This is a New Yorker article on reforestation efforts in the Sahel region. You can find an abstract on the New Yorker website, but the whole article is for paying customers only.

The author describes two different  approaches to contrast desertification, a technical solution involving a box that stores humidity (appropriately called Waterboxx) and the introduction of agroforestry in Niger. The article is written for the general public: it tells an interesting story but does not go in depth into the specifics of the two programs. The author describes Moringa trees (Moringa oleifera) as a miracle tree whose leaves and pods are super nutritious and can be used to cure a variety of medical conditions. I have heard the same claims while monitoring several projects in Africa but I could not find any scientific study that supports those assertions. The article does not provide any source either. I would appreciate inputs on Moringa's health and nutrition wonders.

Is participation the new black?

Great paper by David Parduhn on the consequences of mainstreaming participation in development.

 In my M&E experience, I never found a single project willing to admit that beneficiaries were not adequately consulted/involved in the design of the intervention. However, participation generally means presenting the proposal to groups of beneficiaries before sending it to the donor. Issues of power gaps, group dynamics, gender inequality are rarely taken into account.

Sometimes I try to imagine myself as a beneficiary: would I speak out at a meeting? Would I be brave enough to support a minority opinion or risk losing face by proposing an alternative view of my reality? What kind of barriers would I face?

Here is the full link to the article:
http://www.hiidunia.com/2011/10/‘everyone-is-doing-something-and-calling-it-pra’-a-critical-reflection-on-participatory-methods-in-development/

Discarding information

Here you can find a BBC article worth reading on the importance of discarding information. The amount of information we are exposed to is huge. Saving and storing what is important and throwing away the rest is crucial.